Preoperative Diagnosis of Hepatocellular Carcinoma Patients with Bile Duct Tumor Thrombus using Deep Learning Method

Preoperative diagnosis of bile duct tumor thrombus (BDTT) is clinically important as the surgical prognosis of hepatocellular carcinoma (HCC) patients with BDTT is significantly different from that of patients without BDTT. The current diagnosis of BDTT is usually based on identifying dilated bile ducts (DBDs) on medical images (eg., CT and MRI images). However, it is easy for doctors to ignore DBDs when reporting imaging scan results, leading to a high misdiagnosis rate in practice. The aim of the present study was to develop an artificial intelligence (AI) pipeline for diagnosing HCC patients with BDTT using medical images. The proposed AI pipeline includes two stages. First, the object detection neural network Faster R-CNN is adopted to identify DBDs; then, an HCC patient is diagnosed to have BDTT if the proportion of images with at least one identified DBD exceeds some threshold value. The proposed AI pipeline was applied to a real dataset consisting of 2,611 CT images collected from 34 HCC patients (16 with BDTT and 18 without BDTT). The average true positive rate for identifying DBDs per patient was 0.92, while the patient-level true positive rate for diagnosing BDTT was 0.94. The area under ROC curve for patient-level diagnosis of BDTT was 0.92 (95% CI: 0.83, 1.00), compared with 0.71 (95% CI: 0.52, 0.89) by random forest based on preoperative clinical variables. These results demonstrated that the proposed AI pipeline is successful in the diagnosis of BDTT. The automatic detection of DBDs is a key step in early diagnosis of HCC patients with BDTT, and is helpful in the treatment and prognosis of these patients.

An Artificial Intelligence Pipeline for Diagnosing Hepatocellular Carcinoma Patients with Bile Duct Tumor Thrombus

Background and purpose: Preoperative diagnosis of bile duct tumor thrombus (BDTT) is clinically important as the surgical prognosis of hepatocellular carcinoma (HCC) patients with BDTT is significantly different from that of patients without BDTT. The preoperative diagnosis of BDTT is usually based on identification of dilated bile ducts (DBDs) on medical images (eg., CT and MRI images). However, it is easy for doctors to ignore DBDs when reporting the imaging scan result, leading to a high misdiagnosis rate in practice. The aim of the present study was to develop an artificial intelligence (AI) pipeline for diagnosing HCC patients with BDTT using medical images. Methods: The proposed AI pipeline included two stages. First, the object detection neural network Faster R-CNN was adopted to identify DBDs; then, an HCC patient was diagnosed to have BDTT if the proportion of images with at least one identified DBD exceeds some threshold value. Four-fold cross validation was used to evaluate the performance of the proposed AI pipeline. Results: The proposed AI pipeline was applied on a real dataset consisting of CT images collected from 34 HCC patients (16 with BDTT and 18 without BDTT). The average true positive rate for identifying DBDs per patient was 0.92, while the patient-level true positive rate for diagnosing BDTT was 0.94. The AUC value of patient-level diagnosis of BDTT was 0.92 (95% CI: 0.83, 1.00), compared with 0.71 (95% CI: 0.52, 0.89) by random forest. Conclusions: This study first proposes an AI pipeline to identify DBDs and diagnose BDTT, and the high accuracies demonstrate that it is successful in the diagnosis of BDTT.

Imaging Features of Hepatocellular Carcinoma With Bile Duct Tumor Thrombus: A Multicenter Study

ObjectivesThere are still challenging problems in diagnosis of hepatocellular carcinoma (HCC) with bile duct tumor thrombus (BDTT) before operation. This study aimed to analyze the imaging features of HCC with B1–B3 BDTT.Materials and MethodsThe clinicopathological data and imaging findings of 30 HCC patients with B1–B3 BDTT from three high-volume institutions were retrospectively reviewed. A total of 631 patients without BDTT who were randomly collected from each of the enrolled centers were recorded as the control group to analyze the differences in clinicopathological characteristics and imaging features between the two groups. A total of 453 HCC patients who underwent surgical treatment in the three institutions from January 2020 to December 2020 were collected for a blinded reading test as the validation group.ResultsHCC patients with B1–B3 BDTT had more advanced tumor stages and adverse clinicopathological features. HCC lesions were detected in all patients, and intrahepatic bile duct dilation was observed in 28 (93.3%) patients with B1–B3 BDTT and 9 (1.43%) patients in HCC without BDTT. The intrahepatic bile duct dilation showed no enhancement at hepatic arterial phase (HAP) and no progressively delayed enhancement at portal venous phase (PVP), but it was more obvious at PVP on CT. In the reports of the 30 HCC patients with B1–B3 BDTT generated for the image when the scan was done, BDTT was observed in all 13 B3 patients and 3 of 12 B2 patients, but none of the 5 B1 patients. Fourteen patients were misdiagnosed before surgery. However, when using intrahepatic bile duct dilation in HCC patients as a potential biomarker for BDTT diagnosis, the sensitivity and specificity for BDTT diagnosis were 93.33% and 98.57%, respectively. The blinded reading test showed that intrahepatic bile duct dilation in CT and MRI scans could be for separating HCC patients with B1–B3 BDTT from HCC patients without BDTT.ConclusionsThe HCC lesions and intrahepatic bile duct dilation on CT or MRI scans are imaging features of HCC with BDTT, which might facilitate the early diagnosis of B1–B3 BDTT.