Deep Fusion Models of Multi-Phase CT and Selected Clinical Data for Preoperative Prediction of Early Recurrence in Hepatocellular Carcinoma

Weibin Wang,Hongjie Hu,Yen-Wei Chen,Qingqing Chen,Panyanat Aonpong,Qiaowei Zhang,Yutaro Iwamoto,Lanfen Lin

doi:10.1109/access.2020.3011145

Abstract

Post-operative early recurrence (ER) of hepatocellular carcinoma (HCC) is one of the leading causes of death. The prediction of the ER of HCC before treatment contributes to guiding treatment and follow-up protocols. In recent years, CT radiomics signatures have been proven effective in several studies in predicting early recurrence of HCC, there are still two major challenges. First, the radiomics features extracted were low or mid-level features, which may not fully characterize HCC heterogeneity. Second, the fusion approach of clinical textual data and image information is in little consensus. In this paper, we proposed a deep-learning based prediction model to extract high-level features from the triple-phase CT images and compare its performance with traditional radiomics model and clinical model. The accuracy and area under the curve (AUC) of receiver operating characteristics of three models was 69.52%/0.723, 67.04%/0.64, 76.03%/0.75, respectively. In addition, we proposed four fusion models to combine clinical data and high-level features. Among them, Fusion model D performed best, achieving a higher prediction accuracy of 78.66% and AUC of 0.8248. Moreover, fusion models with a joint loss function can further improve the prediction performance to 80.49% and 0.8331.

Highlights

Hepatocellular carcinoma (HCC) is the sixth most commonly diagnosed malignancy and the third leading cause of cancerrelated mortality worldwide, especially in the Asia-Pacific region [1], [2]
The results showed that the joint loss model improved the prediction accuracy of fusion model D from 78.66% to 80.49% and slightly improved area under the curve (AUC) from 0.8248 to 0.8331
The results showed that the proposed fusion model D can effectively combine computed tomography (CT) images and clinical data, with the accuracy and AUC of model D reaching to 78.66% and 0.8248, respectively, which was better than either deep learning model based on CT images only or clinical model

Summary

Introduction

Hepatocellular carcinoma (HCC) is the sixth most commonly diagnosed malignancy and the third leading cause of cancerrelated mortality worldwide, especially in the Asia-Pacific region [1], [2]. The mainstay treatments of HCC include surgical resection, radiofrequency ablation (RFA), transarterial chemoembolization (TACE), and liver transplantation. For patients with well-preserved liver function, hepatic resection is the first-line treatment strategy [3], [4]. Surgical resection is actively considered even when the patient is. The associate editor coordinating the review of this manuscript and approving it for publication was Ting Li. diagnosed with vascular tumor thrombosis or extrahepatic metastases in many experienced hepatobiliary surgical centers [5]. Postoperative recurrence either intrahepatic or extrahepatic is still the major cause of patient death [6]. The peak time of HCC recurrence is one year after resection, which is defined as ‘‘early recurrence’’ (ER) [7]. Researcher effort to identify high-risk ER individuals before treatment

Objectives

Methods

Findings

Conclusion