基于增强CT影像组学模型术前预测胰腺神经内分泌肿瘤病理分级

目的 观察基于CT影像组学模型术前预测胰腺神经内分泌肿瘤(PNET)病理分级(G1和G2/3级)的价值.方法 回顾性分析145例经病理证实的PNET,分为训练组91例、验证组54例,2组各自来源于同一医疗机构.基于训练组动脉期和门脉期CT图像提取PNET影像组学特征,以Pearson相关分析及ReliefF算法进行筛选...     

机器学习鉴别表现为肺纯磨玻璃结节的浸润性腺癌与非浸润性腺癌

目的 探讨基于影像组学特征构建的机器学习模型鉴别表现为肺纯磨玻璃结节的浸润性腺癌与非浸润性腺癌的可行性。方法 回顾性分析经手术病理证实的87例CT表现为纯磨玻璃结节的肺腺癌患者，其中浸润性腺癌32例，非浸润性腺癌55例（原位癌17例，微浸润性腺癌38例）。应用ITK-SNAP软件勾画ROI，A.K.软件提取影像组学特征。筛选有意义的特征参数，以Spearman相关性分析和Lasso回归分析进行特征降维。选取降维后的特征参数，分别构建支持向量机（SVM）、随机森林（RF）及逻辑回归（LR）3种机器学习模型，采用十折交叉验证法得到最优模型，绘制ROC曲线，评价3种模型的性能。结果 共提取396个影像组学特征，通过特征筛选后最终得到19个影像组学特征。SVM、RF、LR 3种机器学习模型可有效鉴别浸润性腺癌与非浸润性腺癌，准确率分别为93.30%、86.70%和83.30%，AUC分别为0.94、0.92和0.83。结论 基于影像组学特征构建的机器学习模型有较好的分类性能，可于术前有效鉴别肺浸润性腺癌与非浸润性腺癌。     

基于增强CT影像组学模型鉴别肾透明细胞癌与非透明细胞癌

目的 建立基于增强CT的影像组学模型,评估其鉴别肾透明细胞癌（ccRCC）与非透明细胞癌（non-ccRCC）的应用价值。方法 将147例ccRCC及32例non-ccRCC患者随机分为训练集125例和测试集54例。将所有患者的增强CT资料导入ITK-SNAP软件,手动勾画ROI,获得16个特征,分别建立基于特征的随机森林（RF）模型和逻辑回归（LR）模型,采用ROC曲线观察模型对ccRCC的诊断效能。结果 训练集RF模型诊断ccRCC的AUC为0.96（P<0.05）,特异度为1.00,敏感度0.83;LR模型诊断ccRCC的AUC为0.96（P<0.05）,特异度为1.00,敏感度为0.83。测试集RF模型诊断ccRCC的AUC为0.96（P<0.05）,特异度为1.00,敏感度为0.89;LR模型诊断ccRCC的AUC为0.88（P<0.05）,特异度为0.90,敏感度为0.77。结论 基于增强CT影像组学模型可用于鉴别ccRCC与non-ccRCC;RF模型诊断价值较LR模型更高。     

基于增强CT放射组学预测肝细胞肝癌病理分级

目的 观察增强CT放射组学术前预测肝细胞肝癌（HCC）病理分级的可行性及价值。方法 回顾分析429例经手术病理证实的HCC患者，分为训练组（n=329）和测试组（n=100），记录其临床特征；提取动脉期（AP）及静脉期（VP）CT图像的放射组学特征，应用最小绝对值收敛和选择算子（LASSO）回归分析法对其进行降维，筛选最有价值的组学特征后，构建基于AP、VP、AP+VP图像特征的组学模型，计算2组放射学评分并进行二分类判别。根据病理结果定义高级别和低级别HCC，采用10倍交叉验证训练选择最优组学预测模型，筛选对预测HCC病理分级有意义的临床特征后，构建临床模型以及联合组学特征和临床特征的联合模型。绘制3种模型预测训练组和测试组HCC病理分级的ROC曲线，评估其诊断能力。结果 联合组学模型最优，其判别训练组及测试组高级别和低级别HCC的放射学评分的差异均有统计学意义（Z=8.58、3.24，P均<0.05）。测试组中，联合模型预测HCC病理分级的AUC值（0.70）与组学模型（0.69）和临床模型（0.63）差异均无统计学意义（P均>0.05）。结论 基于增强CT图像的放射组学特征可用于术前预测HCC病理分级。     

