3D技术在精准肝胆管结石外科诊治中的应用研究
|
摘要
背景
肝内胆管结石又称为肝胆管结石病,是指肝总管、左右肝管汇合部及其肝内各级胆管内的结石,同时合并胆管的炎症、狭窄、肝纤维化、萎缩及肝功能障碍等多种并发症。肝胆管结石是世界性的疾病,以亚洲国家多发。在我国华南、西南、长江流域及东南沿海等广大区域尤为多见。由于其病变复杂、复发率高且常引起严重的并发症,成为我国良性胆道疾病死亡的主要原因。手术是肝胆管结石的主要治疗方式,目前公认的手术方式为肝切除术,同时辅以各种形式的胆管整形和胆肠吻合术。但由于肝内胆管结石的分布广泛,同时合并不同程度的肝胆管狭窄和肝脏毁损性病变,术后肝内胆管狭窄和引流不畅易致结石残留及复发,而使得肝胆管结石的疗效难以满意。
肝胆管结石病的诊断主要依靠B超、CT和MR等无创性影像学检查,以明确结石的部位、大小、数量和分布情况,胆管狭窄的部位及程度,从而进行定位诊断。并结合患者肝功能状况、Oddi括约肌功能、有无肝胆管炎发作、有无梗阻性黄疸、有无肝门部胆管狭窄、有无门静脉高压症和脾肿大等综合因素制定手术方案。但由于肝内胆管结石的分布广泛,同时合并不同程度的肝胆管狭窄和肝脏毁损性病变,上述各种影像学检查手段各有优缺点,常需几种影像学检查方法的综合运用,才能做出较为全面的诊断。目前尚无一种理想的诊断方法能够对结石大小、数量和分布,胆管狭窄程度和长度,肝脏病理形态,以及胆管和血管的关系做出系统全面的诊断。
目前,国际上尚无统一、公认的肝胆管结石分型标准。对肝胆管结石和狭窄常用的分型方法目前有三种：一种是日本Nakayama分型,一种是Tsunoda分型,还有一种是中华医学会胆道外科学组2007年的中国分型。上述各种分型的依据均基于结石的分布、胆管狭窄及扩张的部位和程度等。由于影像学检查手段的局限,上述分型对结石分布位置、胆管狭窄部位、程度、长度等均缺乏精确的立体定位,缺乏胆管与门静脉、肝静脉解剖关系的详细描述,因此难以制定合理化的手术方案,实现肝胆管结石的术中精准治疗。
近年来,随着计算机三维重建和可视化技术的发展,3D技术已经成功应用于人体解剖结构和组织功能的研究,并应用于疾病诊断和术中导航。2003年本课题组采用数字化虚拟中国女性一号肝脏数据集和肝脏管道灌注标本CT薄层扫描数据进行三维重建和虚拟手术研究,随后开始利用活体人64排螺旋CT扫描数据进行肝胆脾胰等脏器以及腹腔血管的计算机三维重建及虚拟可视化手术的研究。由于三维重建和虚拟手术是基于患者特征性,因此实现了个体化的要求。本课题组目前已对肝胆系统多种疾病,包括肝癌、肝血管瘤、胰腺癌等实现了个体化的三维重建和虚拟仿真手术,显示了良好的临床应用价值。本研究拟将三维可视化技术应用于肝胆管结石病胆道数字化解剖,临床病理分型以及个体化的外科治疗研究,以期进一步提高肝胆管结石病的诊断精准率、手术治疗精准性,从而提高肝胆管结石病的临床疗效。
目的
1、3D技术进行胆道系统数字化三维重建。分别采用PHILIPS CT自带图像后处理Mxview工作站和腹部医学图像三维可视化系统(MI-3DVS)进行肝胆管结石患者肝脏及肝内管道,尤其是结石和胆管的三维重建,实现肝脏胆道系统的数字化解剖。
2、将三维重建和仿真技术引入到肝胆管结石的临床诊断、分型研究以及手术治疗中,探索其在肝胆管结石手术方式合理选择和控制性肝切除治疗上的临床应用价值。
方法
1.基于64排螺旋CT扫描数据的数字化胆道解剖研究
1.1设备
64排螺旋CT—PHILIPS Brilliance 64(荷兰PHILIPS公司),PHILIPS Brilliance64层螺旋CT自带的图像后处理Mxview工作站,FreeForm Modeling System(美国SensAble Technologies公司),力反馈设备PHANToM (PHANToM Desktop), ACDSee9 Message Center (ACD System Ltd.),自主研发医学图像三维可视化系统(MI-3DVS)(主机：内存2G,处理器2.0G*2 Xeon 5130)。
1.2胆道系统结石病例CT扫描参数设定及数据存储
常规平扫时患者取仰卧位,头足方向,由膈顶至双肾下缘。扫描条件：120KV、250mAs,采用0.625×64排探测器组合,层厚5 mm、间隔5mm,螺距0.984,球管旋转一周时间0.5s。患者检查前禁食至少6-8h,扫描前30min口服500 ml清水,作为阴性胃肠道对比剂。先行平扫后再进行增强扫描,平扫最大范围从气管分叉部至耻骨联合上缘水平。增强扫描经肘前静脉采用自动高压注射器经前臂静脉进行团注非离子碘造影剂优维显(ultravist),剂量1.5ml/kg,速度5ml/s,注射对比剂后21～25s启动肝动脉期扫描,30～35s行动脉晚期扫描,50～55s行门静脉期扫描,每期扫描时间约6～8s。扫描结束后将图像数据传至Mxview工作站,在Mxview工作站进行三期数据(平扫期、动脉期和门脉期)的刻盘存贮。数据格式为DICOM(Digital Imaging and Communications in Medicine)3.0,通过DICOM查看器转换为BMP格式。
1.3 PHILIPS Brilliance 64层螺旋CT Mxview工作站上胆道三维重建
对原始数据库中层厚5mm的扫描数据进行薄层重建(层厚0.67mm,间隔0.33mm),薄层处理图像传至Maxview工作站。在Maxview工作站中,对薄层图像数据进行胰胆管和腹腔血管冠状位、矢状位、任意平面或曲面三维重建。主要重建方法包括：最大密度投影法(maximum intensity projection, MIP)、最小密度成像法(minimum intensity projection, MinIP)、曲面重建法(curve reconstyuction, CR)、容积重建(volume rendering, VR)和多平面重建(Inultiplanar reformation, MPR)等。重建后多角度,多方位观察胆管解剖结构,并截图保留。
1.4 MI-3DVS肝胆管系统三维重建
利用Mxview Viewe rDICOM (?)阅读64排CT扫描数据,调整适当的窗宽和窗位,并将以DICOM数据导出并转化成JPEG格式存盘,将JPEG文件导入ACDSee 9 Message Center转化成BMP文件,并调整图像大小后导入自主开发的MI-3DVS中,以自适应的区域生长法对肝脏、胆道系统及其周围血管系统进行序列分割,得到分割后的肝脏、胆道系统、腹主动脉、腹腔动脉及其分支、门静脉、肝静脉系统等STL(STereo Lithography)格式数据。重建后的STL模型导入FreeForm Modeling System进行平滑和修饰,得到光滑逼真的肝脏及其内部管道系统的三维图像。立体观察胆道系统解剖结构,录制视频及多角度视频截图。
2、肝胆管系统三维可视化技术对肝脏分段和肝胆管结石分型研究
2.1仪器与设备同“1.1”
2.2胆道系统三维重建方法及步骤同1.1、1.2、1.4。
2.3基于肝静脉和门静脉解剖的肝胆管结石的肝脏个体化分段
在肝脏及门静脉、肝静脉、胆管和肝动脉三维重建基础上,肝段的划分以门静脉为指示标志,肝静脉为分界标志,胆囊和韧带为辅助分段标志,每一肝段均有一(组)独立的门静脉段支供应。在肝段划分和病变的定位诊断中,门静脉的解剖及病灶与门静脉的关系更为重要。将自门静脉主干发出的分支定义为一级分支,三级分支所供应的肝脏范围定义为一个肝段。根据具体的门静脉三级分支的数量,将肝脏按罗马数字分为I-X个不等的肝段。由于Gelisson系统包括门静脉和胆管,因此可将肝胆管结石所在部位和狭窄或扩张胆管的位置进行精确定位诊断。
2.4基于个体化分段的肝胆管结石解剖分型
利用三维可视化肝脏分段结果,参考中华医学会胆道外科学组2007年分型方法日本分型(?)(?)Tsunoda的分型依据,从如下几个方面进行肝胆管结石的解剖分型。
2.4.1结石或病变胆管分布位置(location)
采用罗马数字的Ⅰ-Ⅹ肝段定位结石和胆管病变(狭窄或扩张)的部位。
2.4.2胆管狭窄程度(stenosis)
由于正常二级胆管直径为2-3mm,因此以2mm作为绝对狭窄的标准。相对狭窄判断标准,近端胆管狭窄与远端扩张胆管的直径比>2/1为相对狭窄(轻度),而<1/2为绝对狭窄(重度),相对和绝对狭窄均需外科手术解除。狭窄分为S0(无狭窄)、S1(轻度狭窄)、S2(重度狭窄)
3.4.3胆管扩张程度(distention)
以肝内胆管内径10mm作为轻度或重度扩张的分界,>2mmm且<10mm为相对扩张(D1)>10mm为扩张(D2),>15mm为显著扩张(D3),。扩张分为0-3级别。D0(无扩张)、D1(轻度扩张)、D2(中度扩张),D3(重度扩张)。
2.4.4病变肝段体积萎缩(atrophy)
通过肝脏三维重建结果观察病变肝段体积的变化,根据有无病变肝脏体积的萎缩或邻近肝段体积的代偿性肥大。对于萎缩肝脏,手术切除为唯一选择,因此,将病变肝段萎缩作为病理分型的指标之一,可以帮助术前确定手术方式为肝段切除还是肝内胆管切开取石。
2.4.5门脉高压症(portal hypertension, PHT)或肝硬化(cirrhosis)
胆汁淤积性肝硬化导致门脉高压症、脾脏肿大是影响肝胆管结石的预后的重要因素之一,常影响肝胆管结石的手术方式的选择。对于存在多处肝内胆管结石合并胆汁淤积性肝硬化,手术方式应选择各种方式的切开取石和内外引流术,肝切除应慎重。三维可视化技术可清晰观察到有脾脏肿大,脾周血管的扩张,肝脏萎缩-肥大综合征等影像改变。因此将肝硬化或门脉高压症列为病理分型指标之一。
其他因素,如有无Oddi括约肌松弛型功能失调,有无肝胆管炎发作,有无需急于处理的梗阻性黄疸、肝脓肿、胆道出血,有无合并肝胆管癌等,虽然对确定手术的时机和确定手术的类型有所帮助,但三维可视化技术难以从影像学上作出相应的判断,因此不列为分型因素。
3、3D技术在肝胆管结石诊治中的应用
3.1研究对象：收集我院自2008年10月至2010年10月60例肝胆管结石患者64排螺旋CT数据。所有患者进行三维重建和病理分型,并进行仿真手术,确定手术方式和手术切除部位,进行临床实际手术,观察术前三维与术中所见、术前仿真手术与实际手术操作符合情况。
3.2仪器与设备同“1.1”
3.3胆道系统三维重建方法及步骤同1.1、1.2、1.4。
3.4图像后处理
将重建后的STL三维模型导入FreeForm Modeling System中,进行光滑、去噪,得到光滑逼真,立体感强的三维模型。为便于立体观察,利用MI-3DVS自带着色功能,分别将肝脏、肝静脉、肝动脉、门静脉、胆道、结石以及腹腔血管和周围脏器渲染不同的颜色,同时采用不同程度的透明化处理或脏器隐去技术,分别观察肝脏和胆道及结石、肝脏和动脉,肝脏和门静脉、肝脏和肝静脉等不同组合的三维解剖关系。通过三维模型的旋转,观察病变部位与相邻脏器的不同角度的空间解剖,分别录制视频、截图保留。在本研究中,主要观察胆管树在肝脏内的立体形态、结石在胆道系统的分布、胆管狭窄的部位和程度、胆道系统与肝动脉、门静脉和肝静脉系统的三维立体关系。
3.5肝内胆管结石的虚拟手术
在FreeForm Modeling System中对三维重建模型及各组成部分进行放大、缩小、旋转、透明等操作,全方位观察各结构或细节,根据结石分布、胆道系统及肝脏病变情况选择合理手术预案,利用力反馈设备PHANTOM和自行开发设计的开腹手术器械进行肝内胆管结石的各种虚拟仿真手术,以确定最佳手术方案。
3.6肝内胆管结石的临床手术
根据三维模型视频和截图的多角度旋转观察,明确肝胆管结石在肝脏内的大小、数量、部位等信息,同时观察有无肝胆管狭窄及扩张、狭窄胆管的长度和程度、有无肝脏萎缩-肥大等信息,进行术前诊断和分型,通过仿真手术制定最佳手术方案。将三维可视化视频及截图的带入手术室,对照术中实际解剖情况,实时指导手术操作。常用手术方法有胆道探查取石(bile duct exploration,BDE)、部分肝叶切除(hepatectomy,HT)、左外叶切除(Left lateral lobectomy, LLL)、右后叶切除(right posterior lobectomy, RPL)、左半肝切除(left hepatectomy,LH)、肝(胆)肠吻合(hepaticojejunstomy,HJS)、肝内胆管置入支撑管(intrahepatic bile duct support tube placement, IBDSTP)等,具体手术方式为上述手术方法的不同组合。通过实际手术,对比术前三维与术中解剖所见,术前仿真手术与实际手术操作情况,评价3D技术对肝胆管结石精准治疗的的临床应用价值。
结果
1、基于64排螺旋CT扫描数据的数字化胆道解剖研究
1.1 64排螺旋CT胆道三维重建
60例胆道结石疾病患者,胆道三维重建成功54例,6例因无明显胆道扩张,肝胆管三维重建不理想,但横截面CT可清楚显示结石病灶。平扫横截面观察肝内胆管结石形态多种多样,呈斑点状、类圆形、条状或串珠样高密度影。结石部位及其近端胆管多有不同程度的狭窄及扩张,伴随病变肝组织萎缩或周围肝组织代偿性增生。肝外胆管结石表现为胆总管内有圆形或环形致密影,周围被低密度胆汁环绕,形成“靶征”或“半月征”。
胆道三维重建后,可见肝内单发、多发或弥散性分布于不同肝段内的高密度结石,伴随肝内胆管不同程度的扩张。结石所在部位胆管狭窄,远端胆管及其分支扩张,扩张胆管呈“树枝状”,有时扩张的胆管成“胆汁湖”或形成局限性密度不均的肝脓肿。MinIP除可观察到高密度结石及扩张的肝内外胆管外,还可显示胆总管内软组织密度的阴性结石、扩张的胆管分支可达3-4级,扩张胆管多为“枯枝状”。一般近肝门胆管扩张明显,而肝内胆管为局限性扩张或狭窄,胆管分支稀少,纤细。还可见肝内胆管高密度结石形态大小外,还可观察到病变肝实质萎缩-增生复合征。对于合并胆总管结石,胆总管内可见不同形态、大小的单个或呈串珠状多发高密度结石影。梗阻上段胆管扩张明显,远端胆管不显影或突然截断。胆总管结石表现为结石以上胆管全部扩张,胆囊亦扩张,肝内胆管不同程度扩张呈“枯枝状”,结石以下胆总管扩张逐渐变细,合并化脓性胆管炎则全程扩张。
1.2 MI-3DVS三维重建胆道解剖
1.2.1肝脏及肝内外血管系统的三维重建
肝脏模型能真实反映肝脏的实际体积和肝脏的解剖标志,并且通过调节肝脏的透明度可同时显示肝脏和肝内的动脉、静脉、门静脉各分支。腹主动脉、腹腔动脉及其分支胃十二指肠动脉、双肾动脉、胃右、左动脉、脾动脉、肠系膜部分动脉、肝固有动脉、左右肝动脉及其肝内分支等结构,形态逼真,立体感强。门静脉系统显示肝外的主干和脾静脉、肠系膜上静脉；肝内门静脉系统能清楚显示门静脉的左主干和右主干,以及各叶、段的分支。肝静脉系统能清晰显示三支肝静脉在肝脏内部的分支分布,三支肝静脉与肝上下腔静脉的汇入情况、各肝静脉之间的空间解剖等。
1.2.2胆道系统及结石的三维重建
利用平扫期的CT图像和门静脉期的CT图的辅助,MI-3DVS自适应区域生长法图像分割法通常能一次性完整分割胆道系统。对一些细微结构如胆囊管与胆总管汇合部、不扩张的肝门部胆管等,可利用不同血管期如平扫期、动脉期、静脉期、门静脉期等,分别对各期胆道系统进行图像分割,再利用系统自动配准功能将各期所含特定解剖学信息整合为完整的胆道三维图像。重建的胆道系统模型能真实反映结石。胆道系统与肝内外血管的空间位置关系。胆道系统三维模型与腹腔脏器及血管模型配准后,整个上腹部脏器及肝脏血管系统和胆道系统的立体关系完整重现。
经MI-3DVS分割重建后,本组54例患者肝脏及其内部管道系统三维重建图像逼真,立体感强。可清晰再现患者肝脏的立体形态和有无肥大和萎缩；肝内一、二、三级胆管树立体形态,以及狭窄或扩张胆管的长度和直径,部分肝内胆管结石合并肝内胆管广泛扩张者,甚至四级胆管(亚肝段胆管)也得以显示；结石的大小、数量、在肝内外胆管的立体分布,以及胆管和肝内血管系统的空间结构情况也清晰可见。
2、肝胆管系统三维可视化技术对肝胆管结石病理分型研究
采用包括结石或病变胆管分布位置(location, L)、胆管狭窄程度(stenosis,S)、胆管扩张程度(distention,D)、病变肝段体积萎缩(atrophy,A)、门脉高压症(portal hypertension, PHT)等因素进行了肝胆管结石的分型诊断,举例如下：例1.肝内胆管结石症(LⅡ、Ⅲ,SⅢ1,DⅢ2,AⅡ、Ⅲ):表明肝第Ⅱ、Ⅲ段结石,同时伴
有Ⅲ段肝内胆管的轻度狭窄和扩张,Ⅱ、Ⅲ段体积萎缩。例2.肝内胆管结石症(LⅥ、Ⅶ,SⅥ1,DⅥ0,AⅥ):表明肝Ⅵ、Ⅷ段结石,Ⅵ段胆管轻度
狭窄,无远端胆管扩张,Ⅵ段体积萎缩。
基于三维可视化技术的肝胆管结石患者肝脏分段符合个体化肝脏解剖特征,对结石分布和胆管病变部位做出了精确的三维定位诊断。结合胆管狭窄和扩张程度范围、有无胆汁性肝硬化和肝脏萎缩肥大等,做出了较为合理的分型诊断。新分型方法考虑到了结石和胆管狭窄和扩张部位、肝脏体积变化和肝硬因素,不仅对肝胆管结石和胆管病变的定位诊断更加精确,对制定手术切除范围、胆道引流术式等外科处理方式有更加实际的临床指导意义。
3、3D技术在肝胆管结石诊治中的应用
3.1三维重建肝内“胆管树”和“血管树”立体形态清晰
本组患者肝脏三维重建的立体模型形态逼真、解剖结构标志正确。在肝脏透明化处理情况下,通过单独或不同组合的联合显示方式,可清楚观察到病变肝脏有无萎缩、肝内胆管和血管的的立体分布形态、结石的大小及在肝胆管内的分布范围、胆管狭窄程度和范围、有无合并肝外胆管结石等。3D模型除了观察到肝内结石分布、胆管系统及肝实质的病变情况以及病灶与周围组织的相互关系,还可明确肝脏的血供类型及血管变异情况等,从而准确进行肝胆管结石的分型诊断和仿真手术。
3.2肝胆管结石仿真手术
在虚拟手术环境中,立体模型能通过放大、缩小、旋转及透明化进行观察组织结构的组织器官,可明确肝内结石分布以及胆道系统与肝脏病理改变情况,以及肝内血管树的改变,据此进行临床分型及选择最佳的手术方式,进而虚拟上述各种手术过程,观察手术效果。本组54例肝胆管结石患者选择30例进行仿真手术,其余诊断明确的的Ⅰ型和不伴有胆管狭窄的24例肝胆管结石患者未进行仿真手术,30例患者经仿真手术观察切除平面内重要血管和胆管的解剖关系,制定了最终优化手术方案,指导实际手术过程。
3.3胆道三维可视化技术临床应用
根据三维视频和截图的观察分析,以及仿真手术的演练,54例患者均制定了手术方案。实际手术方式包括BDE+LLL+HJS30例,BDE+LLL+IBDSTP 8例,LLL+RPL+HJS3例,LH+HJS 3例,LLL+ HJS+IBDSTP 10例。54例病人肝脏重建模型与术中所见均符合。30例仿真手术方案与实际手术方式符合者为90%(27/30,3例急诊患者难以行根治性手术)。所有患者术后未出现术后严重并发症,无死亡,51例非急诊患者术后胆管造影未见结石残留,3例患者半年后结石复发,半年复发率5.58%(3/51)。
结论
1、三维重建技术可实现胆道数字化解剖
基于64排螺旋CT扫描数据,CT自带Mxview工作站和MI-3DVS均能实现胆道数字化解剖。相对于二维CT图像,3D图像能立体显示胆道系统的胆管树形态和结石分布情况。但CT工作站的三维重建图像仅是某一血管期的胆道像,且提供给临床医生的仅为三维图像的矢状位或冠状位的二维平片,并非真正意义上的三维立体图形。难以将动脉期、门静脉期及静脉期的三期血管与肝脏及胆管树的立体解剖关系同时显现,难以不同角度和方向观察病变胆管及结石与肝内血管系统的空间解剖。MI-3DVS可将肝脏、周围脏器、腹腔血管、肝内不同管道系统分别渲染不同颜色,通过可视化技术,获得了整个上腹部脏器的整体立体图像；并能将肝脏、肝动脉、肝静脉、门静脉、腹腔血管以及周围脏器等单独、同时或不同组合方式分别显现；通过局部放大、旋转立体观察,肝内胆管树形态、结石的分布、胆管狭窄部位和程度可清晰显示。与传统的黑白二维图像比较,三维图像对病变部位与周围血管及脏器等重要解剖结构的显示更加清晰直观,实现了真正意义上的三维立体显现,对肝胆管结石可作出精确的术前诊断。
2、三维可视化科实现肝胆管结石精确分型诊断
基于三维可视化技术的肝胆管结石患者肝脏分段符合个体化肝脏解剖特征,对结石分布和胆管病变部位做出了精确的三维定位诊断,结合胆管狭窄和扩张程度范围、有无胆汁性肝硬化和肝脏萎缩肥大等,做出了较为合理的病理分析。新分析方法考虑到了结石和胆管狭窄和扩张部位、肝脏体积变化和肝硬因素,不仅对肝胆管结石和胆管病变的定位诊断更加精确,对制定手术切除范围、胆道引流术式等外科处理方式有更加实际的临床指导意义。
3、三维可视化技术对肝胆管结石的临床治疗有重要指导价值
胆道三维可视化和仿真手术能够实现肝胆管结石的术前精确诊断,通过术前仿真手术反复演练,针对不同手术方案和不同手术方法和路径的比较,制定最优手术方案,指导术中精确操作。三维可视化技术对复杂肝胆管结石手术方式的合理选择上具有重要参考价值。
Background
Hepatolithiasis, also called calculosis, is the presence of gallstones in the common hepatic duct, the region of hepatic biliary confluence as well as all levels of intrahepatic bile ducts with accompanied complications such as inflammation or stenosis of bile ducts, hepatic fibrosis, hepatatrophia, liver dysfunction, and so on. Hepatolithiasis is a worldwide disease and is prevalent in Asian countries, especially in South China, Southwest China, the Yangtze valley and southeast coastal areas. Due to the complicated pathological changes, relative high recurrence rate and severe accompanied complications, hepatolithiasis is the main cause of death from benign hepatobiliary diseases. As the main therapy, surgical treatment, especially hepatectomy is widely accepted with the aid of bile-duct anaplasty and choledochojejunostomy. However, due to extensive distribution of intrahepatic duct calculus with accompanied different degrees of devastated liver lesion and stenosis of hepatic ducts, stenosis of bile duct and inadequate drainage-caused recurrence of calculosis after surgical operation, surgical treatment could not always lead to satisfactory results.
