Liver Cirrhosis Stage Estimation from MRI with Deep Learning

Jun Zeng, Debesh Jha, Ertugrul Aktas, Elif Keles, Alpay Medetalibeyoglu, Matthew Antalek, Federica Proietto Salanitri, Amir A. Borhani, Daniela P. Ladner, Gorkem Durak, Ulas Bagci · 2025

Read Liver Cirrhosis Stage Estimation from MRI with Deep Learning by Jun Zeng, Debesh Jha, Ertugrul Aktas, Elif Keles, Alpay Medetalibeyoglu, Matthew Antalek, Federica Proietto Salanitri, Amir A. Borhani, Daniela P. Ladner, Gorkem Durak, Ulas Bagci (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-18225.txt.

Description

Jun Zeng and co-authors build a deep learning system that grades liver scarring from MRI scans, in a research paper from 2025. Cirrhosis is severe scarring of the liver and ends many long-term liver diseases. Finding it early matters, but early stages are hard to spot and patients often arrive with grave problems.

The system reads more than one type of MRI scan at once and learns signs at several scales. A part of the model weighs which scan type matters most for each sign. The team trained and tested it on a public set of 628 scans from 339 patients. Their best model sorted scans into three stages with 72.8 percent accuracy on one scan type. That score beat older methods based on hand-built image measures.

Catalog record

CREATORJun Zeng, Debesh Jha, Ertugrul Aktas, Elif Keles, Alpay Medetalibeyoglu, Matthew Antalek, Federica Proietto Salanitri, Amir A. Borhani, Daniela P. Ladner, Gorkem Durak, Ulas Bagci
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE23 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-18225

Browse the full library · About the Tower of Babel