Written by : Saloni Tyagi
March 22, 2025
The tool employs a machine learning model trained on over 100,000 annotations made by 59 pathologists who analyzed over 5,000 liver biopsy samples across nine large clinical trials.
The European Medicines Agency (EMA) has approved AIM-NASH, a tool that utilizes artificial intelligence (AI) for evaluating the severity of metabolic dysfunction-associated steatohepatitis (MASH) in clinical trials.
Developed by Boston-based PathAI, AIM-NASH assists pathologists in analyzing liver biopsy scans to assess histology in a standardized way. The tool employs a machine learning model trained on over 100,000 annotations made by 59 pathologists who analyzed over 5,000 liver biopsy samples across nine large clinical trials.
The tool provides reliable assessments of disease activity with less variability than traditional methods. According to EMA, this breakthrough is expected to enhance consistency in clinical trials and improve the evaluation of new treatments.
Metabolic dysfunction-associated steatohepatitis (MASH) is a progressive liver disease linked to obesity, Type 2 diabetes, hypertension, abnormal cholesterol levels, and abdominal fat. As per the American Liver Foundation, the condition affects approximately 1.5% to 6.5% of adults in the United States. If left untreated, it can lead to severe liver complications.
Currently, Rezdiffra by Madrigal Pharmaceuticals is the only US-approved medication for MASH. Pharmaceutical companies such as Novo Nordisk and Eli Lilly are also conducting clinical trials for their GLP-1 treatments to address the disease. The EMA is currently supervising technological advancements through a multi-year plan to ensure the responsible integration of such innovations into the European medicines network.
The Big Data Steering Group created the technology work plan, which received approval from the Heads of Medicines Agencies (HMA) in November 2023. It was subsequently endorsed by the Management Board of the EMA in December 2023.