The diagnosis of hepatolithiasis is mainly relied on non-invasive imaging examination, such as type-B Ultrasound, computed tomography (CT) and magnetic resonance imaging (MR), etc. The classification diagnosis is based on the detection of the location, size, number and distribution of calculi and the location and degree of stenosis of bile ducts. To make a surgery plan, diverse and combinatory factors should be considered, such as the function of liver and Oddi sphincter, emergence of hepatocholangeitis, obstructive jaundice, hepatic portal bile duct stricture, portal hypertension or splenomegaly and so on. Due to the extensive distribution of calculi, different degrees of stenosis of intrahepatic ducts and devastated liver lesions, each of the above-mentioned imaging examinations has its advantages and disadvantages. Such situation requires the combinatory application of several imaging examinations to get final comprehensive diagnosis. Currently, there is no single perfect method can reach a final systematic and comprehensive diagnosis conclusion on the number and distribution of calculi, the degree and length of stenosis of bile ducts, the liver pathomorphism, the relationship between bile duct and blood vessel.
Currently, there isn't a worldwide-recognized classification criterion of hepatolithiasis. There are only three classification methods have been commonly used for the classification of hepatolithiasis, including the Japanese Nakayama classification, the Tsunoda classification and the Chinese classification which was proposed by the biliary-tract surgery group of Chinese Medical Association (CMA) in 2007. The above classification methods are all based on the distribution of Calculi and the location and degree of cholangiectasis and stenosis of bile ducts. However, these methods are lack of precise stereotaxis of the distribution of Calculi and the location and degree of cholangiectasis due to the limitation of imaging examinations, which also can not give detailed anatomical relationship among bile ducts, portal vein and hepatic vein. Thus it's hard to design a suitable surgery plan to achieve the precise operation for hepatolithiasis.
In recent years, with the development of computerized techniques for three-dimensional (3D) reconstruction and visualization technology,3D visualization technique have been successfully applied to the studies of human body anatomy structure and organ functions, and also to the clinical applications of disease diagnosis and intra-operative navigation. Since 2003, our group used the hepatic CT scan data set from digitalized virtual Chinese Female No. one and specimens liver with perfusion of hepatic ducts to perform the studies of the 3D reconstruction and virtual surgery. After then, using the data of Live humans liver, gall, spleen, pancreas, other visceral organs and vessels scaned by the 64 multi-detector helical CT, we performed the study of 3D reconstruction and virtual surgery.Because the study of 3D reconstruction and virtual surgery is based on the individual characteristics of patients, it satisfied the individual requirement. Currently, our group has completed the individual 3D reconstruction and virtual simulation of many hepatobiliary diseases, including liver cancer, hepatic hemangioma and pancreatic carcinoma and so on, which shows important value of clinical application. In this work, we apply the 3D visualization technique to study the digital dissection of biliary tract, clinical pathological classification and of individualized surgical therapy for hepatolithiasis, which shall improve diagnostic accuracy and precision of surgical treatment for hepatolithiasis, and finally will contribute to the clinical efficacy of hepatolithiasis.
Objects
1. Establishment of digitalized 3D reconstruction of biliary system. We perform 3D reconstruction of liver and it's internal ducts, especially calculus and bile duct, to establish digitized anatomy of hepatobiliary system by using the CT image-postprocessing Mxview (Philips) workstation and abdomen medical image-3D visualization system (MI-3DVS) respectively.
2. We apply 3D reconstruction and simulation techniques to the clinical diagnosis, classification and surgical treatment of hepatolithiasis, then to explore the clinical value of this techniques for reasonable selection of surgical operation and controlled hepatectomy of hepatolithiasis.
Methods
1. The study of digitalized anatomy of biliary system based on 64 multi-detector helical CT scanning data
1.1 Equipments
64 multi-detector helical CT machine (Holland PHILIPS); The Mxview workstation accompanied with 64 multi-detector helical CT machine; FreeForm Modeling System (USA, SensAble Technologies); Force feedback system PHANToM (PHANToM Desktop); ACDSee9 Message Center (ACD System Ltd.); medical image-3D visualization system (MI-3DVS) developed and set up by ourselves [Host machine containing memory (2G), microprocessor (2.0G×2Xeon 5130)].
1.2 The set of CT scanning parameters and data storage of biliary system from clinical hepatolithiasis cases
When performing routine plain CT scan, the patient was lying down in supine position and scanned from diaphragmatic dome to lower margin of pancreas. The scanning parameter was set as following:120 KV and 250 mAs with 0.625 mm×64 multi-detector combination,5 mm-thick reconstruction layer,5 mm layer intervals, thread pitch 0.984 and 0.5s for one rotation of the tube. The patients were fasted at least 6 to 8 hours before examination. As negative gastrointestinal contrast reagent,500-ml clean water was taken orally by patients. Plain scan was performed from the bifurcation of bronchi to the upper margin of pubic symphysis ahead of contrast enhanced scan which was performed through the antecubital vein and forearm vein using automatic high pressure injector to do bolus injection of nonionic diodone ultravist with the dosage of 1.5ml/kg at a rate of 5ml/s. Hepatic arterial phase scan, late arterial scan and portal phase scanning was performed 21-25s,30-35s and 50-55s respectively after injection of contrast reagent and each of the scan lasted about 6-8s. After the scan, the image data was transmitted to Mxview workstation and all the three phase (plain scan phase, arterial phase and portal phase) data was burned into disc for storage. The data format is in DICOM (Digital Imaging and Communications in Medicine) 3.0 which can be transformed to BMP format through DICOM viewer.
1.3 3D reconstruction of biliary system on the Mxview workstation accompanied with PHILIPS Brilliance 64 multi-detector helical CT
The scanning data of 5-mm-thick layer in the original database were 3D reconstructed using thin slice width reconstruction (0.67-mm thickness and 0.33-mm layer interval). The reconstructed CT images were transmitted to Maxview workstation. In Maxview workstation,3D reconstruction of coronal, sagittal, arbitrary plane surface or curved surface of pancreaticobiliary duct and celiac blood vessels were performed using those thin-thick image data. The main reconstruction methods include maximum intensity projection (MIP), minimum intensity projection (MinIP), curve reconstruction (CR), Volume rendering (VR) and multiplanar reformation (MPR). Thus we can observe the anatomic structure of bile duct through multi-angle and multi-dimension.
1.4 3D reconstruction of hepatobiliary system using MI-3DVS
In this work, the segmentation method combined the algorithms of threshold segmentation and region growing,3D segmentation and two-dimensional segmentation. The scanning data from 64 multi-detector helical CT were read by MxliteView DICOM Viewer, adjusted with suitable window width and level, and exported as JPEG format files. Then the JPEG files were imported into ACDSee 9 Message Center, transformed into BMP files and imported into MI-3DVS after re-sizing the images. In MI-3DVS, the sequential segmentation of biliary tract system and around blood vessel system were performed using self-adaptive region growing algorithm to get the data with Stereo Lithography (STL) format of hepatobiliary system, abdominal aorta, celiac artery and its branches, portal vein and its branches, and hepatic veins, et al. Finally, the reconstructed STL models were imported into FreeForm Modeling System for smoothing and retouching to get the smooth and lifelike 3D images of hepatobiliary system. The video and the multi-angle video recordings were recorded for 3D observation of the anatomic structure of biliary system.
2. Classification of hepatolithiasis using 3D visualization technique of hepatobiliary system
2.1 Equipment same as 1.1
2.2 Methods for 3D reconstruction of biliary tract system:same as 1.1,1.2,1.4
2.3 Individual liver segmentation of hepatolithiasis based on the anatomy of hepatic vein and portal vein
Based on the 3D reconstruction of liver, portal vein, hepatic vein and bile duct, the segmentation of liver is indicated by portal veins as indicative marks, hepatic veins as boundary marks, gall bladder and ligament as auxiliary segmentation marks. Each segment of liver is supported by an independent branch of portal vein. For the segmentation of liver and location diagnosis of lesion, the anatomy of portal veins and the relationship between lesion and portal veins is more important. Normally, the branches stem from main trunk of portal vein was defined as the first level branch, and the part of liver supported by a third-level branches of portal veins is defined as one hepatic segment. Thus we can divide the liver into 1 to 10 hepatic segments named with Roman numerals fromⅠtoⅩ. Because the Gelisson system contains the portal veins and bile duct, we can do accurate location diagnosis of hepatolithiasis and cholangiectasis or stenosis of bile duct.