基于增强CT影像组学术前预测胃腺癌淋巴结转移

目的 观察增强CT影像组学模型术前预测胃腺癌淋巴结转移（LNM）的价值。方法 回顾性分析193例经术后病理证实的单发胃腺癌的腹部双期增强CT资料,将其分为训练集（n=97）和验证集（n=96）,比较LNM （+）与LNM （－）肿瘤CT表现的差异。分别于增强动脉期和静脉期CT提取病灶影像组学特征,构建相应影像组学标签;将单因素分析有统计学意义的CT参数及其影像组学标签纳入多因素logistic回归分析,筛选胃腺癌LNM的独立预测因素,分别建立临床模型及影像组学列线图。采用受试者工作特征（ROC）曲线评估各模型预测胃腺癌LNM的效能,计算曲线下面积（AUC）,比较其差异。结果 训练集含54例LNM （+）和43例LNM （－）,验证集含58例LNM （+）和38例LNM （－）。LNM （+）患者肿瘤厚度和阳性淋巴结占比均高于LNM （－）者（P均<0.05）。肿瘤厚度及淋巴结状态均为LNM的独立预测因素（P均<0.01）并用于构建临床模型。淋巴结状态和静脉期影像组学标签是胃腺癌LNM的独立预测因素（P均<0.01）,以之构建的影像组学列线图在训练集和验证集中的AUC分别为0.810和0.778,与临床模型AUC差异均无统计学意义（0.772、0.762,Z=1.11、0.27,P=0.27、0.78）。结论 基于增强CT影像组学模型术前预测胃腺癌LNM效能较佳。     

Utility of the portal venous phase for diagnosing pancreatic necrosis in acute pancreatitis using the CT severity index

Purpose

The purpose of the study was to evaluate the value of portal venous phase (PVP) images in the diagnosis of pancreatic necrosis in patients with acute pancreatitis using computed tomography severity index (CTSI).

Methods

This retrospective study was approved by our Institutional Review Board, and written informed consent was waived. Dynamic contrast-enhanced CT images, with the pancreatic parenchymal phase (PPP) and the PVP, were obtained from 56 consecutive patients with acute pancreatitis. Two radiologists reviewed two sets of images, namely PPP images alone (image set A) and combined PPP and PVP images (image set B) to evaluate the CTSI. Cases were categorized as necrotizing pancreatitis if ensuing walled-off necrosis formation was identified 4 weeks after onset of symptoms. The relationship between pancreatic necrosis and CTSI was compared between image sets A and B. Logistic regression analysis was performed to evaluate the significance of clinical and radiological factors associated with the diagnosis of pancreatic necrosis.

Results

Pancreatic necrosis was confirmed in 14 out of 56 (25%) patients. The area under the receiver-operating-characteristic curve (AUC) for the diagnosis of pancreatic necrosis was 0.70 and 0.78 for image sets A and B, respectively. The AUC for image set B was significantly greater than that for image set A (P = 0.0002). Logistic regression analysis demonstrated that among clinical and radiological factors tested, CTSI for image set B was independently correlated with pancreatic necrosis (P = 0.025).

Conclusions

Combined PPP and PVP images significantly improved the diagnostic accuracy of pancreatic necrosis following acute pancreatitis.

