arXiv cs.AI· Emma Sarfati, Alexandre B\^one, Pietro Gori, Sisi Yang, Marc-Michel Roh\'e, Fran\c{c}ois Nicolas, Jeong-Min Lee, Jeong Hee Yoon, Maxime Ronot, Isabelle Bloch, Christophe Aub\'e·· 13 小时前AI 评分38
dLI-RADS：基于多期 CT 的可解释深度学习肝细胞癌早期诊断方法
An interpretable deep learning method for early hepatocellular carcinoma diagnosis on multiphase CT
AI 导读
研究者提出 dLI-RADS，一种从多期 CT 图像估计肝细胞癌概率的可解释 AI 方法，将神经网络预测与反映 LI-RADS 主要特征的定量影像测量结合，经逻辑回归整合。在多中心回顾性研究中，其病灶级准确率 80%，优于 LI-RADS 的 74%（p = 0.04），非劣效性检验通过。
正文 · 原文
Authors:Emma Sarfati, Alexandre Bône, Pietro Gori, Sisi Yang, Marc-Michel Rohé, François Nicolas, Jeong-Min Lee, Jeong Hee Yoon, Maxime Ronot, Isabelle Bloch, Christophe Aubé
Abstract:Early diagnosis of hepatocellular carcinoma is critical to preserve access to curative treatment options. The Liver Imaging Reporting and Data System (LI-RADS) provides a standardized framework for non-invasive diagnosis on multiphase computed tomography, based on major visual imaging features. However, these features may be difficult to assess consistently, particularly in small liver lesions. We developed dLI-RADS, an interpretable artificial intelligence method designed to estimate the probability of hepatocellular carcinoma from multiphase computed tomography images of manually segmented liver lesions, while remaining aligned with LI-RADS-related imaging concepts. dLI-RADS combines neural network-derived predictions with handcrafted quantitative imaging measurements reflecting LI-RADS major features and integrates these inputs using logistic regression. In this multicenter retrospective study, patients with multiphase computed tomography and at least one liver lesion were included, while lesions larger than 40 mm were excluded. Histology served as the reference standard for hepatocellular carcinoma, and a composite reference standard was used for non-hepatocellular carcinoma lesions. The lesion-level accuracy of dLI-RADS and LI-RADS were compared on the internal evaluation set using 10,000-repetition bootstrap non-inferiority tests, with 5% margins and significance levels. On the external test set, sensitivities were similarly compared. dLI-RADS was non-inferior to LI-RADS for lesion-level accuracy in the internal evaluation set, achieving 80% accuracy compared with 74% for LI-RADS (p = 0.04)...
| Comments: | Accepted at Scientific Reports |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.29616 [cs.CV] |
| (or arXiv:2610.29616v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.29616 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Pietro Gori [view email]
[v1]
Wed, 21 Oct 2026 11:56:23 UTC (8,312 KB)
来源：arXiv cs.AI · arxiv.org