2.4 Anatomic classification of hepatolithiasis based on the individualized liver segmentation
Based on the Japanese Nakayama classification, the Tsunoda classification and the Chinese classification which was proposed by the biliary-tract surgery group of Chinese Medical Association (CMA) in 2007, we used the results of hepatic segmentation from 3D visualization and classified the anatomy of hepatolithiasis through the following criteria.
2.4.1 Location:using roman-numers-labeled hepatic segmentation to localize the calculi and cholangiectasis or stenosis of bile duct.
2.4.2 Stenosis:Because the diameter of normal bile duct is 2 to 3 mm,2mm was selected as the criterion of absolute stenosis. The criterion of relative stenosis is defined as the diameter ratio of proximal terminal to distal terminal. Thus the stenosis is classified as SO (no stenosis), S1 (mild stenosis is defined as the diameter ratio above 1/2), SO (severe stenosis, also called absolute stenosis, is defined as the diameter ration below 1/2).
2.4.3 Distention:To evaluate the distention degree of bile duct,10 mm was used as the criterion to discriminate the mild and sever distention of the internal diameter of intrahepatic duct. Thus the distention the internal diameter is classified as DO (no distention), D1 (mild distention,>2mm and <10mm), D2 (moderate distention,>10mm), D3 (severe distention,> 15mm).
2.4.4 Atrophy:We observed the volume variation of the pathological hepatic segments through 3D reconstruction to confirm whether there is atrophy of the pathological liver or the compensatory hypertrophy of neighboring hepatic segments. The surgical removal is unique choice for the hepatic segments showing atrophy. So using the hepatic segment with atrophy as a pathological typing index can help to select the surgical operation with surgical excision of hepatic segment or incision removal of intrahepatic calculus.
2.4.5 Portal hypertension (PHT) or Cirrhosis. One of the important factors affecting the prognosis of hepatolithiasis is the portal vein hypertension and splenomegaly caused by biliary cirrhosis, which commonly influence the selection of surgical operation of hepatolithiasis. For the existence of multiple calculi of intrahepatic duct combined with biliary cirrhosis, the surgical operation shall choose incision removal of calculus in different kind of ways and internal and external drainage and be cautious for hepatectomy. The 3D visualization technique can clearly observe the pathological changes of splenomegaly, the distention of around blood vessels, atrophy and hypertrophy of liver. Thus we use portal hypertension as a pathological typing index.
Other factors, such as Oddi's sphincter functional disturbance, hepatocholangeitis, obstructive jaundice, hepatic abscesses, hemobilia and accompanied cholangiocarcinoma, etc, which are helpful to choose the type and timing of surgical operation, but are hard to made a correct diagnosis from imaging analysis.
3. The application of 3D reconstruction technique in the diagnosis and treatment of hepatolithiasis
3.1 Research objects. The data of 64 multi-detector helical CT of total 60 patients with hepatolithiasis were collected from Oct.2008 to Oct.2010. All the patients'data were analyzed with 3D reconstruction and pathological classification, and subjected to surgery simulation to determine the surgical procedure and resection sites. After then the real surgical operations were performed and compare the real situation with the results from previous 3D reconstruction and surgery simulation.
3.2 Instruments and equipment:same as 1.1
3.3 Methods for 3D reconstruction of biliary tract system:same as 1.1,1.2,1.4.
3.4 Post-processing after imaging
The reconstructed STL models were imported into FreeForm Modeling System for smoothing and denoising to get the smooth, lifelike 3D models. For the convenience of 3D observation, we use MIPS to render the liver, hepatic veins, hepatic artery, portal veins, biliary tract, calculus, abdominal vessels and around visceral organs with different colors. At the same time, the 3D anatomic relationships were observed.
3.5 virtual operation of hepatolithiasis
The 3D model was magnified, minified, rotated, and transparentized to observe every parts of organs and lesions in all directions with the FreeForm Modeling System, so that the distribution of calculi, the states of intra- and extra-hepatic bile ducts and liver would be explicit, based on which a proper operation plan can be made. Additionally, virtual reality surgery could be applied to hepatolithiasis to choose an optimal operation method based on PHANTOM, an instrument with force feedback, and self-developed surgical instruments.
3.6 Actual operation of hepatolithiasis
Based on the observation of the video of 3D models and printscreens in any directions, the size, the amount and position of hepatolithiasis will be explicit; the stenosis and distention of hepatic duct, the length of the narrow bile ducts, and hepatatrophia and megalohepatia could be observed simultaneously; so as to make preoperative prognosis, pathological classification and operation plans. When brought to operating room, the video of 3D models and printscreens compared to the actual operation, can offer the real-time direction of operations. The common surgical methods are listed as follows:bile duct exploration (BDE), hepatectomy (HT), left lateral lobectomy (LLL), right posterior lobectomy (RPL), left hepatectomy(LH), hepaticojejunstomy(HJS), intrahepatic bile duct support tube placement (IBDSTP), and so on. The actual operation is the combination of different surgical methods mentioned above. After actual operation, on the basement of the comparison between 3D models and the true structure of organs in operation, and that between the virtual reality operation and the process of the real one,then the clinical value of 3D technique in the accurate treatment of hepatolithiasis is evaluated.
Results
1. Digitized biliary tract anatomy study based on 64 multi-detector helical CT scanning data
1.13D reconstruction of biliary tract by 64 multi-detector helical CT
The 3D reconstruction of biliary tract was successfully performed in 54 out of 60 patients with cholelithes. The 3D reconstruction of intrahepatic ducts was unsatisfactory in 6 patients due to unobvious distention of biliary tract system. Diverse shapes of calculi in intrahepatic ducts were observed in the cross section of plain CT scanning and appeared as mottled, quasi-circular, striped or beadlike high density images. The position of calculi and nearby bile ducts showed various degrees of stenosis and distention, combination with atrophy of Liver lesion or compensatory hypertrophy of surrounding liver tissue. Extrahepatic cholelithes were marked by rounded or annular high-density shadow surrounded by low density gall images and appeared as "target sign" or "meniscus sign".
Through 3D reconstruction of biliary tract, single, multiple calculi or disseminated distribution of high density intrahepatic calculi in different subsections of liver could be observed, complicating with various distention of intrahepatic ducts. The bile ducts at the calculus region were narrowed,and the distal bile ducts branches dilated like "biliary tree". Sometimes the dilated bile ducts became a "Gall Lake" or regional hepatapostema with different density. MinIP could show not only high density calculi and dilated intra- and extra-hepatic bile ducts, but also present radioparent calculus with soft tissue density in the choledoch. The dilated bile ducts could branch up to three to four levels and mostly appeared as a "dry stick". Commonly, the distension of bile ducts near porta was more serious than that of bile ducts far from porta, while distension or stenosis of intrahepatic duct was regional and the branch was sparse and slim. Besides the shape and volume of high density calculi in the intrahepatic ducts, the lesion liver prenchyma atrophied and hyperplastic complex was also observed. As far as complicated choledocholithiasis was concerned, single or multiple beadlike high density calculi images with various shapes and volume could be seen in the choledoch. The distension of bile ducts at upside obstruction was obvious, while distal bile ducts did not show or suddenly obstructed. Oledocholithiasis was marked by distension of all bile ducts above calculus as well as cholecyst; intrahepatic ducts dilated to different extent and appeared as a "dry stick". The choledoch under calculus was gradually narrowed, while dilated in full range in complicated pyogenic cholangitis.
1.23D reconstruction of biliary tracts by MI-3DVS
1.2.1 3D reconstruction of live, intra- and extra-hepatic vascular system
The liver model can truly reflect the actual volume and anatomy marker of liver, and simultaneously present liver and branches of intrahepatic artery, vein and portal vein through adjusting the transparency of liver. The model presented vivid shapes and stereoscopic identification of various structures, including abdominal aorta, celiac arteries and branches, such as gastroduodenal artery, right and left renal artery, right and left gastric artery, splenic artery, part of mesenteric artery, proper hepatic artery, right and left hepatic artery as well as intrahepatic branches, and so on. The model of the portal system presented portal trunk, splenic vein and superior mesenteric vein out of liver, and the intrahepatic portal system could clearly present right and left main portal vein as well as branches in various hepatic lobes and subsections in the liver. The model of the hepatic vein system could clearly present the intrahepatic distribution of three hepatic veins and their branches, the confluence of three hepatic veins, suprahepatic and inferior vena cava as well as the spatial anatomy relationship of all hepatic veins.
1.2.2 3D reconstruction of biliary tract system and calculus
Aided by plain CT scanning images and portal phase scanning images, MI-3DVS medical images segmentation with adaptive region growing algorithm usually could segment biliary tract system completely in single-time. For some fine structures such as the region where cystic duct and choledoch converge, and the hilar bile duct without expansion, scanning data of different blood vessel phases including plain scanning phase, arterial phase, venous phase and portal phase could be used for image segmentation of biliary tract system of various phases. The specific anatomy information in various phases could be integrated into an complete 3D image of biliary tract by system automatic registration function. The reconstructed biliary tract system mode could truly reflect the spatial ubiety relationship among calculi, biliary tract system and intra- and extra-hepatic blood vessels. After the 3D biliary tract system model was completed with celiac viscera and blood vessel model, the stereoscopic relationship of the whole epigastric viscera, the hepatic vascular system and the biliary tract system was completely presented.
After segmentation and reconstruction by MI-3DVS and through accurate segmentation of biliary tract system and calculus, the 3D reconstruction images of livers as well as intrahepatic duct system of 54 patients in this study were presented with vivid shapes and stereoscopic effects. These images clearly presented the 3D shape of patient liver and indicated the existence of megalohepatia or hepatatrophia, the 3D shape of three levels of intrahepatic bile duct tree, as well as the length and diameter of stenotic or dilated bile ducts. In part of patients with hepatolithiasis combined with extensive intrahepatic duct distension, the forth level of bile ducts, e.g., bile ducts in hepatic subsegments, could also be shown. Besides, the volume, number, in intra- and extra-hepatic 3D distribution of calculi as well as the spatial structure of bile ducts and intrahepatic vascular system could be clearly presented too.
2. Study on pathology classification of hepatolithiasis by 3D visualization technique
Typing diagnosis of hepatolithiasis was carried out by combining multiple factors including the location of calculi orlesion bile ducts (location, L), the degree of biliary stenosis (stenosis, S), the degree of biliary distention (distention, D), the atrophy of lesion liver subsection (atrophy, A), the portal hypertension (PHT), and so on. For example:
Hepatolithiasis (LⅡ,Ⅲ, SⅢ1, DⅢ2, AⅡ,Ⅲ):indicating lithiasis in theⅡandⅢsubsection of liver, combined with slight stenosis and distention of intrahepatic duct in theⅢsubsection as well as the atrophy of theⅡandⅢsubsections of liver.
Hepatolithiasis (LⅥ,Ⅶ,SⅥ1, DⅥ0, AⅥ):indicating lithiasis in theⅥandⅦsubsection of liver, complicating with slight stenosis ofⅥsubsection of bile duct without distention of remote bile duct, as well as atrophy of theⅥsubsections of liver.
The liver subsections of patients with hepatolithiasis based on the technique of 3D visualization complicating wit the individual characters of liver anatomy. It brings out accurate position diagnosis for the distribution of calculi and lesion regions of bile ducts which combined with the degree and range of biliary stenosis or distention and having or not biliary cirrhosis, hepatatrophia as well as megalohepatia, could draw out relatively reasonable clinicopathologic analysis could be drawn out. This new analysis method considering multi-factors including distribution of calculi, regions of biliary stenosis or distention, liver volume variation and cirrhosis and so on, thus could make out more accurate position diagnosis for hepatolithiasis and the lesion region of bile duct, and is more practically significant in clinic guidance for determining the excision range and the surgical processing modes such as bile tract drainage.
3. Application of 3D reconstruction technique in diagnosis and therapy of hepatolithiasis
3.1 Definite 3D shape of intrahepatic "biliary tree" and "blood vessel tree" by 3D reconstruction
The 3D shape of liver model of patients in this study has vivid shape and correct anatomy structure marker. After Transparency processing on liver, through single or various combination of video mode, it will be clearly observed the existence of hepatatrophia,3D distribution and shape of intrahepatic ducts and blood vessels, volume and distribution range of calculi in intrahepatic ducts, the degree of biliary stenosis and distention, having or not combination with extrahepatic cholangiolithiasis. By virtue of the 3D model, not only the distribution of intrahepatic calculi, the pathological change of biliary system and liver parenchyma as well as the relationship between disease focus and surrounding tissues could be observed, but also the blood supply patterns of liver and variation types of hepatic arteries could be determined, thus the typing diagnosis and virtual operation of hepatolithiasis could be precisely carried out.
3.2 Simulated operation of hepatolithiasis
In the simulated environment of operation, the tissues and organs could be observed in any angle through magnifying, minifying, rotating and hyalinizing the 3D model. The distribution of calculi, the pathologic lesion of biliary tract system and liver as well as changes in intrahepatic blood vessel tree could be determined. Based on these observations, clinic classification of hepatolithiasis and choice of optimal operation mode were carried out. The operation process mentioned above was simulated in turn to observe the surgical outcome. In this study, simulated operations were carried out on 30 cases of complex hepatolithiasis, while other 30 patients with positively diagnostic type I hepatolithiasis or hepatolithiasis without stenosis of bile duct did not performed simulated operations. After observing the dissection relationship of important blood vessels and bile ducts in the excision plane of 30 patients by simulated operation, the optimal operation plan were formulated to direct the actual operation process.
3.3 Clinic application of 3D visualization technique of biliary tract
Based on the observation and analysis of 3D video and sectional drawing as well as drilling by simulated operation on 30 patients with complex hepatolithiasis, operation plans were made on all patients. The actual plan included 30 cases of BDE+LLL+HJS,8 cases of BDE+LLL+IBDSTP,3 cases of LLL+RPL+HJS,3 cases of LH+HJS, and 10 cases of LLL+ HJS+1BDSTP. The reconstruction model of liver complied with the actual operation condition in 54 patients well. The accordance rate between simulated operation plans and actual operation was 90% (27/30. The radical surgery was difficult to be carried out on 3 emergency patients). No serious complications were observed post-operation, and no case of death. Postoperative radiography of bile duct on 51 patients presents no residual of calculus. Lithiasis relapsed in 3 patients after half a year, and the recurrence rate in half a year was 5.58%(3/51).
Conclusion
1. Technique of 3D reconstruction can realize the digitized anatomy of biliary tract.
Both the CT 3D reconstruction by 64 multi-detector helical CT MxliteView workstation or by MI-3DVS, the 3D reconstruction can achieve the digitized anatomy of biliary tracts. Compared with 2D CT imaging, the digitized dissection can show the form of the bile duct tree of the biliary tract system and the distribution of calculi in three-dimension. However, the 3D images reconstructed by the CT MxliteView workstation were only the biliary tract images in a certain blood vessel phase. which only showed the 2D sagittal or coronal plain films of the 3D images in vision to surgeons,but not the 3D images in the true sense. It is difficult for these 2D plain films to present the 3D spatial relationship among blood vessels, including hepatic artery, portal vein, hepatic vein, and the biliary tree simultaneously.Thus it is difficult to observe the spatial dissection of bile ducts, calculi and the intrahepatic vascular system in various angles or directions. However, MI-3DVS could render the liver surrounding viscus, celiac blood vessels, and various intrahepatic duct systems with different colors. Then through the visualization technique the complete 3D images of the whole epigastric viscus are obtained. In these 3D images, the liver, hepatic artery, hepatic vein, portal vein, celiac blood vessels as well as surrounding viscus could be presented independently, or simultaneously, or in various combinations. The pattern of intrahepatic duct tree, distribution of calculi, as well as the location and degree of biliary stricture could be clearly presented through partially magnifying or rotating this 3D image. Compared with the traditional black-and-white 2D image, the 3D image presents the critical dissection structures including the region of lesion as well as surrounding blood vessels and viscus with higher definition and visualization, and realizes the 3D presentation in the true sense and thus will contribute to accurate preoperative diagnosis for hepatolithiasis.
2.3D visualization can realize precise classification diagnosis of hepatolithiasis.
The liver subsection of patients with hepatolithiasis based on the technique of 3D visualization complies with the individual characters of liver anatomy. It brings out accurate position diagnosis for the distribution of calculi and lesion regions of bile duct. Combining the degree and range of biliary stenosis or distention and having or not biliary cirrhosis, hepatatrophia as well as megalohepatia, relatively reasonable clinicopathologic analysis could be drawn out. This new analysis method considers multi-factors including distribution of calculi, regions of biliary stenosis or distention, liver volume variation and cirrhosis and so on, thus could make out more accurate position diagnosis for hepatolithiasis and the lesion region of bile duct, and is more practically significant in clinic guidance for determining the excision range and the surgical process such as bile tract drainage.