     

基于增强T2*加权血管成像序列R2*图肿瘤全域纹理分析预测子宫内膜癌微卫星不稳定性

目的评估基于增强T2^*加权血管成像(ESWAN)序列R2^*图的肿瘤全域纹理分析(TA)预测子宫内膜癌(EC)微卫星不稳定性(MSI)的价值。方法回顾性分析38例经术后病理证实、术前接受ESWAN序列盆腔MR扫描的EC患者,其中12例MSI(MSI组)、26例微卫星稳定(MSS,MSS组),经后处理获得R2^*图。于R2^*图像上逐层手动勾画肿瘤ROI,融合后获得全域感兴趣容积(VOI);采用A.K.分析软件提取其纹理特征,以Spearman相关性分析和梯度提升决策树(GBDT)方法筛选最优纹理特征,构建多元logistic回归模型,用于预测EC MSI状态;以受试者工作特征(ROC)曲线评价模型的诊断效能。结果共提取74个纹理特征,最终筛选出6个最优纹理特征,以之构建预测EC MSI的回归模型。ROC曲线显示,模型的曲线下面积(AUC)、准确率、敏感度及特异度分别为0.95、89.50%、83.30%及92.30%。结论基于ESWAN序列R2^*图的肿瘤全域TA有助于术前预测EC MSI。     

基于CT影像组学术前预测胃癌淋巴血管侵犯

目的 探讨基于CT影像组学术前预测胃癌淋巴血管侵犯的价值。方法 回顾性收集经手术病理证实的181例胃癌患者,将其随机分为训练集（n=120）和验证集（n=61）。首先基于增强CT静脉期图像分割肿瘤区域并提取影像组学特征;然后利用训练集筛选与淋巴血管侵犯相关特征,构建影像组学标签;最后基于验证集验证模型,采用ROC曲线及校准曲线评估模型的预测效能及拟合度。结果 最终提取7个与胃癌淋巴管血管侵犯最相关的影像组学特征构建影像组学标签,其在训练集的ROC曲线AUC为0.742[P=0.001,95%CI（0.652,0.831）],验证集AUC为0.727[P=0.002,95%CI（0.593,0.853）]。基于训练集所得最优阈值为0.422,模型在训练集中的准确率、敏感度和特异度分别为0.708、0.586、0.806,将此阈值用于验证集,其准确率、敏感度和特异度为0.689、0.519、0.824。校准曲线显示影像组学标签在训练集及验证集均具有较好的拟合度（P均>0.05）。结论 CT影像组学可作为预测胃癌术前淋巴血管侵犯提供的全新的无创影像学方法。     

基于CT影像组学结合机器学习模型预测食管胃结合部腺癌人表皮生长因子受体2状态

目的 评估基于C T影像组学结合机器学习模型术前预测食管胃交界处腺癌(A EG)人表皮生长因子受体2(HER2)状态的价值.方法 回顾性分析101例经术后病理证实的AEG患者,按7:3比例将其分为训练集(n=70)和验证集(n=31).基于门静脉期增强CT提取AEG影像组学特征,以最小绝对值选择与收缩算子回归模型针对训...     

临床及CT影像组学特征预测胃癌微卫星高度不稳定状态

目的 观察临床和CT影像组学特征用于预测胃癌微卫星高度不稳定(MSI-H)状态的价值。方法 纳入150例胃癌患者,MSI-H阳性30例、阴性120例;按7∶3比例将其分为训练集(n=105)和验证集(n=45)。基于腹部静脉期增强CT图提取病灶影像组学特征并加以筛选,计算影像组学评分(Radscore);比较训练集和验证集MSI-H阳性与阴性患者临床资料及Radscore差异;分别基于其间差异有统计学意义的临床因素和Radscore构建临床模型、CT影像组学模型及临床-CT影像组学联合模型,评估其预测胃癌MSI-H状态的价值。结果 训练集和验证集中,MSI-H阳性与阴性肿瘤位置、Radscore差异均有统计学意义(P均<0.05)。临床模型、CT影像组学模型及联合模型评估训练集胃癌MSI-H状态的曲线下面积(AUC)分别0.760、0.799及0.864,在验证集分别为0.735、0.812及0.849;联合模型的AUC大于2种单一模型(P均<0.05)。结论 基于肿瘤位置和Radscore的临床-CT影像组学联合特征可有效预测胃癌MSI-H状态。     

Gadoxetic acid enhanced magnetic resonance imaging for prediction of the postoperative prognosis of intrahepatic mass-forming cholangiocarcinoma

Purpose

To identify imaging markers that independently predict the post-operative outcome of intrahepatic mass-forming cholangiocarcinoma (IMCC) using gadoxetate disodium-enhanced magnetic resonance imaging (MRI).

