Machine-Learning-Enhanced Non-Invasive Testing for MASLD Fibrosis: Shallow-Deep Neural Networks Versus FIB-4, Tabular Foundation Models, and Large Language Models
Researchers have developed a machine-learning enhanced non-invasive testing method for detecting advanced fibrosis in MASLD patients. This new approach, utilizing a shallow-deep neural network (s-DNN), demonstrated improved diagnostic accuracy compared to the traditional FIB-4 method in external validation cohorts. The s-DNN achieved better ROC-AUC scores and maintained a balanced operating profile with significantly fewer trainable parameters than other models like TabPFN and GPT-4o. AI
IMPACT Presents a novel machine learning approach that could improve diagnostic accuracy for liver disease.