3.3D visualization technique plays an important guide in the treatment of hepatolithiasis.
3D visualization of biliary tract and virtual operation could realize accurate preoperative diagnosis of hepatolithiasis and simulated operation; formulate the optimum operation plan through comparing various plans and surgical pathways or methods of critical steps, thus implementing accurate surgical operation. For these reasons, the 3D visualization technique has an important reference value in reasonable choice of operation mode for complex hepatolithiasis.
肝内胆管结石又称为肝胆管结石病,是指肝总管、左右肝管汇合部及其肝内各级胆管内的结石,同时合并胆管的炎症、狭窄、肝纤维化、萎缩及肝功能障碍等多种并发症。肝胆管结石是世界性的疾病,以亚洲国家多发。在我国华南、西南、长江流域及东南沿海等广大区域尤为多见。由于其病变复杂、复发率高且常引起严重的并发症,成为我国良性胆道疾病死亡的主要原因。手术是肝胆管结石的主要治疗方式,目前公认的手术方式为肝切除术,同时辅以各种形式的胆管整形和胆肠吻合术。但由于肝内胆管结石的分布广泛,同时合并不同程度的肝胆管狭窄和肝脏毁损性病变,术后肝内胆管狭窄和引流不畅易致结石残留及复发,而使得肝胆管结石的疗效难以满意。
肝胆管结石病的诊断主要依靠B超、CT和MR等无创性影像学检查,以明确结石的部位、大小、数量和分布情况,胆管狭窄的部位及程度,从而进行定位诊断。并结合患者肝功能状况、Oddi括约肌功能、有无肝胆管炎发作、有无梗阻性黄疸、有无肝门部胆管狭窄、有无门静脉高压症和脾肿大等综合因素制定手术方案。但由于肝内胆管结石的分布广泛,同时合并不同程度的肝胆管狭窄和肝脏毁损性病变,上述各种影像学检查手段各有优缺点,常需几种影像学检查方法的综合运用,才能做出较为全面的诊断。目前尚无一种理想的诊断方法能够对结石大小、数量和分布,胆管狭窄程度和长度,肝脏病理形态,以及胆管和血管的关系做出系统全面的诊断。
目前,国际上尚无统一、公认的肝胆管结石分型标准。对肝胆管结石和狭窄常用的分型方法目前有三种：一种是日本Nakayama分型,一种是Tsunoda分型,还有一种是中华医学会胆道外科学组2007年的中国分型。上述各种分型的依据均基于结石的分布、胆管狭窄及扩张的部位和程度等。由于影像学检查手段的局限,上述分型对结石分布位置、胆管狭窄部位、程度、长度等均缺乏精确的立体定位,缺乏胆管与门静脉、肝静脉解剖关系的详细描述,因此难以制定合理化的手术方案,实现肝胆管结石的术中精准治疗。
近年来,随着计算机三维重建和可视化技术的发展,3D技术已经成功应用于人体解剖结构和组织功能的研究,并应用于疾病诊断和术中导航。2003年本课题组采用数字化虚拟中国女性一号肝脏数据集和肝脏管道灌注标本CT薄层扫描数据进行三维重建和虚拟手术研究,随后开始利用活体人64排螺旋CT扫描数据进行肝胆脾胰等脏器以及腹腔血管的计算机三维重建及虚拟可视化手术的研究。由于三维重建和虚拟手术是基于患者特征性,因此实现了个体化的要求。本课题组目前已对肝胆系统多种疾病,包括肝癌、肝血管瘤、胰腺癌等实现了个体化的三维重建和虚拟仿真手术,显示了良好的临床应用价值。本研究拟将三维可视化技术应用于肝胆管结石病胆道数字化解剖,临床病理分型以及个体化的外科治疗研究,以期进一步提高肝胆管结石病的诊断精准率、手术治疗精准性,从而提高肝胆管结石病的临床疗效。
目的
1、3D技术进行胆道系统数字化三维重建。分别采用PHILIPS CT自带图像后处理Mxview工作站和腹部医学图像三维可视化系统(MI-3DVS)进行肝胆管结石患者肝脏及肝内管道,尤其是结石和胆管的三维重建,实现肝脏胆道系统的数字化解剖。
2、将三维重建和仿真技术引入到肝胆管结石的临床诊断、分型研究以及手术治疗中,探索其在肝胆管结石手术方式合理选择和控制性肝切除治疗上的临床应用价值。
方法
1.基于64排螺旋CT扫描数据的数字化胆道解剖研究
1.1设备
64排螺旋CT—PHILIPS Brilliance 64(荷兰PHILIPS公司),PHILIPS Brilliance64层螺旋CT自带的图像后处理Mxview工作站,FreeForm Modeling System(美国SensAble Technologies公司),力反馈设备PHANToM (PHANToM Desktop), ACDSee9 Message Center (ACD System Ltd.),自主研发医学图像三维可视化系统(MI-3DVS)(主机：内存2G,处理器2.0G*2 Xeon 5130)。
1.2胆道系统结石病例CT扫描参数设定及数据存储
常规平扫时患者取仰卧位,头足方向,由膈顶至双肾下缘。扫描条件：120KV、250mAs,采用0.625×64排探测器组合,层厚5 mm、间隔5mm,螺距0.984,球管旋转一周时间0.5s。患者检查前禁食至少6-8h,扫描前30min口服500 ml清水,作为阴性胃肠道对比剂。先行平扫后再进行增强扫描,平扫最大范围从气管分叉部至耻骨联合上缘水平。增强扫描经肘前静脉采用自动高压注射器经前臂静脉进行团注非离子碘造影剂优维显(ultravist),剂量1.5ml/kg,速度5ml/s,注射对比剂后21～25s启动肝动脉期扫描,30～35s行动脉晚期扫描,50～55s行门静脉期扫描,每期扫描时间约6～8s。扫描结束后将图像数据传至Mxview工作站,在Mxview工作站进行三期数据(平扫期、动脉期和门脉期)的刻盘存贮。数据格式为DICOM(Digital Imaging and Communications in Medicine)3.0,通过DICOM查看器转换为BMP格式。
1.3 PHILIPS Brilliance 64层螺旋CT Mxview工作站上胆道三维重建
对原始数据库中层厚5mm的扫描数据进行薄层重建(层厚0.67mm,间隔0.33mm),薄层处理图像传至Maxview工作站。在Maxview工作站中,对薄层图像数据进行胰胆管和腹腔血管冠状位、矢状位、任意平面或曲面三维重建。主要重建方法包括：最大密度投影法(maximum intensity projection, MIP)、最小密度成像法(minimum intensity projection, MinIP)、曲面重建法(curve reconstyuction, CR)、容积重建(volume rendering, VR)和多平面重建(Inultiplanar reformation, MPR)等。重建后多角度,多方位观察胆管解剖结构,并截图保留。
1.4 MI-3DVS肝胆管系统三维重建
利用Mxview Viewe rDICOM (?)阅读64排CT扫描数据,调整适当的窗宽和窗位,并将以DICOM数据导出并转化成JPEG格式存盘,将JPEG文件导入ACDSee 9 Message Center转化成BMP文件,并调整图像大小后导入自主开发的MI-3DVS中,以自适应的区域生长法对肝脏、胆道系统及其周围血管系统进行序列分割,得到分割后的肝脏、胆道系统、腹主动脉、腹腔动脉及其分支、门静脉、肝静脉系统等STL(STereo Lithography)格式数据。重建后的STL模型导入FreeForm Modeling System进行平滑和修饰,得到光滑逼真的肝脏及其内部管道系统的三维图像。立体观察胆道系统解剖结构,录制视频及多角度视频截图。
2、肝胆管系统三维可视化技术对肝脏分段和肝胆管结石分型研究
2.1仪器与设备同“1.1”
2.2胆道系统三维重建方法及步骤同1.1、1.2、1.4。
2.3基于肝静脉和门静脉解剖的肝胆管结石的肝脏个体化分段
在肝脏及门静脉、肝静脉、胆管和肝动脉三维重建基础上,肝段的划分以门静脉为指示标志,肝静脉为分界标志,胆囊和韧带为辅助分段标志,每一肝段均有一(组)独立的门静脉段支供应。在肝段划分和病变的定位诊断中,门静脉的解剖及病灶与门静脉的关系更为重要。将自门静脉主干发出的分支定义为一级分支,三级分支所供应的肝脏范围定义为一个肝段。根据具体的门静脉三级分支的数量,将肝脏按罗马数字分为I-X个不等的肝段。由于Gelisson系统包括门静脉和胆管,因此可将肝胆管结石所在部位和狭窄或扩张胆管的位置进行精确定位诊断。
2.4基于个体化分段的肝胆管结石解剖分型
利用三维可视化肝脏分段结果,参考中华医学会胆道外科学组2007年分型方法日本分型(?)(?)Tsunoda的分型依据,从如下几个方面进行肝胆管结石的解剖分型。
2.4.1结石或病变胆管分布位置(location)
采用罗马数字的Ⅰ-Ⅹ肝段定位结石和胆管病变(狭窄或扩张)的部位。
2.4.2胆管狭窄程度(stenosis)
由于正常二级胆管直径为2-3mm,因此以2mm作为绝对狭窄的标准。相对狭窄判断标准,近端胆管狭窄与远端扩张胆管的直径比>2/1为相对狭窄(轻度),而<1/2为绝对狭窄(重度),相对和绝对狭窄均需外科手术解除。狭窄分为S0(无狭窄)、S1(轻度狭窄)、S2(重度狭窄)
3.4.3胆管扩张程度(distention)
以肝内胆管内径10mm作为轻度或重度扩张的分界,>2mmm且<10mm为相对扩张(D1)>10mm为扩张(D2),>15mm为显著扩张(D3),。扩张分为0-3级别。D0(无扩张)、D1(轻度扩张)、D2(中度扩张),D3(重度扩张)。
2.4.4病变肝段体积萎缩(atrophy)
通过肝脏三维重建结果观察病变肝段体积的变化,根据有无病变肝脏体积的萎缩或邻近肝段体积的代偿性肥大。对于萎缩肝脏,手术切除为唯一选择,因此,将病变肝段萎缩作为病理分型的指标之一,可以帮助术前确定手术方式为肝段切除还是肝内胆管切开取石。
2.4.5门脉高压症(portal hypertension, PHT)或肝硬化(cirrhosis)
胆汁淤积性肝硬化导致门脉高压症、脾脏肿大是影响肝胆管结石的预后的重要因素之一,常影响肝胆管结石的手术方式的选择。对于存在多处肝内胆管结石合并胆汁淤积性肝硬化,手术方式应选择各种方式的切开取石和内外引流术,肝切除应慎重。三维可视化技术可清晰观察到有脾脏肿大,脾周血管的扩张,肝脏萎缩-肥大综合征等影像改变。因此将肝硬化或门脉高压症列为病理分型指标之一。
其他因素,如有无Oddi括约肌松弛型功能失调,有无肝胆管炎发作,有无需急于处理的梗阻性黄疸、肝脓肿、胆道出血,有无合并肝胆管癌等,虽然对确定手术的时机和确定手术的类型有所帮助,但三维可视化技术难以从影像学上作出相应的判断,因此不列为分型因素。
3、3D技术在肝胆管结石诊治中的应用
3.1研究对象：收集我院自2008年10月至2010年10月60例肝胆管结石患者64排螺旋CT数据。所有患者进行三维重建和病理分型,并进行仿真手术,确定手术方式和手术切除部位,进行临床实际手术,观察术前三维与术中所见、术前仿真手术与实际手术操作符合情况。
3.2仪器与设备同“1.1”
3.3胆道系统三维重建方法及步骤同1.1、1.2、1.4。
3.4图像后处理
将重建后的STL三维模型导入FreeForm Modeling System中,进行光滑、去噪,得到光滑逼真,立体感强的三维模型。为便于立体观察,利用MI-3DVS自带着色功能,分别将肝脏、肝静脉、肝动脉、门静脉、胆道、结石以及腹腔血管和周围脏器渲染不同的颜色,同时采用不同程度的透明化处理或脏器隐去技术,分别观察肝脏和胆道及结石、肝脏和动脉,肝脏和门静脉、肝脏和肝静脉等不同组合的三维解剖关系。通过三维模型的旋转,观察病变部位与相邻脏器的不同角度的空间解剖,分别录制视频、截图保留。在本研究中,主要观察胆管树在肝脏内的立体形态、结石在胆道系统的分布、胆管狭窄的部位和程度、胆道系统与肝动脉、门静脉和肝静脉系统的三维立体关系。
3.5肝内胆管结石的虚拟手术
在FreeForm Modeling System中对三维重建模型及各组成部分进行放大、缩小、旋转、透明等操作,全方位观察各结构或细节,根据结石分布、胆道系统及肝脏病变情况选择合理手术预案,利用力反馈设备PHANTOM和自行开发设计的开腹手术器械进行肝内胆管结石的各种虚拟仿真手术,以确定最佳手术方案。
3.6肝内胆管结石的临床手术
根据三维模型视频和截图的多角度旋转观察,明确肝胆管结石在肝脏内的大小、数量、部位等信息,同时观察有无肝胆管狭窄及扩张、狭窄胆管的长度和程度、有无肝脏萎缩-肥大等信息,进行术前诊断和分型,通过仿真手术制定最佳手术方案。将三维可视化视频及截图的带入手术室,对照术中实际解剖情况,实时指导手术操作。常用手术方法有胆道探查取石(bile duct exploration,BDE)、部分肝叶切除(hepatectomy,HT)、左外叶切除(Left lateral lobectomy, LLL)、右后叶切除(right posterior lobectomy, RPL)、左半肝切除(left hepatectomy,LH)、肝(胆)肠吻合(hepaticojejunstomy,HJS)、肝内胆管置入支撑管(intrahepatic bile duct support tube placement, IBDSTP)等,具体手术方式为上述手术方法的不同组合。通过实际手术,对比术前三维与术中解剖所见,术前仿真手术与实际手术操作情况,评价3D技术对肝胆管结石精准治疗的的临床应用价值。
结果
1、基于64排螺旋CT扫描数据的数字化胆道解剖研究
1.1 64排螺旋CT胆道三维重建
60例胆道结石疾病患者,胆道三维重建成功54例,6例因无明显胆道扩张,肝胆管三维重建不理想,但横截面CT可清楚显示结石病灶。平扫横截面观察肝内胆管结石形态多种多样,呈斑点状、类圆形、条状或串珠样高密度影。结石部位及其近端胆管多有不同程度的狭窄及扩张,伴随病变肝组织萎缩或周围肝组织代偿性增生。肝外胆管结石表现为胆总管内有圆形或环形致密影,周围被低密度胆汁环绕,形成“靶征”或“半月征”。
胆道三维重建后,可见肝内单发、多发或弥散性分布于不同肝段内的高密度结石,伴随肝内胆管不同程度的扩张。结石所在部位胆管狭窄,远端胆管及其分支扩张,扩张胆管呈“树枝状”,有时扩张的胆管成“胆汁湖”或形成局限性密度不均的肝脓肿。MinIP除可观察到高密度结石及扩张的肝内外胆管外,还可显示胆总管内软组织密度的阴性结石、扩张的胆管分支可达3-4级,扩张胆管多为“枯枝状”。一般近肝门胆管扩张明显,而肝内胆管为局限性扩张或狭窄,胆管分支稀少,纤细。还可见肝内胆管高密度结石形态大小外,还可观察到病变肝实质萎缩-增生复合征。对于合并胆总管结石,胆总管内可见不同形态、大小的单个或呈串珠状多发高密度结石影。梗阻上段胆管扩张明显,远端胆管不显影或突然截断。胆总管结石表现为结石以上胆管全部扩张,胆囊亦扩张,肝内胆管不同程度扩张呈“枯枝状”,结石以下胆总管扩张逐渐变细,合并化脓性胆管炎则全程扩张。
1.2 MI-3DVS三维重建胆道解剖
1.2.1肝脏及肝内外血管系统的三维重建
肝脏模型能真实反映肝脏的实际体积和肝脏的解剖标志,并且通过调节肝脏的透明度可同时显示肝脏和肝内的动脉、静脉、门静脉各分支。腹主动脉、腹腔动脉及其分支胃十二指肠动脉、双肾动脉、胃右、左动脉、脾动脉、肠系膜部分动脉、肝固有动脉、左右肝动脉及其肝内分支等结构,形态逼真,立体感强。门静脉系统显示肝外的主干和脾静脉、肠系膜上静脉；肝内门静脉系统能清楚显示门静脉的左主干和右主干,以及各叶、段的分支。肝静脉系统能清晰显示三支肝静脉在肝脏内部的分支分布,三支肝静脉与肝上下腔静脉的汇入情况、各肝静脉之间的空间解剖等。
1.2.2胆道系统及结石的三维重建
利用平扫期的CT图像和门静脉期的CT图的辅助,MI-3DVS自适应区域生长法图像分割法通常能一次性完整分割胆道系统。对一些细微结构如胆囊管与胆总管汇合部、不扩张的肝门部胆管等,可利用不同血管期如平扫期、动脉期、静脉期、门静脉期等,分别对各期胆道系统进行图像分割,再利用系统自动配准功能将各期所含特定解剖学信息整合为完整的胆道三维图像。重建的胆道系统模型能真实反映结石。胆道系统与肝内外血管的空间位置关系。胆道系统三维模型与腹腔脏器及血管模型配准后,整个上腹部脏器及肝脏血管系统和胆道系统的立体关系完整重现。
经MI-3DVS分割重建后,本组54例患者肝脏及其内部管道系统三维重建图像逼真,立体感强。可清晰再现患者肝脏的立体形态和有无肥大和萎缩；肝内一、二、三级胆管树立体形态,以及狭窄或扩张胆管的长度和直径,部分肝内胆管结石合并肝内胆管广泛扩张者,甚至四级胆管(亚肝段胆管)也得以显示；结石的大小、数量、在肝内外胆管的立体分布,以及胆管和肝内血管系统的空间结构情况也清晰可见。
2、肝胆管系统三维可视化技术对肝胆管结石病理分型研究
采用包括结石或病变胆管分布位置(location, L)、胆管狭窄程度(stenosis,S)、胆管扩张程度(distention,D)、病变肝段体积萎缩(atrophy,A)、门脉高压症(portal hypertension, PHT)等因素进行了肝胆管结石的分型诊断,举例如下：例1.肝内胆管结石症(LⅡ、Ⅲ,SⅢ1,DⅢ2,AⅡ、Ⅲ):表明肝第Ⅱ、Ⅲ段结石,同时伴
有Ⅲ段肝内胆管的轻度狭窄和扩张,Ⅱ、Ⅲ段体积萎缩。例2.肝内胆管结石症(LⅥ、Ⅶ,SⅥ1,DⅥ0,AⅥ):表明肝Ⅵ、Ⅷ段结石,Ⅵ段胆管轻度
狭窄,无远端胆管扩张,Ⅵ段体积萎缩。
基于三维可视化技术的肝胆管结石患者肝脏分段符合个体化肝脏解剖特征,对结石分布和胆管病变部位做出了精确的三维定位诊断。结合胆管狭窄和扩张程度范围、有无胆汁性肝硬化和肝脏萎缩肥大等,做出了较为合理的分型诊断。新分型方法考虑到了结石和胆管狭窄和扩张部位、肝脏体积变化和肝硬因素,不仅对肝胆管结石和胆管病变的定位诊断更加精确,对制定手术切除范围、胆道引流术式等外科处理方式有更加实际的临床指导意义。
3、3D技术在肝胆管结石诊治中的应用
3.1三维重建肝内“胆管树”和“血管树”立体形态清晰
本组患者肝脏三维重建的立体模型形态逼真、解剖结构标志正确。在肝脏透明化处理情况下,通过单独或不同组合的联合显示方式,可清楚观察到病变肝脏有无萎缩、肝内胆管和血管的的立体分布形态、结石的大小及在肝胆管内的分布范围、胆管狭窄程度和范围、有无合并肝外胆管结石等。3D模型除了观察到肝内结石分布、胆管系统及肝实质的病变情况以及病灶与周围组织的相互关系,还可明确肝脏的血供类型及血管变异情况等,从而准确进行肝胆管结石的分型诊断和仿真手术。
3.2肝胆管结石仿真手术
在虚拟手术环境中,立体模型能通过放大、缩小、旋转及透明化进行观察组织结构的组织器官,可明确肝内结石分布以及胆道系统与肝脏病理改变情况,以及肝内血管树的改变,据此进行临床分型及选择最佳的手术方式,进而虚拟上述各种手术过程,观察手术效果。本组54例肝胆管结石患者选择30例进行仿真手术,其余诊断明确的的Ⅰ型和不伴有胆管狭窄的24例肝胆管结石患者未进行仿真手术,30例患者经仿真手术观察切除平面内重要血管和胆管的解剖关系,制定了最终优化手术方案,指导实际手术过程。
3.3胆道三维可视化技术临床应用
根据三维视频和截图的观察分析,以及仿真手术的演练,54例患者均制定了手术方案。实际手术方式包括BDE+LLL+HJS30例,BDE+LLL+IBDSTP 8例,LLL+RPL+HJS3例,LH+HJS 3例,LLL+ HJS+IBDSTP 10例。54例病人肝脏重建模型与术中所见均符合。30例仿真手术方案与实际手术方式符合者为90%(27/30,3例急诊患者难以行根治性手术)。所有患者术后未出现术后严重并发症,无死亡,51例非急诊患者术后胆管造影未见结石残留,3例患者半年后结石复发,半年复发率5.58%(3/51)。
结论
1、三维重建技术可实现胆道数字化解剖
基于64排螺旋CT扫描数据,CT自带Mxview工作站和MI-3DVS均能实现胆道数字化解剖。相对于二维CT图像,3D图像能立体显示胆道系统的胆管树形态和结石分布情况。但CT工作站的三维重建图像仅是某一血管期的胆道像,且提供给临床医生的仅为三维图像的矢状位或冠状位的二维平片,并非真正意义上的三维立体图形。难以将动脉期、门静脉期及静脉期的三期血管与肝脏及胆管树的立体解剖关系同时显现,难以不同角度和方向观察病变胆管及结石与肝内血管系统的空间解剖。MI-3DVS可将肝脏、周围脏器、腹腔血管、肝内不同管道系统分别渲染不同颜色,通过可视化技术,获得了整个上腹部脏器的整体立体图像；并能将肝脏、肝动脉、肝静脉、门静脉、腹腔血管以及周围脏器等单独、同时或不同组合方式分别显现；通过局部放大、旋转立体观察,肝内胆管树形态、结石的分布、胆管狭窄部位和程度可清晰显示。与传统的黑白二维图像比较,三维图像对病变部位与周围血管及脏器等重要解剖结构的显示更加清晰直观,实现了真正意义上的三维立体显现,对肝胆管结石可作出精确的术前诊断。
2、三维可视化科实现肝胆管结石精确分型诊断
基于三维可视化技术的肝胆管结石患者肝脏分段符合个体化肝脏解剖特征,对结石分布和胆管病变部位做出了精确的三维定位诊断,结合胆管狭窄和扩张程度范围、有无胆汁性肝硬化和肝脏萎缩肥大等,做出了较为合理的病理分析。新分析方法考虑到了结石和胆管狭窄和扩张部位、肝脏体积变化和肝硬因素,不仅对肝胆管结石和胆管病变的定位诊断更加精确,对制定手术切除范围、胆道引流术式等外科处理方式有更加实际的临床指导意义。
3、三维可视化技术对肝胆管结石的临床治疗有重要指导价值
胆道三维可视化和仿真手术能够实现肝胆管结石的术前精确诊断,通过术前仿真手术反复演练,针对不同手术方案和不同手术方法和路径的比较,制定最优手术方案,指导术中精确操作。三维可视化技术对复杂肝胆管结石手术方式的合理选择上具有重要参考价值。
Background
Hepatolithiasis, also called calculosis, is the presence of gallstones in the common hepatic duct, the region of hepatic biliary confluence as well as all levels of intrahepatic bile ducts with accompanied complications such as inflammation or stenosis of bile ducts, hepatic fibrosis, hepatatrophia, liver dysfunction, and so on. Hepatolithiasis is a worldwide disease and is prevalent in Asian countries, especially in South China, Southwest China, the Yangtze valley and southeast coastal areas. Due to the complicated pathological changes, relative high recurrence rate and severe accompanied complications, hepatolithiasis is the main cause of death from benign hepatobiliary diseases. As the main therapy, surgical treatment, especially hepatectomy is widely accepted with the aid of bile-duct anaplasty and choledochojejunostomy. However, due to extensive distribution of intrahepatic duct calculus with accompanied different degrees of devastated liver lesion and stenosis of hepatic ducts, stenosis of bile duct and inadequate drainage-caused recurrence of calculosis after surgical operation, surgical treatment could not always lead to satisfactory results.
