Deep learning for liver lesion segmentation and classification on staging CT scans of colorectal cancer patients: a multi-site technical validation study.

Abstract

AIM: To validate a liver lesion detection and classification model using staging computed tomography (CT) scans of colorectal cancer (CRC) patients. MATERIALS AND METHODS: A UNet-based deep learning model was trained on 272 public liver tumour CT scans and tested on 220 CRC staging CTs acquired from a single institution (2014-2019). Performance metrics included lesion detection rates by size (20 mm), segmentation accuracy (dice similarity coefficient, DSC), volume measurement agreement (Bland-Altman limits of agreement, LOAs; intraclass correlation coefficient, ICC), and classification accuracy (malignant vs benign) at patient and lesion levels (detected lesions only). RESULT(S): The model detected 743 out of 884 lesions (84%), with detection rates of 75%, 91.3%, and 96% for lesions 20 mm, respectively. The median DSC was 0.76 (95% CI: 0.72-0.80) for lesions 20 mm. Bland-Altman analysis showed a mean volume bias of -0.12 cm3 (LOAs: -1.68 to +1.43 cm3), and ICC was 0.81. Lesion-level classification showed 99.5% sensitivity, 65.7% specificity, 53.8% positive predictive value (PPV), 99.7% negative predictive value (NPV), and 75.4% accuracy. Patient-level classification had 100% sensitivity, 27.1% specificity, 59.2% PPV, 100% NPV, and 64.5% accuracy. CONCLUSION(S): The model demonstrates strong lesion detection and segmentation performance, particularly for sub-centimetre lesions. Although classification accuracy was moderate, the 100% NPV suggests strong potential as a CRC staging screening tool. Future studies will assess its impact on radiologist performance and efficiency. Copyright © 2025 The Royal College of Radiologists