Methods

Data from 54 patients who underwent pre-operative gadoxetate disodium-enhanced MRI and curative surgery for IMCC were retrospectively evaluated. The prognostic power of various imaging and pathological features reportedly associated with recurrence-free survival (RFS) and overall survival (OS) was analyzed using Cox regression models. A model combining imaging and pathological features was developed and its performance was evaluated using the Harrell C-index and Akaike information criterion.

Results

Capsule penetration (P = 0.016) and tumor size (P = 0.015) were independent markers for worse RFS, while capsule penetration (P = 0.012) and hepatic vein obstruction (HVO, P = 0.016) were independent markers for worse OS, respectively, in the imaging-based model. Capsule penetration was the only imaging marker identified in the combined prediction model of RFS, and the combined model showed a higher C-index and lower AIC value compared with the model based on pathological features alone.

Conclusions

Capsule penetration and HVO on MRI are significantly worse imaging prognostic factors for post-operative outcomes in patients with IMCC. Incorporation of capsule penetration and HVO into a surgical staging system may improve prediction of the post-operative prognosis of IMCC.

     

基于灰阶超声的影像组学模型预测乳腺癌新辅助化疗效果

目的 探讨基于灰阶超声的影像组学模型预测乳腺癌新辅助化疗（NACT）效果的应用价值。方法 选取53例乳腺癌患者，根据NACT疗效分为临床应答与临床无应答组，比较二组临床资料及灰阶超声特征。提取基于灰阶超声的乳腺癌影像组学特征，采用Logistic回归分析建立基于上述特征的模型，采用ROC曲线评价模型预测乳腺癌NACT后临床应答的效能。结果 NACT后临床应答组32例、临床无应答组21例，2组间年龄、绝经比例、分期及分子分型差异均无统计学意义（P均>0.05），声像图所示病灶最大径、内部回声、钙化、边缘、后方回声、形态差异均无统计学意义（P均>0.05）。共6个影像学特征纳入Logistic回归模型，该模型预测乳腺癌NACT后临床应答的AUC为0.88[95% CI（0.78，0.99）]，敏感度0.88，特异度0.81。结论 基于灰阶超声的影像组学模型对评价乳腺癌NACT效果有一定价值。     

灰阶超声影像组学鉴别诊断皮下组织血管瘤与卡波西型血管内皮瘤

目的 观察灰阶超声影像组学鉴别诊断皮下组织血管瘤(HE)与卡波西型血管内皮瘤(KHE)的价值。方法 回顾性分析143例皮下组织HE和70例KHE共252处病灶,按7∶3比例将病灶随机分为训练集(n=176)和验证集(n=76);提取病灶灰阶超声影像组学特征,构建影像组学模型,结合临床资料建立联合模型,观察各模型鉴别诊断皮下组织HE与KHE的效能。结果 共选取22个系数非零的稳定特征。影像组学模型鉴别训练集皮下组织HE与KHE的曲线下面积(AUC)、准确率、敏感度、特异度、阳性预测值及阴性预测值分别为0.91[95%CI(0.89,0.93)]、91.41%、83.20%、93.92%、95.79%及89.00%;用于验证集分别为0.85[95%CI(0.83,0.87)]、90.78%、79.32%、97.90%、96.71%及88.68%。联合模型鉴别训练集皮下组织HE与KHE的AUC、准确率、敏感度、特异度、阳性预测值及阴性预测值分别为0.94[95%CI(0.92,0.96)]、94.33%、90.77%、96.38%、94.23%及94.90%;用于验证集分别为0.90[95%...     