The diagnosis of hepatolithiasis is mainly relied on non-invasive imaging examination, such as type-B Ultrasound, computed tomography (CT) and magnetic resonance imaging (MR), etc. The classification diagnosis is based on the detection of the location, size, number and distribution of calculi and the location and degree of stenosis of bile ducts. To make a surgery plan, diverse and combinatory factors should be considered, such as the function of liver and Oddi sphincter, emergence of hepatocholangeitis, obstructive jaundice, hepatic portal bile duct stricture, portal hypertension or splenomegaly and so on. Due to the extensive distribution of calculi, different degrees of stenosis of intrahepatic ducts and devastated liver lesions, each of the above-mentioned imaging examinations has its advantages and disadvantages. Such situation requires the combinatory application of several imaging examinations to get final comprehensive diagnosis. Currently, there is no single perfect method can reach a final systematic and comprehensive diagnosis conclusion on the number and distribution of calculi, the degree and length of stenosis of bile ducts, the liver pathomorphism, the relationship between bile duct and blood vessel.
Currently, there isn't a worldwide-recognized classification criterion of hepatolithiasis. There are only three classification methods have been commonly used for the classification of hepatolithiasis, including the Japanese Nakayama classification, the Tsunoda classification and the Chinese classification which was proposed by the biliary-tract surgery group of Chinese Medical Association (CMA) in 2007. The above classification methods are all based on the distribution of Calculi and the location and degree of cholangiectasis and stenosis of bile ducts. However, these methods are lack of precise stereotaxis of the distribution of Calculi and the location and degree of cholangiectasis due to the limitation of imaging examinations, which also can not give detailed anatomical relationship among bile ducts, portal vein and hepatic vein. Thus it's hard to design a suitable surgery plan to achieve the precise operation for hepatolithiasis.
In recent years, with the development of computerized techniques for three-dimensional (3D) reconstruction and visualization technology,3D visualization technique have been successfully applied to the studies of human body anatomy structure and organ functions, and also to the clinical applications of disease diagnosis and intra-operative navigation. Since 2003, our group used the hepatic CT scan data set from digitalized virtual Chinese Female No. one and specimens liver with perfusion of hepatic ducts to perform the studies of the 3D reconstruction and virtual surgery. After then, using the data of Live humans liver, gall, spleen, pancreas, other visceral organs and vessels scaned by the 64 multi-detector helical CT, we performed the study of 3D reconstruction and virtual surgery.Because the study of 3D reconstruction and virtual surgery is based on the individual characteristics of patients, it satisfied the individual requirement. Currently, our group has completed the individual 3D reconstruction and virtual simulation of many hepatobiliary diseases, including liver cancer, hepatic hemangioma and pancreatic carcinoma and so on, which shows important value of clinical application. In this work, we apply the 3D visualization technique to study the digital dissection of biliary tract, clinical pathological classification and of individualized surgical therapy for hepatolithiasis, which shall improve diagnostic accuracy and precision of surgical treatment for hepatolithiasis, and finally will contribute to the clinical efficacy of hepatolithiasis.
Objects
1. Establishment of digitalized 3D reconstruction of biliary system. We perform 3D reconstruction of liver and it's internal ducts, especially calculus and bile duct, to establish digitized anatomy of hepatobiliary system by using the CT image-postprocessing Mxview (Philips) workstation and abdomen medical image-3D visualization system (MI-3DVS) respectively.
2. We apply 3D reconstruction and simulation techniques to the clinical diagnosis, classification and surgical treatment of hepatolithiasis, then to explore the clinical value of this techniques for reasonable selection of surgical operation and controlled hepatectomy of hepatolithiasis.
Methods
1. The study of digitalized anatomy of biliary system based on 64 multi-detector helical CT scanning data
1.1 Equipments
64 multi-detector helical CT machine (Holland PHILIPS); The Mxview workstation accompanied with 64 multi-detector helical CT machine; FreeForm Modeling System (USA, SensAble Technologies); Force feedback system PHANToM (PHANToM Desktop); ACDSee9 Message Center (ACD System Ltd.); medical image-3D visualization system (MI-3DVS) developed and set up by ourselves [Host machine containing memory (2G), microprocessor (2.0G×2Xeon 5130)].
1.2 The set of CT scanning parameters and data storage of biliary system from clinical hepatolithiasis cases
When performing routine plain CT scan, the patient was lying down in supine position and scanned from diaphragmatic dome to lower margin of pancreas. The scanning parameter was set as following:120 KV and 250 mAs with 0.625 mm×64 multi-detector combination,5 mm-thick reconstruction layer,5 mm layer intervals, thread pitch 0.984 and 0.5s for one rotation of the tube. The patients were fasted at least 6 to 8 hours before examination. As negative gastrointestinal contrast reagent,500-ml clean water was taken orally by patients. Plain scan was performed from the bifurcation of bronchi to the upper margin of pubic symphysis ahead of contrast enhanced scan which was performed through the antecubital vein and forearm vein using automatic high pressure injector to do bolus injection of nonionic diodone ultravist with the dosage of 1.5ml/kg at a rate of 5ml/s. Hepatic arterial phase scan, late arterial scan and portal phase scanning was performed 21-25s,30-35s and 50-55s respectively after injection of contrast reagent and each of the scan lasted about 6-8s. After the scan, the image data was transmitted to Mxview workstation and all the three phase (plain scan phase, arterial phase and portal phase) data was burned into disc for storage. The data format is in DICOM (Digital Imaging and Communications in Medicine) 3.0 which can be transformed to BMP format through DICOM viewer.
1.3 3D reconstruction of biliary system on the Mxview workstation accompanied with PHILIPS Brilliance 64 multi-detector helical CT
The scanning data of 5-mm-thick layer in the original database were 3D reconstructed using thin slice width reconstruction (0.67-mm thickness and 0.33-mm layer interval). The reconstructed CT images were transmitted to Maxview workstation. In Maxview workstation,3D reconstruction of coronal, sagittal, arbitrary plane surface or curved surface of pancreaticobiliary duct and celiac blood vessels were performed using those thin-thick image data. The main reconstruction methods include maximum intensity projection (MIP), minimum intensity projection (MinIP), curve reconstruction (CR), Volume rendering (VR) and multiplanar reformation (MPR). Thus we can observe the anatomic structure of bile duct through multi-angle and multi-dimension.
1.4 3D reconstruction of hepatobiliary system using MI-3DVS
In this work, the segmentation method combined the algorithms of threshold segmentation and region growing,3D segmentation and two-dimensional segmentation. The scanning data from 64 multi-detector helical CT were read by MxliteView DICOM Viewer, adjusted with suitable window width and level, and exported as JPEG format files. Then the JPEG files were imported into ACDSee 9 Message Center, transformed into BMP files and imported into MI-3DVS after re-sizing the images. In MI-3DVS, the sequential segmentation of biliary tract system and around blood vessel system were performed using self-adaptive region growing algorithm to get the data with Stereo Lithography (STL) format of hepatobiliary system, abdominal aorta, celiac artery and its branches, portal vein and its branches, and hepatic veins, et al. Finally, the reconstructed STL models were imported into FreeForm Modeling System for smoothing and retouching to get the smooth and lifelike 3D images of hepatobiliary system. The video and the multi-angle video recordings were recorded for 3D observation of the anatomic structure of biliary system.
2. Classification of hepatolithiasis using 3D visualization technique of hepatobiliary system
2.1 Equipment same as 1.1
2.2 Methods for 3D reconstruction of biliary tract system:same as 1.1,1.2,1.4
2.3 Individual liver segmentation of hepatolithiasis based on the anatomy of hepatic vein and portal vein
Based on the 3D reconstruction of liver, portal vein, hepatic vein and bile duct, the segmentation of liver is indicated by portal veins as indicative marks, hepatic veins as boundary marks, gall bladder and ligament as auxiliary segmentation marks. Each segment of liver is supported by an independent branch of portal vein. For the segmentation of liver and location diagnosis of lesion, the anatomy of portal veins and the relationship between lesion and portal veins is more important. Normally, the branches stem from main trunk of portal vein was defined as the first level branch, and the part of liver supported by a third-level branches of portal veins is defined as one hepatic segment. Thus we can divide the liver into 1 to 10 hepatic segments named with Roman numerals fromⅠtoⅩ. Because the Gelisson system contains the portal veins and bile duct, we can do accurate location diagnosis of hepatolithiasis and cholangiectasis or stenosis of bile duct.
