Interpretable ai matches experts in early liver cancer ct diagnosis
An interpretable deep learning method for early hepatocellular carcinoma diagnosis on multiphase CT
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Early detection of liver cancer is very important to enable effective treatment. The authors created an artificial intelligence method called dLI-RADS that analyzes specialized liver CT scans to estimate the chance a lesion is cancerous. This method combines neural networks with known imaging features doctors use for diagnosis, making it easier to understand. Their tests show dLI-RADS performs at least as well as the existing expert system LI-RADS in identifying small liver tumors.
What this means in practice
- •For radiology departments: Assist radiologists in reliably diagnosing early liver cancer from multiphase CT scans of small liver lesions without losing interpretability.
- •For medical imaging software developers: Incorporate dLI-RADS interpretability features combining neural network predictions with traditional imaging measures to enhance liver cancer diagnostic tools.
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)...