Ultrasound grading improved by view-based regional evidence for liver fibrosis
VCRE-Fib: View-Conditioned Regional Evidence for Fine-Grained Ultrasound Grading of Schistosoma japonicum-Associated Liver Fibrosis
Computer Vision and Pattern RecognitionMachine Learning
Summary
Assessing liver damage caused by Schistosoma japonicum infection is important but hard to do in detail with ultrasound images. The authors developed a computer method called VCRE-Fib that considers how the ultrasound image was taken and focuses on different parts of the liver to better judge the damage. This approach not only predicts the overall liver condition but also shows where abnormalities might be in the images, allowing doctors to check the results spatially. Their tests on a large collection of ultrasound images from many patients showed improvements over previous methods.
What this means in practice
- •For medical imaging teams: Provide more accurate and interpretable ultrasound grading of liver fibrosis caused by Schistosoma japonicum using view-conditioned regional deep learning analysis.
- •For public health monitoring teams: Enhance disease management and follow-up for endemic regions by delivering better fibrosis severity assessments from large ultrasound datasets.
Authors
Ziyang Xu, Shuli An, Hao Zhou, Haitian Zhong, Tingting Wu, Tao Wang, Kun Yang, Tieyong Zeng
Abstract
Accurate assessment of Schistosoma japonicum-associated liver fibrosis is essential for disease management and long-term follow-up in endemic regions. Ultrasound provides non-invasive imaging, but complex local echogenic patterns and anatomical structures make fine-grained grading challenging. Existing deep learning methods can predict fibrosis scores, yet directly incorporating acquisition views and regional cues into grading while retaining spatial information for inspection remains an open problem. Here we present VCRE-Fib, a view-conditioned regional evidence framework that integrates anatomical context, local information, and global image assessment for fine-grained ultrasound grading. The framework forms view-conditioned local grading evidence before spatial pooling, uses weak localization to guide its aggregation, and combines it with global predictions. Image-only inference jointly returns a fibrosis score, acquisition view, and candidate abnormal-region map. We developed and evaluated the method on a re-curated cohort of 108,709 ultrasound images from 6,373 patients across 35 centers. On a patient-disjoint test set of 4,107 images from 240 patients across four centers, VCRE-Fib reduced the prespecified composite grading risk by 7.115% relative to SFibAI trained and evaluated on the same data split. Image-level mean absolute error decreased from 0.391 to 0.378, alongside lower patient-max, patient-median, and center-balanced risks. The full model also achieved lower composite grading risk than variants that separately removed view conditioning or weak localization. These results support incorporating anatomical context and regional evidence into ultrasound grading while exposing spatial predictions for inspection alongside severity estimates.