2.4 Anatomic classification of hepatolithiasis based on the individualized liver segmentation
Based on the Japanese Nakayama classification, the Tsunoda classification and the Chinese classification which was proposed by the biliary-tract surgery group of Chinese Medical Association (CMA) in 2007, we used the results of hepatic segmentation from 3D visualization and classified the anatomy of hepatolithiasis through the following criteria.
2.4.1 Location:using roman-numers-labeled hepatic segmentation to localize the calculi and cholangiectasis or stenosis of bile duct.
2.4.2 Stenosis:Because the diameter of normal bile duct is 2 to 3 mm,2mm was selected as the criterion of absolute stenosis. The criterion of relative stenosis is defined as the diameter ratio of proximal terminal to distal terminal. Thus the stenosis is classified as SO (no stenosis), S1 (mild stenosis is defined as the diameter ratio above 1/2), SO (severe stenosis, also called absolute stenosis, is defined as the diameter ration below 1/2).
2.4.3 Distention:To evaluate the distention degree of bile duct,10 mm was used as the criterion to discriminate the mild and sever distention of the internal diameter of intrahepatic duct. Thus the distention the internal diameter is classified as DO (no distention), D1 (mild distention,>2mm and <10mm), D2 (moderate distention,>10mm), D3 (severe distention,> 15mm).
2.4.4 Atrophy:We observed the volume variation of the pathological hepatic segments through 3D reconstruction to confirm whether there is atrophy of the pathological liver or the compensatory hypertrophy of neighboring hepatic segments. The surgical removal is unique choice for the hepatic segments showing atrophy. So using the hepatic segment with atrophy as a pathological typing index can help to select the surgical operation with surgical excision of hepatic segment or incision removal of intrahepatic calculus.
2.4.5 Portal hypertension (PHT) or Cirrhosis. One of the important factors affecting the prognosis of hepatolithiasis is the portal vein hypertension and splenomegaly caused by biliary cirrhosis, which commonly influence the selection of surgical operation of hepatolithiasis. For the existence of multiple calculi of intrahepatic duct combined with biliary cirrhosis, the surgical operation shall choose incision removal of calculus in different kind of ways and internal and external drainage and be cautious for hepatectomy. The 3D visualization technique can clearly observe the pathological changes of splenomegaly, the distention of around blood vessels, atrophy and hypertrophy of liver. Thus we use portal hypertension as a pathological typing index.
Other factors, such as Oddi's sphincter functional disturbance, hepatocholangeitis, obstructive jaundice, hepatic abscesses, hemobilia and accompanied cholangiocarcinoma, etc, which are helpful to choose the type and timing of surgical operation, but are hard to made a correct diagnosis from imaging analysis.
3. The application of 3D reconstruction technique in the diagnosis and treatment of hepatolithiasis
3.1 Research objects. The data of 64 multi-detector helical CT of total 60 patients with hepatolithiasis were collected from Oct.2008 to Oct.2010. All the patients'data were analyzed with 3D reconstruction and pathological classification, and subjected to surgery simulation to determine the surgical procedure and resection sites. After then the real surgical operations were performed and compare the real situation with the results from previous 3D reconstruction and surgery simulation.
3.2 Instruments and equipment:same as 1.1
3.3 Methods for 3D reconstruction of biliary tract system:same as 1.1,1.2,1.4.
3.4 Post-processing after imaging
The reconstructed STL models were imported into FreeForm Modeling System for smoothing and denoising to get the smooth, lifelike 3D models. For the convenience of 3D observation, we use MIPS to render the liver, hepatic veins, hepatic artery, portal veins, biliary tract, calculus, abdominal vessels and around visceral organs with different colors. At the same time, the 3D anatomic relationships were observed.
3.5 virtual operation of hepatolithiasis
The 3D model was magnified, minified, rotated, and transparentized to observe every parts of organs and lesions in all directions with the FreeForm Modeling System, so that the distribution of calculi, the states of intra- and extra-hepatic bile ducts and liver would be explicit, based on which a proper operation plan can be made. Additionally, virtual reality surgery could be applied to hepatolithiasis to choose an optimal operation method based on PHANTOM, an instrument with force feedback, and self-developed surgical instruments.
3.6 Actual operation of hepatolithiasis
Based on the observation of the video of 3D models and printscreens in any directions, the size, the amount and position of hepatolithiasis will be explicit; the stenosis and distention of hepatic duct, the length of the narrow bile ducts, and hepatatrophia and megalohepatia could be observed simultaneously; so as to make preoperative prognosis, pathological classification and operation plans. When brought to operating room, the video of 3D models and printscreens compared to the actual operation, can offer the real-time direction of operations. The common surgical methods are listed as follows:bile duct exploration (BDE), hepatectomy (HT), left lateral lobectomy (LLL), right posterior lobectomy (RPL), left hepatectomy(LH), hepaticojejunstomy(HJS), intrahepatic bile duct support tube placement (IBDSTP), and so on. The actual operation is the combination of different surgical methods mentioned above. After actual operation, on the basement of the comparison between 3D models and the true structure of organs in operation, and that between the virtual reality operation and the process of the real one,then the clinical value of 3D technique in the accurate treatment of hepatolithiasis is evaluated.
Results
1. Digitized biliary tract anatomy study based on 64 multi-detector helical CT scanning data
1.13D reconstruction of biliary tract by 64 multi-detector helical CT
The 3D reconstruction of biliary tract was successfully performed in 54 out of 60 patients with cholelithes. The 3D reconstruction of intrahepatic ducts was unsatisfactory in 6 patients due to unobvious distention of biliary tract system. Diverse shapes of calculi in intrahepatic ducts were observed in the cross section of plain CT scanning and appeared as mottled, quasi-circular, striped or beadlike high density images. The position of calculi and nearby bile ducts showed various degrees of stenosis and distention, combination with atrophy of Liver lesion or compensatory hypertrophy of surrounding liver tissue. Extrahepatic cholelithes were marked by rounded or annular high-density shadow surrounded by low density gall images and appeared as "target sign" or "meniscus sign".
Through 3D reconstruction of biliary tract, single, multiple calculi or disseminated distribution of high density intrahepatic calculi in different subsections of liver could be observed, complicating with various distention of intrahepatic ducts. The bile ducts at the calculus region were narrowed,and the distal bile ducts branches dilated like "biliary tree". Sometimes the dilated bile ducts became a "Gall Lake" or regional hepatapostema with different density. MinIP could show not only high density calculi and dilated intra- and extra-hepatic bile ducts, but also present radioparent calculus with soft tissue density in the choledoch. The dilated bile ducts could branch up to three to four levels and mostly appeared as a "dry stick". Commonly, the distension of bile ducts near porta was more serious than that of bile ducts far from porta, while distension or stenosis of intrahepatic duct was regional and the branch was sparse and slim. Besides the shape and volume of high density calculi in the intrahepatic ducts, the lesion liver prenchyma atrophied and hyperplastic complex was also observed. As far as complicated choledocholithiasis was concerned, single or multiple beadlike high density calculi images with various shapes and volume could be seen in the choledoch. The distension of bile ducts at upside obstruction was obvious, while distal bile ducts did not show or suddenly obstructed. Oledocholithiasis was marked by distension of all bile ducts above calculus as well as cholecyst; intrahepatic ducts dilated to different extent and appeared as a "dry stick". The choledoch under calculus was gradually narrowed, while dilated in full range in complicated pyogenic cholangitis.
1.23D reconstruction of biliary tracts by MI-3DVS
1.2.1 3D reconstruction of live, intra- and extra-hepatic vascular system
The liver model can truly reflect the actual volume and anatomy marker of liver, and simultaneously present liver and branches of intrahepatic artery, vein and portal vein through adjusting the transparency of liver. The model presented vivid shapes and stereoscopic identification of various structures, including abdominal aorta, celiac arteries and branches, such as gastroduodenal artery, right and left renal artery, right and left gastric artery, splenic artery, part of mesenteric artery, proper hepatic artery, right and left hepatic artery as well as intrahepatic branches, and so on. The model of the portal system presented portal trunk, splenic vein and superior mesenteric vein out of liver, and the intrahepatic portal system could clearly present right and left main portal vein as well as branches in various hepatic lobes and subsections in the liver. The model of the hepatic vein system could clearly present the intrahepatic distribution of three hepatic veins and their branches, the confluence of three hepatic veins, suprahepatic and inferior vena cava as well as the spatial anatomy relationship of all hepatic veins.
1.2.2 3D reconstruction of biliary tract system and calculus
Aided by plain CT scanning images and portal phase scanning images, MI-3DVS medical images segmentation with adaptive region growing algorithm usually could segment biliary tract system completely in single-time. For some fine structures such as the region where cystic duct and choledoch converge, and the hilar bile duct without expansion, scanning data of different blood vessel phases including plain scanning phase, arterial phase, venous phase and portal phase could be used for image segmentation of biliary tract system of various phases. The specific anatomy information in various phases could be integrated into an complete 3D image of biliary tract by system automatic registration function. The reconstructed biliary tract system mode could truly reflect the spatial ubiety relationship among calculi, biliary tract system and intra- and extra-hepatic blood vessels. After the 3D biliary tract system model was completed with celiac viscera and blood vessel model, the stereoscopic relationship of the whole epigastric viscera, the hepatic vascular system and the biliary tract system was completely presented.
After segmentation and reconstruction by MI-3DVS and through accurate segmentation of biliary tract system and calculus, the 3D reconstruction images of livers as well as intrahepatic duct system of 54 patients in this study were presented with vivid shapes and stereoscopic effects. These images clearly presented the 3D shape of patient liver and indicated the existence of megalohepatia or hepatatrophia, the 3D shape of three levels of intrahepatic bile duct tree, as well as the length and diameter of stenotic or dilated bile ducts. In part of patients with hepatolithiasis combined with extensive intrahepatic duct distension, the forth level of bile ducts, e.g., bile ducts in hepatic subsegments, could also be shown. Besides, the volume, number, in intra- and extra-hepatic 3D distribution of calculi as well as the spatial structure of bile ducts and intrahepatic vascular system could be clearly presented too.
2. Study on pathology classification of hepatolithiasis by 3D visualization technique
Typing diagnosis of hepatolithiasis was carried out by combining multiple factors including the location of calculi orlesion bile ducts (location, L), the degree of biliary stenosis (stenosis, S), the degree of biliary distention (distention, D), the atrophy of lesion liver subsection (atrophy, A), the portal hypertension (PHT), and so on. For example:
Hepatolithiasis (LⅡ,Ⅲ, SⅢ1, DⅢ2, AⅡ,Ⅲ):indicating lithiasis in theⅡandⅢsubsection of liver, combined with slight stenosis and distention of intrahepatic duct in theⅢsubsection as well as the atrophy of theⅡandⅢsubsections of liver.
Hepatolithiasis (LⅥ,Ⅶ,SⅥ1, DⅥ0, AⅥ):indicating lithiasis in theⅥandⅦsubsection of liver, complicating with slight stenosis ofⅥsubsection of bile duct without distention of remote bile duct, as well as atrophy of theⅥsubsections of liver.
The liver subsections of patients with hepatolithiasis based on the technique of 3D visualization complicating wit the individual characters of liver anatomy. It brings out accurate position diagnosis for the distribution of calculi and lesion regions of bile ducts which combined with the degree and range of biliary stenosis or distention and having or not biliary cirrhosis, hepatatrophia as well as megalohepatia, could draw out relatively reasonable clinicopathologic analysis could be drawn out. This new analysis method considering multi-factors including distribution of calculi, regions of biliary stenosis or distention, liver volume variation and cirrhosis and so on, thus could make out more accurate position diagnosis for hepatolithiasis and the lesion region of bile duct, and is more practically significant in clinic guidance for determining the excision range and the surgical processing modes such as bile tract drainage.
3. Application of 3D reconstruction technique in diagnosis and therapy of hepatolithiasis
3.1 Definite 3D shape of intrahepatic "biliary tree" and "blood vessel tree" by 3D reconstruction
The 3D shape of liver model of patients in this study has vivid shape and correct anatomy structure marker. After Transparency processing on liver, through single or various combination of video mode, it will be clearly observed the existence of hepatatrophia,3D distribution and shape of intrahepatic ducts and blood vessels, volume and distribution range of calculi in intrahepatic ducts, the degree of biliary stenosis and distention, having or not combination with extrahepatic cholangiolithiasis. By virtue of the 3D model, not only the distribution of intrahepatic calculi, the pathological change of biliary system and liver parenchyma as well as the relationship between disease focus and surrounding tissues could be observed, but also the blood supply patterns of liver and variation types of hepatic arteries could be determined, thus the typing diagnosis and virtual operation of hepatolithiasis could be precisely carried out.
3.2 Simulated operation of hepatolithiasis
In the simulated environment of operation, the tissues and organs could be observed in any angle through magnifying, minifying, rotating and hyalinizing the 3D model. The distribution of calculi, the pathologic lesion of biliary tract system and liver as well as changes in intrahepatic blood vessel tree could be determined. Based on these observations, clinic classification of hepatolithiasis and choice of optimal operation mode were carried out. The operation process mentioned above was simulated in turn to observe the surgical outcome. In this study, simulated operations were carried out on 30 cases of complex hepatolithiasis, while other 30 patients with positively diagnostic type I hepatolithiasis or hepatolithiasis without stenosis of bile duct did not performed simulated operations. After observing the dissection relationship of important blood vessels and bile ducts in the excision plane of 30 patients by simulated operation, the optimal operation plan were formulated to direct the actual operation process.
3.3 Clinic application of 3D visualization technique of biliary tract
Based on the observation and analysis of 3D video and sectional drawing as well as drilling by simulated operation on 30 patients with complex hepatolithiasis, operation plans were made on all patients. The actual plan included 30 cases of BDE+LLL+HJS,8 cases of BDE+LLL+IBDSTP,3 cases of LLL+RPL+HJS,3 cases of LH+HJS, and 10 cases of LLL+ HJS+1BDSTP. The reconstruction model of liver complied with the actual operation condition in 54 patients well. The accordance rate between simulated operation plans and actual operation was 90% (27/30. The radical surgery was difficult to be carried out on 3 emergency patients). No serious complications were observed post-operation, and no case of death. Postoperative radiography of bile duct on 51 patients presents no residual of calculus. Lithiasis relapsed in 3 patients after half a year, and the recurrence rate in half a year was 5.58%(3/51).
Conclusion
1. Technique of 3D reconstruction can realize the digitized anatomy of biliary tract.
Both the CT 3D reconstruction by 64 multi-detector helical CT MxliteView workstation or by MI-3DVS, the 3D reconstruction can achieve the digitized anatomy of biliary tracts. Compared with 2D CT imaging, the digitized dissection can show the form of the bile duct tree of the biliary tract system and the distribution of calculi in three-dimension. However, the 3D images reconstructed by the CT MxliteView workstation were only the biliary tract images in a certain blood vessel phase. which only showed the 2D sagittal or coronal plain films of the 3D images in vision to surgeons,but not the 3D images in the true sense. It is difficult for these 2D plain films to present the 3D spatial relationship among blood vessels, including hepatic artery, portal vein, hepatic vein, and the biliary tree simultaneously.Thus it is difficult to observe the spatial dissection of bile ducts, calculi and the intrahepatic vascular system in various angles or directions. However, MI-3DVS could render the liver surrounding viscus, celiac blood vessels, and various intrahepatic duct systems with different colors. Then through the visualization technique the complete 3D images of the whole epigastric viscus are obtained. In these 3D images, the liver, hepatic artery, hepatic vein, portal vein, celiac blood vessels as well as surrounding viscus could be presented independently, or simultaneously, or in various combinations. The pattern of intrahepatic duct tree, distribution of calculi, as well as the location and degree of biliary stricture could be clearly presented through partially magnifying or rotating this 3D image. Compared with the traditional black-and-white 2D image, the 3D image presents the critical dissection structures including the region of lesion as well as surrounding blood vessels and viscus with higher definition and visualization, and realizes the 3D presentation in the true sense and thus will contribute to accurate preoperative diagnosis for hepatolithiasis.
