The European Medicines Agency (EMA) is paving the way for advancements in medical technology with its recent endorsement of an artificial intelligence (AI) tool called AIM-NASH. This innovative tool is designed to help pathologists more accurately assess liver biopsy scans, particularly for determining the severity of metabolic dysfunction-associated steatohepatitis (MASH). If left untreated, MASH can lead to serious liver disease, so this development is a big deal for both patients and healthcare providers.
Previously known as non-alcoholic steatohepatitis (NASH), MASH involves fat buildup in the liver, causing inflammation and scarring without alcohol being a factor. It’s often linked to conditions like obesity, type 2 diabetes, and high cholesterol. With AIM-NASH, clinical trials for MASH treatments are expected to become more accurate and efficient. The tool aims to reduce variability in measuring disease indicators like inflammation and fibrosis, making trial results more reliable and consistent.
After a period of public consultation, the EMA’s Committee for Medicinal Products for Human Use (CHMP) gave AIM-NASH the green light. This means that any future evidence generated using this tool will be scientifically valid, potentially speeding up the development of effective therapies for patients.
Currently, liver biopsies are the gold standard for evaluating new treatments. They involve taking small tissue samples to confirm inflammation and scarring. However, inconsistencies in evaluating these samples have been a hurdle. AIM-NASH addresses this challenge by providing consistent results, verified by a single expert pathologist rather than relying on the consensus of three independent specialists.
The AI system behind AIM-NASH is powered by a machine learning model trained on over 100,000 annotations from 59 pathologists across nine extensive clinical trials. The model is ‘locked’ to ensure it remains unchanged, although any significant updates will require re-qualification by the CHMP.
The EMA is actively working to ensure AI applications in medicine are safe and responsibly integrated across Europe’s regulatory networks. Their coordinated AI workplan, developed with the Heads of Medicines Agencies, reflects this commitment.