Quantitative dynamic contrast-enhanced MR imaging for differentiating benign,borderline, and malignant ovarian tumors

Purpose

This study aimed to investigate the diagnostic performance of quantitative DCE-MRI for characterizing ovarian tumors.

Methods

We prospectively assessed the differences of quantitative DCE-MRI parameters (K^trans, k_ep, and v_e) among 15 benign, 28 borderline, and 66 malignant ovarian tumors; and between type I (n = 28) and type II (n = 29) of epithelial ovarian carcinomas (EOCs). DCE-MRI data were analyzed using whole solid tumor volume region of interest (ROI) method, and quantitative parameters were calculated based on a modified Tofts model. The non-parametric Kruskal–Wallis test, Mann–Whitney U test, Pearson’s chi-square test, intraclass correlation coefficient (ICC), variance test, and receiver operating characteristic curves (ROC) were used for statistical analysis.

Results

The largest K^trans and k_ep values were observed in ovarian malignant tumors, followed by borderline and benign tumors (all P < 0.001). K_ep was the better parameter for differentiating benign tumors from borderline and malignant tumors, with a sensitivity of 89.3% and 95.5%, a specificity of 86.7% and 100%, an accuracy of 88.4% and 96.3%, and an area under the curve (AUC) of 0.94 and 0.992, respectively, whereas K^trans was better for differentiating borderline from malignant tumors with a sensitivity of 60.7%, a specificity of 78.8%, an accuracy of 73.4%, and an AUC of 0.743. In addition, a combination with k_ep could further improve the sensitivity to 78.9%. The median K^trans and k_ep values were significantly higher in type II than in type I EOCs.

Conclusion

DCE-MRI with volume quantification is a technically feasible method, and can be used for the differentiation of ovarian tumors and for discriminating between type I and type II EOCs.

     

Noncontrast computed tomography factors predictive of extracorporeal shock wave lithotripsy outcomes in patients with pancreatic duct stones

Purpose

To assess the usefulness of factors unique to NCCT for the prediction of ESWL outcomes in patients with pancreatic duct stones.

Materials and methods

We retrospectively evaluated 148 patients with multiple PDS who had undergone ESWL therapy. All patients received an examination for NCCT both before and after ESWL. The following parameters were measured and recorded: patient characteristics including sex and age; NCCT parameters including mean stone length, mean stone volumes before and after ESWL, mean value of CT attenuation, standard deviation of CT attenuation, variation coefficient of CT attenuation, skin-to-stone distance, and pancreatic duct diameter; ESWL outcome indexes including stone clearance rate calculated using the formula \(\frac{V0 - V1}{V0} \times 100\%\), and the number of ESWL sessions. All patients were divided into groups based on their SCR: A group (SCR ≥ 90%), B group (SCR between 50% and 90%), and C group (SCR < 50%). Analysis of variance was used among the three groups to evaluate the potential predictors of SCR, and a receiver-operating curve was established to determine the optimal cutoff value.

Results

ANOVA analysis revealed that MSD was the only significant predictor for SCR (p < 0.05), and ROC indicated an optimal cutoff value of +1000.45 HU, with a sensitivity up to 78.0% and specificity of 48.6%. Stones with MSD lower than +1000.45 HU had higher SCR (69.3%) than that of higher-density ones (59.6%). Pearson correlation analysis and histogram indicated a significant positive correlation between ESWL No. and MSL (r = 0.536), MSD (r = 0.250), SDSD (r = 0.247), and PDD (r = 0.227), all values being p < 0.01.

Conclusion

MSD is the optimal predictor of ESWL efficacy, and PDS with lower MSD had a better clearance rate with fewer fragmentation sessions.