2.3D visualization can realize precise classification diagnosis of hepatolithiasis.
The liver subsection of patients with hepatolithiasis based on the technique of 3D visualization complies with the individual characters of liver anatomy. It brings out accurate position diagnosis for the distribution of calculi and lesion regions of bile duct. Combining the degree and range of biliary stenosis or distention and having or not biliary cirrhosis, hepatatrophia as well as megalohepatia, relatively reasonable clinicopathologic analysis could be drawn out. This new analysis method considers multi-factors including distribution of calculi, regions of biliary stenosis or distention, liver volume variation and cirrhosis and so on, thus could make out more accurate position diagnosis for hepatolithiasis and the lesion region of bile duct, and is more practically significant in clinic guidance for determining the excision range and the surgical process such as bile tract drainage.
3.3D visualization technique plays an important guide in the treatment of hepatolithiasis.
3D visualization of biliary tract and virtual operation could realize accurate preoperative diagnosis of hepatolithiasis and simulated operation; formulate the optimum operation plan through comparing various plans and surgical pathways or methods of critical steps, thus implementing accurate surgical operation. For these reasons, the 3D visualization technique has an important reference value in reasonable choice of operation mode for complex hepatolithiasis.
引文
[1]姚向庆,陈燕凌.原发性肝内胆管结石的治疗进展[J].医学综述,2007,13(8)：583-584
[2]黄志强.当今胆道外科的发展与方向[J].中华外科杂志,2006,44(23)：1585-1586.
[3]陈水平,余新,邵江华,等.原发性胆管结石手术引流方式的分析及其再认识[J].世界华人消化杂志,2009,17(19)：2005-2009
[4]黄志强主编,胆道外科手术学(第二版),人民卫生出版社。2010,129-141
[5]秦明放,赵宏志,王庆,等.微创治疗肝外胆管结石阶梯性方案研究[J].88-90
[6]范应方,方驰华,朱新勇,等.64层螺旋CT胆道三维成像对肝胆管结石病的诊断价值[J].中华消化外科杂志,2007(6)6：428-432.
[7]中华医学会外科学分会胆道外科学组.肝胆管结石病诊断治疗指南[J].中华消化外科杂志,2007,6(2)：156-160
[8]Xu Qingzhen, Bao Susu, Peng Fengping, et al. Research on modified ivewire interaction segmentation algorithm [J]. Journal of Computational Information Systemsv,2007,3(1):117-123
[9]廖其光,鲍苏苏,潘家辉,等.基于VTK肝脏三维模型可视化研究与实现[J].计算机与数字工程,2008,220(2)：102-104.
[10]ZM Zhou, CH Fang, LW Huang, et al. Three-dimensional Reconstruction of Pancreas Based on the Virtual Chinese Human-Female No.1[J]. Postgraduate Medical Journal,2006,82; 392-396.
[11]周五一,方驰华,黄立伟,等.肝脏管道灌注后数字化虚拟肝脏及其手术[J].第四军医大学学报,2006,27(8)：712-716.
[12]方驰华,周五一,虞春堂,等.肝脏管道系统灌注后薄层CT扫描和三维重建的研究[J].中华外科杂志,2004,42(9)：562-565.
[13]朱新勇,方驰华,鲍苏苏,等.基于64排螺旋CT扫描数据的肝脏图像分割和三维重建[J].南方医科大学学报,2008,28(3)：345-347.
[14]方驰华,杨剑,范应方.肝脏仿真手术的研究[J].中华外科杂志,2007(45)11：753-755.
[15]范应方,方驰华,项楠,杨剑,陈智翔,苏仲和.三维可视化技术在中段胰腺切除术的临床应用[J].中华实验外科杂志,2010,27(9)：1338-1340.
[16]FANG Chi-hua, LI Xiao-Feng, LI Zhou, FAN Ying-Fang, LU Chao-Min, HUANG Yan-Peng and PENG Feng-Ping. Application of medical image processing system in liver transplantation[J]. Hepatobiliary Pancreat Dis Int. 2010,9(4):346-351.
[17]FANG Chi-hua, HUANG Yan-peng, CHEN Mian-ling, LU Chao-min, LI Xiao-feng and QIU Wen-feng. Digital medical technology based on 64-slice computed tomography in hepatic surgery[J]. Chinese medical Journal. 2010,123(9)1149-1153.
[1]冉瑞图.发展中国特色的胆道外科[J].中华肝胆外科杂志,2000,6(3)：163-165
[2]赵振美,树伟,林祥涛,等.国人肝段的再认识[J].解剖学杂志,2008,31(3)：400-403.
[3]钟世镇.现代临床解剖学的回顾与展望[J].解剖学报,1999,30(2)：97-99.
[4]范应方,方驰华,朱新勇,等.64层螺旋CT胆道三维成像对肝胆管结石病的诊断价值[J].中华消化外科杂志,2007,6(6)：428-432.
[5]方驰华,周五一,钟世镇.虚拟人研究现状及展望[J].中华外科杂志,2004,42(15)：953-955.
[6]方驰华,周五一,黄立伟,等.虚拟中国人女性一号肝脏图像三维重建和虚拟手术的切割[J].中华外科杂志,2005,43(11)：748-752.
[7]方驰华,周五一,虞春堂,等.肝脏管道系统灌注后薄层CT扫描和三维重建的研究[J].中华外科杂志,2004,42(9)：562-565.
[8]方驰华,钟世镇,原林,等.数字化虚拟肝脏图像三维重建的初步研究[J].中华外科杂志,2004,42(2)：94-96.
[9]方驰华,范应方,朱新勇,等.64层螺旋CT胰胆管和腹腔血管三维成像在胆胰疾病中的应用[J].中华消化外科杂志,2008,7(1)：31-34
[10]方驰华,陈智翔,范应方,等.腹部医学图像三维可视化系统辅助肝胆管结石诊治决策的价值[J].中国实用外科杂志,2010,30(01)：40-43
[11]方驰华,黄燕鹏,鲁朝敏,等.数字医学技术在肝胆管结石诊治中的应用价值[J].中华外科杂志,2009 47(12)：909-911
[1]张学文,杨永生,张丹.肝内胆管结石分型及治疗方法选择[J].中国实用外科杂志,2009,29(9)：790-792.
[2]Ti TK,Wong CWW,Yuen RY, et al. The chemical composition of gallstones:its relevance to surgeons in southeast Asia[J].Ann Acad Med Singapore, 1996,25(2):255-258.
[3]Cetta F. The classification of biliary calculi and the clicico-ther-apeutic implications[J]. Ann Ital Chir,1998,69(6):701-708.
[4]Nakayama F. Hepatolithiasis:Classification of hepatolithiasis and its statistical distribution in Japan[J]. Nippon Rinsho,1987,5(7):560-563.
[5]何小东,贵勇,张振寰,等.肝内胆管结石和狭窄的外科治疗[J].肝胆外科杂志,2000,8(5)：327-328
[6]中华医学会外科学分会胆道外科学组.肝内胆管结石病诊断治疗指南[J].中华消化外科杂志,2007,6(2)：156-161.
[7]方驰华,常旭,鲁朝敏,等.肝内外胆管结石64排CT数据三维重建及其临床意义[J].南方医科大学学报,2008,28(3)：370-372.
[8]方驰华,黄燕鹏.数字医学技术在肝胆外科中的临床应用现况[J].岭南外科杂志,2008,8(5)：333-335
[9]沈柏用,施源.肝脏分段解剖的新认识[J].世界华人消化杂志,2008,16(9)：913-918
[1]黄志强.胆道外科手术学(第二版),人民卫生出版社。2010,129-141.
[2]范应方,方驰华,朱新勇,等.64层螺旋CT胆道三维成像对肝胆管结石病的诊断价值[J].中华消化外科杂志,2007,6(6)：428-432.
[3]方驰华,钟世镇,吴坤成,等.适用于CT薄层扫描和三维重建肝脏管道系统的灌注和铸型的建模研究[J].第四军医大学学报,2003,24(22)：2076-80.
[4]方驰华,周五一,虞春堂,等.肝脏管道系统灌注后薄层CT扫描和三维重建的研究[J].中华外科杂志,2004,42(9)：562-565.
[5]张学文,杨永生,张丹.肝内胆管结石分型及治疗方法选择[J].中国实用外科杂 志,2009,29(9)：790-792.
[6]中华医学会外科学分会胆道外科学组.肝胆管结石病诊断治疗指南[J].中华消化外科杂志,2007,6(2)：156-160.
[7]黄志强.当今胆道外科的发展与方向[J].中华外科杂志,2006,44(23)：1585-1586.
[8]周宁新,张效东.肝内胆管结石外科治疗的方法与选择[J].肝胆外科杂志,2007,15(6)：401-403.
[9]陈燕凌,朱金海,韩圣华,等.解剖式肝段切除治疗左肝胆管结石-保肝手术的探讨[J].中国现代普通外科进展,2009,12(2)：93-96.
[10]董家鸿,杨世忠.精准肝切除的技术特征与临床应用[J].中国实用外科杂志,2010,30(8)：638-640.
[11]H.C.Kim, S.H.Park, S.I.Park, et al. Three-dimensional reconstructed images using multidector computed tomography in evaluation of the biliary tract:an illustrative review[J]. Abdominal Imaging,2004,29:472-478.
[12]方驰华,常旭,鲁朝敏,等.肝内外胆管结石64排CT数据三维重建及其临床意义[J].南方医科大学学报,2008,28(3)：370-372.
[13]方驰华,黄燕鹏.数字医学技术在肝胆外科中的临床应用现况[J].岭南外科杂志,2008,8(5)：333-335.
[1]黄志强.色素性胆石形成的研究[J].中华外科杂志,1984,22(5)：261-262.
[2]黄家驷编著.黄家驷外科学第六版[M].人民卫生出版社,2000,1283-1286.
[3]段炼,李宜雄.肝胆管结石发病机制的研究进展[J].国外医学外科学分册,2005,32(4)：255-259.
[4]韩天权,张圣道.胆石病流行病学研究的现状和发展[J].胃肠病学,2003,8(3)：166-167.
[5]张学文,杨永生,张丹.肝内胆管结石分型及治疗方法选择[J].中国实用外科杂志,2009,29(9)：790-792.
[6]中华医学会外科学分会胆道外科学组.肝胆管结石病诊断治疗指南[J].中华消化外科杂志,2007,6(2)：156-160
[7]Cetta F. The classification of biliary calculi and the clicico-ther-apeutic implications[J]. Ann Ital Chir,1998,69(6):701-708.
[8]Nakayama F. Hepatolithiasis:Classification of hepatolithiasis and its statistical distribution in Japan[J]. Nippon Rinsho,1987,5(7):560-563.
[9]何小东,贵勇,张振寰,等.肝内胆管结石和狭窄的外科治疗[J].肝胆外科杂志,2000,8(5)：327-328
[10]范应方,方驰华,朱新勇,等.64层螺旋CT胆道三维成像对肝胆管结石病的诊断价值[J].中华消化外科杂志,2007,6(6)：428-432.
[11]巨邦律,方驰华,钟洪才,等.磁共振胆胰管成像在肝内胆管结石外科治疗中的价值[J].世界华人消化杂志,2005,13(5)：690-691.
[12]梁力建,李绍强.复杂肝胆管结石诊断和治疗原则[J].中国实用外科杂志,2009,29(7)：542-545.
[13]梁力建,李绍强,彭宝岗,等.肝切除治疗肝内胆管结石[J].中华肝胆外科杂志,2006,12(1)：796-800.
[14]邹声泉,郭伟,秦仁义,等.肝内胆管结石外科手术治疗疗效的分析[J].中华外科杂志,2003,41(7)：509-511.
[15]M.Lesurtel, J.M.Regimbeau, O.Farges, et al. Intrahepatic cholangio carcinoma and hepatolithiasis:an unusual associationin Western countries [J]. European Journal of Gastroenterology&Hepatology,2002,14(9):1025-1027.
[16]周宁新,张效东.肝内胆管结石外科治疗的方法与选择[J].肝胆外科杂志,2007,15(6)：401-403.
[17]梁力建,李绍强.关于胆肠吻合术一些问题的思考[J].中国实用外科杂志,2004,24(1)：41-42.
[18]孟翔凌,杨文奇,徐阿曼,等.胆肠内引流在肝内胆管结石治疗中的应用[J].腹部外科,2005,18(3)：169-170.
[19]黄志强.微创外科与外科微创化的总体研究[J].中国微创科杂志,2004,4(1)：1-2.
[20]Li ZH, Chen M, Liu JK, et al. Endoscopic sphincterotomy in the treatment of cholangiopancreatic diseases [J]. World JGastroenterol,2005,11(17):2678
[21]江文枢,卢榜裕.腹腔镜肝切除治疗肝内胆管结石[J].中国内镜杂志,2006,12(1)：22-23.
[22]孙学军,卢乐,石景森.肝胆管结石的微创外科治疗进展[J].中国普外基础与临床杂志,2006,13(4)：403-405.
[23]石景森,禄韶英.胆道外科微创技术的现状与展望[J].中国微创外科杂志,2006,6(1)：8-10.
[24]Shoda J, Oda K, Suzuki H, et al. Etiological significance of defects in cholesterol, phospholipid,and bile acid metabolism inthe liver of patients with intrahepatic calculi[J]. Hepatology,2001,33(5):1194-1205.
[25]石景森,王炳煌.胆道外科基础与临床[M].北京：人民卫生出版社,2003：467-481.
[26]马玉刚,武林枫,代文杰,等.胆道结石术后复发与残留的成因及预防[J].肝胆胰外科杂志,2008,20(2)：147-149.
[27]Zhong SZ, Yuan L, Tang L, et al. Research report of experimental database establishment of digitized virtual Chinese No.1 female [J]. DiYi JunYi DaXue XueBao,2003,23(3):196-209.
[28]张绍祥,刘正津,谭立文.首例中国数字化可视人体完成[J].第三军医大学学报,2002,24(10)：1231-1232.
[29]Kohler U, Marshall I, Robertson MB, et al. MRI measurement of wall shear stress vectors in bifurcation models and comparison with CFD predictions [J]. J Magn Reson Imaging,2001,14(5):563-573.
[30]'Flynn PM, Roche ET, Pandit AS. Generating an ex vivo sascular model [J]. ASAIO Journal.2005,51(4):426-433.
[31]范毓东,董家鸿.虚拟手术系统在肝脏手术方案设计中的应用[J].外科理论与实践,2004,9(4)：272-274
[32]方驰华,周五一,黄立伟,等.虚拟中国人女性一号肝脏图像三维重建和虚拟手术的切割[J].中华外科杂志,2005,43(11)：748-752.
[33]Andreas P,Karl H,Bernhard P, et al. Creating a high resolution spatial model of the inner organs based on the Visible Human [J]. Med Image Anal, 2001,5:220-221.
[34]Soler L,Delingette H,Malandain G,et al. An automatic virtual patient reconst ruction from CT-scans for hepaticsurgical planning [J]. Stud Health Technol Inform,2000,70:315-316.
[35]张绍祥.中国数字化可视人体研究进展[J].中国科学基金,2003,17(1)：4-6.
[36]方驰华,周五一,黄立伟,等.虚拟中国人女性一号肝脏图像三维重建和虚拟手术的切割[J].中华外科杂志,2005,43(11)：748-752.
[37]周泽民,方驰华,黄立伟,等.基于数字化虚拟中国人女性一号胰腺图像的三维重建及可视化研究[J].中华外科杂志,2005,43(21)：1401-1404.
[38]金立亭,冯贤松.三维超声成像在肝胆疾病诊断中的应用[J].腹部外科,2007,20(1)：55-56.
[39]方驰华,周五一,黄立伟,等.虚拟中国人女性一号肝脏图像三维重建和虚拟手术的切割[J].中华外科杂志,2005,43(11)：748-752.
[40]方驰华,周五一,虞春堂,等.肝脏管道系统灌注后薄层CT扫描和三维重建的研究[J].中华外科杂志,2004,42(9)：562-565.
[41]彭丰平,鲍苏苏,曾碧卿.基于64排CT数据的肝胆虚拟手术系统的研究[J].系统仿真学报,2009,21(17)：5458-5461.
[42]方驰华,唐云强,鲁朝敏,等.胆囊切开取石和胆囊切除术的可视化研究[J].南方医科大学学报,2008,28(3)：356-359.
[43]方驰华,刘宇斌,唐云强,等.胰体尾肿瘤的可视化仿真手术研究[J].南方医科大学学报,2008,28(6)：926-929.
[44]方驰华,常旭,鲁朝敏,等.肝内外胆管结石64排CT数据三维重建及其临床意义[J].南方医科大学学报,2008,28(3)：370-372.
[45]方驰华,陈智翔,范应方,等.腹部医学图像三维可视化系统辅助肝胆管结石诊治决策的价值[J].中国实用外科杂志,2010,30(1)：40-43.
[46]方驰华,黄燕鹏.数字医学技术在肝内胆管结石外科治疗中的应用[J].外科理论与实践,2009,14(2)：156-158.
[47]方驰华,项楠.数字化微创技术在肝胆胰外科的应用[J].中国微创外科杂志,2011,11(1)：15-19.
[2]黄志强.当今胆道外科的发展与方向[J].中华外科杂志,2006,44(23)：1585-1586.
[3]陈水平,余新,邵江华,等.原发性胆管结石手术引流方式的分析及其再认识[J].世界华人消化杂志,2009,17(19)：2005-2009
[4]黄志强主编,胆道外科手术学(第二版),人民卫生出版社。2010,129-141
[5]秦明放,赵宏志,王庆,等.微创治疗肝外胆管结石阶梯性方案研究[J].88-90
[6]范应方,方驰华,朱新勇,等.64层螺旋CT胆道三维成像对肝胆管结石病的诊断价值[J].中华消化外科杂志,2007(6)6：428-432.
[7]中华医学会外科学分会胆道外科学组.肝胆管结石病诊断治疗指南[J].中华消化外科杂志,2007,6(2)：156-160
[8]Xu Qingzhen, Bao Susu, Peng Fengping, et al. Research on modified ivewire interaction segmentation algorithm [J]. Journal of Computational Information Systemsv,2007,3(1):117-123
[9]廖其光,鲍苏苏,潘家辉,等.基于VTK肝脏三维模型可视化研究与实现[J].计算机与数字工程,2008,220(2)：102-104.
[10]ZM Zhou, CH Fang, LW Huang, et al. Three-dimensional Reconstruction of Pancreas Based on the Virtual Chinese Human-Female No.1[J]. Postgraduate Medical Journal,2006,82; 392-396.
[11]周五一,方驰华,黄立伟,等.肝脏管道灌注后数字化虚拟肝脏及其手术[J].第四军医大学学报,2006,27(8)：712-716.
[12]方驰华,周五一,虞春堂,等.肝脏管道系统灌注后薄层CT扫描和三维重建的研究[J].中华外科杂志,2004,42(9)：562-565.
[13]朱新勇,方驰华,鲍苏苏,等.基于64排螺旋CT扫描数据的肝脏图像分割和三维重建[J].南方医科大学学报,2008,28(3)：345-347.
[14]方驰华,杨剑,范应方.肝脏仿真手术的研究[J].中华外科杂志,2007(45)11：753-755.
[15]范应方,方驰华,项楠,杨剑,陈智翔,苏仲和.三维可视化技术在中段胰腺切除术的临床应用[J].中华实验外科杂志,2010,27(9)：1338-1340.
[16]FANG Chi-hua, LI Xiao-Feng, LI Zhou, FAN Ying-Fang, LU Chao-Min, HUANG Yan-Peng and PENG Feng-Ping. Application of medical image processing system in liver transplantation[J]. Hepatobiliary Pancreat Dis Int. 2010,9(4):346-351.
[17]FANG Chi-hua, HUANG Yan-peng, CHEN Mian-ling, LU Chao-min, LI Xiao-feng and QIU Wen-feng. Digital medical technology based on 64-slice computed tomography in hepatic surgery[J]. Chinese medical Journal. 2010,123(9)1149-1153.
[1]冉瑞图.发展中国特色的胆道外科[J].中华肝胆外科杂志,2000,6(3)：163-165
[2]赵振美,树伟,林祥涛,等.国人肝段的再认识[J].解剖学杂志,2008,31(3)：400-403.
[3]钟世镇.现代临床解剖学的回顾与展望[J].解剖学报,1999,30(2)：97-99.
[4]范应方,方驰华,朱新勇,等.64层螺旋CT胆道三维成像对肝胆管结石病的诊断价值[J].中华消化外科杂志,2007,6(6)：428-432.
[5]方驰华,周五一,钟世镇.虚拟人研究现状及展望[J].中华外科杂志,2004,42(15)：953-955.
[6]方驰华,周五一,黄立伟,等.虚拟中国人女性一号肝脏图像三维重建和虚拟手术的切割[J].中华外科杂志,2005,43(11)：748-752.
[7]方驰华,周五一,虞春堂,等.肝脏管道系统灌注后薄层CT扫描和三维重建的研究[J].中华外科杂志,2004,42(9)：562-565.
[8]方驰华,钟世镇,原林,等.数字化虚拟肝脏图像三维重建的初步研究[J].中华外科杂志,2004,42(2)：94-96.
[9]方驰华,范应方,朱新勇,等.64层螺旋CT胰胆管和腹腔血管三维成像在胆胰疾病中的应用[J].中华消化外科杂志,2008,7(1)：31-34
[10]方驰华,陈智翔,范应方,等.腹部医学图像三维可视化系统辅助肝胆管结石诊治决策的价值[J].中国实用外科杂志,2010,30(01)：40-43
[11]方驰华,黄燕鹏,鲁朝敏,等.数字医学技术在肝胆管结石诊治中的应用价值[J].中华外科杂志,2009 47(12)：909-911
[1]张学文,杨永生,张丹.肝内胆管结石分型及治疗方法选择[J].中国实用外科杂志,2009,29(9)：790-792.
[2]Ti TK,Wong CWW,Yuen RY, et al. The chemical composition of gallstones:its relevance to surgeons in southeast Asia[J].Ann Acad Med Singapore, 1996,25(2):255-258.
[3]Cetta F. The classification of biliary calculi and the clicico-ther-apeutic implications[J]. Ann Ital Chir,1998,69(6):701-708.
[4]Nakayama F. Hepatolithiasis:Classification of hepatolithiasis and its statistical distribution in Japan[J]. Nippon Rinsho,1987,5(7):560-563.
[5]何小东,贵勇,张振寰,等.肝内胆管结石和狭窄的外科治疗[J].肝胆外科杂志,2000,8(5)：327-328
[6]中华医学会外科学分会胆道外科学组.肝内胆管结石病诊断治疗指南[J].中华消化外科杂志,2007,6(2)：156-161.
[7]方驰华,常旭,鲁朝敏,等.肝内外胆管结石64排CT数据三维重建及其临床意义[J].南方医科大学学报,2008,28(3)：370-372.
[8]方驰华,黄燕鹏.数字医学技术在肝胆外科中的临床应用现况[J].岭南外科杂志,2008,8(5)：333-335
[9]沈柏用,施源.肝脏分段解剖的新认识[J].世界华人消化杂志,2008,16(9)：913-918
[1]黄志强.胆道外科手术学(第二版),人民卫生出版社。2010,129-141.
[2]范应方,方驰华,朱新勇,等.64层螺旋CT胆道三维成像对肝胆管结石病的诊断价值[J].中华消化外科杂志,2007,6(6)：428-432.
[3]方驰华,钟世镇,吴坤成,等.适用于CT薄层扫描和三维重建肝脏管道系统的灌注和铸型的建模研究[J].第四军医大学学报,2003,24(22)：2076-80.
[4]方驰华,周五一,虞春堂,等.肝脏管道系统灌注后薄层CT扫描和三维重建的研究[J].中华外科杂志,2004,42(9)：562-565.
[5]张学文,杨永生,张丹.肝内胆管结石分型及治疗方法选择[J].中国实用外科杂 志,2009,29(9)：790-792.
[6]中华医学会外科学分会胆道外科学组.肝胆管结石病诊断治疗指南[J].中华消化外科杂志,2007,6(2)：156-160.
[7]黄志强.当今胆道外科的发展与方向[J].中华外科杂志,2006,44(23)：1585-1586.
[8]周宁新,张效东.肝内胆管结石外科治疗的方法与选择[J].肝胆外科杂志,2007,15(6)：401-403.
[9]陈燕凌,朱金海,韩圣华,等.解剖式肝段切除治疗左肝胆管结石-保肝手术的探讨[J].中国现代普通外科进展,2009,12(2)：93-96.
[10]董家鸿,杨世忠.精准肝切除的技术特征与临床应用[J].中国实用外科杂志,2010,30(8)：638-640.
[11]H.C.Kim, S.H.Park, S.I.Park, et al. Three-dimensional reconstructed images using multidector computed tomography in evaluation of the biliary tract:an illustrative review[J]. Abdominal Imaging,2004,29:472-478.
[12]方驰华,常旭,鲁朝敏,等.肝内外胆管结石64排CT数据三维重建及其临床意义[J].南方医科大学学报,2008,28(3)：370-372.
[13]方驰华,黄燕鹏.数字医学技术在肝胆外科中的临床应用现况[J].岭南外科杂志,2008,8(5)：333-335.
[1]黄志强.色素性胆石形成的研究[J].中华外科杂志,1984,22(5)：261-262.
[2]黄家驷编著.黄家驷外科学第六版[M].人民卫生出版社,2000,1283-1286.
[3]段炼,李宜雄.肝胆管结石发病机制的研究进展[J].国外医学外科学分册,2005,32(4)：255-259.
[4]韩天权,张圣道.胆石病流行病学研究的现状和发展[J].胃肠病学,2003,8(3)：166-167.
[5]张学文,杨永生,张丹.肝内胆管结石分型及治疗方法选择[J].中国实用外科杂志,2009,29(9)：790-792.
[6]中华医学会外科学分会胆道外科学组.肝胆管结石病诊断治疗指南[J].中华消化外科杂志,2007,6(2)：156-160
[7]Cetta F. The classification of biliary calculi and the clicico-ther-apeutic implications[J]. Ann Ital Chir,1998,69(6):701-708.
[8]Nakayama F. Hepatolithiasis:Classification of hepatolithiasis and its statistical distribution in Japan[J]. Nippon Rinsho,1987,5(7):560-563.
[9]何小东,贵勇,张振寰,等.肝内胆管结石和狭窄的外科治疗[J].肝胆外科杂志,2000,8(5)：327-328
[10]范应方,方驰华,朱新勇,等.64层螺旋CT胆道三维成像对肝胆管结石病的诊断价值[J].中华消化外科杂志,2007,6(6)：428-432.
[11]巨邦律,方驰华,钟洪才,等.磁共振胆胰管成像在肝内胆管结石外科治疗中的价值[J].世界华人消化杂志,2005,13(5)：690-691.
[12]梁力建,李绍强.复杂肝胆管结石诊断和治疗原则[J].中国实用外科杂志,2009,29(7)：542-545.
[13]梁力建,李绍强,彭宝岗,等.肝切除治疗肝内胆管结石[J].中华肝胆外科杂志,2006,12(1)：796-800.
[14]邹声泉,郭伟,秦仁义,等.肝内胆管结石外科手术治疗疗效的分析[J].中华外科杂志,2003,41(7)：509-511.
[15]M.Lesurtel, J.M.Regimbeau, O.Farges, et al. Intrahepatic cholangio carcinoma and hepatolithiasis:an unusual associationin Western countries [J]. European Journal of Gastroenterology&Hepatology,2002,14(9):1025-1027.
[16]周宁新,张效东.肝内胆管结石外科治疗的方法与选择[J].肝胆外科杂志,2007,15(6)：401-403.
[17]梁力建,李绍强.关于胆肠吻合术一些问题的思考[J].中国实用外科杂志,2004,24(1)：41-42.
[18]孟翔凌,杨文奇,徐阿曼,等.胆肠内引流在肝内胆管结石治疗中的应用[J].腹部外科,2005,18(3)：169-170.
[19]黄志强.微创外科与外科微创化的总体研究[J].中国微创科杂志,2004,4(1)：1-2.
[20]Li ZH, Chen M, Liu JK, et al. Endoscopic sphincterotomy in the treatment of cholangiopancreatic diseases [J]. World JGastroenterol,2005,11(17):2678
[21]江文枢,卢榜裕.腹腔镜肝切除治疗肝内胆管结石[J].中国内镜杂志,2006,12(1)：22-23.
[22]孙学军,卢乐,石景森.肝胆管结石的微创外科治疗进展[J].中国普外基础与临床杂志,2006,13(4)：403-405.
[23]石景森,禄韶英.胆道外科微创技术的现状与展望[J].中国微创外科杂志,2006,6(1)：8-10.
[24]Shoda J, Oda K, Suzuki H, et al. Etiological significance of defects in cholesterol, phospholipid,and bile acid metabolism inthe liver of patients with intrahepatic calculi[J]. Hepatology,2001,33(5):1194-1205.
[25]石景森,王炳煌.胆道外科基础与临床[M].北京：人民卫生出版社,2003：467-481.
[26]马玉刚,武林枫,代文杰,等.胆道结石术后复发与残留的成因及预防[J].肝胆胰外科杂志,2008,20(2)：147-149.
[27]Zhong SZ, Yuan L, Tang L, et al. Research report of experimental database establishment of digitized virtual Chinese No.1 female [J]. DiYi JunYi DaXue XueBao,2003,23(3):196-209.
[28]张绍祥,刘正津,谭立文.首例中国数字化可视人体完成[J].第三军医大学学报,2002,24(10)：1231-1232.
[29]Kohler U, Marshall I, Robertson MB, et al. MRI measurement of wall shear stress vectors in bifurcation models and comparison with CFD predictions [J]. J Magn Reson Imaging,2001,14(5):563-573.
[30]'Flynn PM, Roche ET, Pandit AS. Generating an ex vivo sascular model [J]. ASAIO Journal.2005,51(4):426-433.
[31]范毓东,董家鸿.虚拟手术系统在肝脏手术方案设计中的应用[J].外科理论与实践,2004,9(4)：272-274
[32]方驰华,周五一,黄立伟,等.虚拟中国人女性一号肝脏图像三维重建和虚拟手术的切割[J].中华外科杂志,2005,43(11)：748-752.
[33]Andreas P,Karl H,Bernhard P, et al. Creating a high resolution spatial model of the inner organs based on the Visible Human [J]. Med Image Anal, 2001,5:220-221.
[34]Soler L,Delingette H,Malandain G,et al. An automatic virtual patient reconst ruction from CT-scans for hepaticsurgical planning [J]. Stud Health Technol Inform,2000,70:315-316.
[35]张绍祥.中国数字化可视人体研究进展[J].中国科学基金,2003,17(1)：4-6.
[36]方驰华,周五一,黄立伟,等.虚拟中国人女性一号肝脏图像三维重建和虚拟手术的切割[J].中华外科杂志,2005,43(11)：748-752.
[37]周泽民,方驰华,黄立伟,等.基于数字化虚拟中国人女性一号胰腺图像的三维重建及可视化研究[J].中华外科杂志,2005,43(21)：1401-1404.
[38]金立亭,冯贤松.三维超声成像在肝胆疾病诊断中的应用[J].腹部外科,2007,20(1)：55-56.
[39]方驰华,周五一,黄立伟,等.虚拟中国人女性一号肝脏图像三维重建和虚拟手术的切割[J].中华外科杂志,2005,43(11)：748-752.
[40]方驰华,周五一,虞春堂,等.肝脏管道系统灌注后薄层CT扫描和三维重建的研究[J].中华外科杂志,2004,42(9)：562-565.
[41]彭丰平,鲍苏苏,曾碧卿.基于64排CT数据的肝胆虚拟手术系统的研究[J].系统仿真学报,2009,21(17)：5458-5461.
[42]方驰华,唐云强,鲁朝敏,等.胆囊切开取石和胆囊切除术的可视化研究[J].南方医科大学学报,2008,28(3)：356-359.
[43]方驰华,刘宇斌,唐云强,等.胰体尾肿瘤的可视化仿真手术研究[J].南方医科大学学报,2008,28(6)：926-929.
[44]方驰华,常旭,鲁朝敏,等.肝内外胆管结石64排CT数据三维重建及其临床意义[J].南方医科大学学报,2008,28(3)：370-372.
[45]方驰华,陈智翔,范应方,等.腹部医学图像三维可视化系统辅助肝胆管结石诊治决策的价值[J].中国实用外科杂志,2010,30(1)：40-43.
[46]方驰华,黄燕鹏.数字医学技术在肝内胆管结石外科治疗中的应用[J].外科理论与实践,2009,14(2)：156-158.
[47]方驰华,项楠.数字化微创技术在肝胆胰外科的应用[J].中国微创外科杂志,2011,11(1)：15-19.