     

New intra-abdominal mass after operation for colorectal cancer: desmoid tumor versus peritoneal seeding

Purpose

To identify differential clinical and imaging findings between intra-abdominal desmoid tumors and peritoneal seeding that developed after surgery for colorectal cancer.

Methods

8 patients (9 desmoid tumors) and 11 patients (13 peritoneal seeding masses) were enrolled in our retrospective study. Patients with three or more tumors were excluded. Clinical findings including location of initial tumors, type of surgery, T- and N-stages of initial tumors, time interval between initial surgery and development of intra-abdominal tumors, and level of carcinoembryonic antigen (CEA) were evaluated. Imaging findings of intra-abdominal tumors including size, number, growth rate, location, shape, homogeneity, relative enhancement, and maximum standardized uptake value were evaluated. The Mann–Whitney U test and Fisher’s exact test were used to compare clinical and imaging findings between desmoid tumors and peritoneal seeding.

Results

In patients with a desmoid tumor, initial T-stage, initial N-stage, and level of CEA at the time of surgery for intra-abdominal tumor were lower than in patients with peritoneal seeding (p = 0.027, p = 0.033, and p = 0.017). The desmoid tumors were frequently located in the small bowel mesentery (p = 0.018) and were larger at detection (p = 0.041). Round or ovoid shapes on CT images were more frequently observed with the desmoid tumors (p = 0.035).

Conclusions

Stage of colorectal cancer, CEA level, and location, size, and shape of new intra-abdominal tumors can be helpful for differentiating between intra-abdominal desmoid tumors and peritoneal seeding in patients with a history of colorectal cancer surgery.

     

Preoperative Prediction of Extracapsular Extension: Radiomics Signature Based on Magnetic Resonance Imaging to Stage Prostate Cancer

Purpose

To investigate and validate the potential role of a radiomics signature in predicting the side-specific probability of extracapsular extension (ECE) of prostate cancer (PCa).

Procedures

The preoperative magnetic resonance imaging data of 238 prostatic samples from 119 enrolled PCa patients were retrospectively assessed. The samples with were randomized in a two-to-one ratio into training (n?=?74) and validation (n?=?45) datasets. The radiomics features were derived from T2-weighted images (T2WIs). The optimal radiomics features were identified from the least absolute shrinkage and selection operator (LASSO) logistic regression model and were used to construct a predictive radiomics signature via dimension reduction and selection approaches. The association between the radiomics signatures and pathological ECE status was explored. Receiver operating characteristic (ROC) analysis was used to assess the discriminatory ability of the signature. The calibration performance and clinical usefulness of the radiomics signature were subsequently assessed by calibration curve and decision curve analyses.

Results

The proposed radiomics signature that incorporated 17 selected radiomics features was significantly associated with pathological ECE outcomes (P?<?0.001) in both the training and validation datasets. The constructed model displayed good discrimination, with areas under the curve (AUC) of 0.906 (95 % confidence interval (CI), 0.847, 0.948) and 0.821 (95 % CI, 0.726, 0.894) for the training and validation datasets, respectively, and had a good calibration performance. The clinical utility of this model was confirmed through decision curve analysis.

Conclusions

The radiomics signature based on T2WIs showed the potential to predict the side-specific probability of pathological ECE status and can facilitate the preoperative individualized predictions for PCa patients.

     

18F-FDG PET/CT影像组学判断乳腺癌人表皮生长因子受体2表达状态

目的 观察¹⁸F-FDG PET/CT影像组学判断乳腺癌人表皮生长因子受体2(HER2)表达状态的价值。方法 纳入100例乳腺癌患者,包括HER2(+)组28例、HER2(-)组72例,比较组间PET/CT参数,包括病灶最大标准摄取值(SUV_max)及其标准差(SD)、平均标准摄取值(SUV_mean)及肿瘤代谢体积的差异。于标准化后的PET图像中手动分割病灶ROI,获取其影像组学特征,根据降维后的影像组学特征及其系数的线性加权获得病灶影像组学风险评分(RRS);针对差异有统计学意义的指标绘制受试者工作特征曲线,计算曲线下面积(AUC),评价其判断乳腺癌患者HER2状态的效能。采用Bootstrap 1000重复抽样进行内部验证,计算校正AUC,并以DeLong检验比较。以决策曲线分析评价患者净获益情况。结果 HER2(+)组病灶SUV_max、SUV_mean及SD均大于HER2(-)组(P均<0.05)。共获取704个影像组学特征,经筛选最终获得10个非零系数的特征用于计算...     

2D shear wave elastography: measurement acquisition and reliability criteria in noninvasive assessment of liver fibrosis

Purpose

The objective was to evaluate the accuracy of 2D shear wave elastography (SWE) in predicting stages of liver fibrosis using five individual versus grouped measurements and different reliability criteria.

Materials and methods

This is a prospective study of 109 patients who underwent hepatic 2D SWE (Canon Aplio 500) prior to liver biopsy for varied indications. Liver fibrosis was staged using the METAVIR scoring system (F = 0–4). Propagation mapping was used to guide ten SWE measurements from the liver parenchyma: five individual measurements and five grouped measurements. IQR/median, SD/median, and SD/mean were examined as quality criteria for patient inclusion at various thresholds (IQR/median ≤ 0.15, 0.2, 0.3, 0.4, 0.5; SD/median ≤ 0.15, 0.2, 0.3; SD/mean ≤ 0.2, 0.3, 0.5). Threshold for clinically significant fibrosis (F ≥ 2) was determined with receiver operating characteristic (ROC) analysis.

Results

There was high agreement between individual and grouped measurements without statistically significant differences (intraclass correlation coefficient = 0.82; p = 0.26–0.96). When no quality criterion was used (n = 103), the optimal threshold was 11.3 kPa [AUROC 0.78, 95% CI (0.69, 0.88)] with sensitivity and specificity of 80% and 66%, respectively. All quality criteria were associated with equal or higher AUROC ranging from 0.78 to 0.87. IQR/median ≤ 0.5 (n = 88) achieved the highest sensitivity of 85% and only excluded a small subset of patients. The AUROC and specificity were 0.83 [95% CI (0.74, 0.92)] and 72%, respectively.

Significance

Quality criterion IQR/median ≤ 0.5 increases sensitivity and specificity in prediction of clinically significant liver fibrosis while excluding only a small subset of patients. Grouped measurements are comparable to individual measurements and may help increase procedural efficiency.

     

Differences in early imaging features and pattern of progression on CT between intrahepatic biliary metastasis of colorectal origin and intrahepatic non-mass-forming cholangiocarcinoma in patients with extrabiliary malignancy

Purpose

To assess the differences in early imaging features and progression pattern on CT between intrahepatic biliary metastasis (IBM) and non-mass-forming cholangiocarcinoma (NMFC) in patients with extrabiliary malignancy.

Methods

This retrospective study included 35 patients who were surgically and pathologically confirmed with IBM (n = 14) or NMFC (n = 21) at the time of or after surgery for extrabiliary malignancy. Two observers evaluated the following aspects of biliary lesions on initial or follow-up CT images: location, characteristics of intrahepatic duct (IHD) dilatation, presence of duct wall thickening, and periductal infiltration lesion or periductal expansile mass.

Results

All IBMs were associated with colorectal cancer (p = 0.032). As early imaging features on CT, smooth tapered localized IHD dilatation without duct wall thickening and peripheral duct involvement were observed significantly more often in IBM, and IHD dilatation with abrupt tapering or irregularity of transition site and bile duct wall thickening were significantly more common in NMFC (all p < 0.05). Regarding progression pattern, periductal expansile mass was present only in IBM, whereas periductal infiltrative lesion was present only in NMFC (p < 0.001).

Conclusion

In the differentiation between IBM and NMFC in patients with extrabiliary malignancy, the differences in early imaging features and progression pattern of the two diseases revealed in this study would be helpful for diagnosis.