Document resource
In this issue of the Journal of Clinical Pathology, Dylan Windell and colleagues present a machine-learning model for evaluating liver inflammation in patients with metabolic-dysfunction-associated steatohepatitis (MASH) and autoimmune hepatitis (AIH).1 Unlike many recent pathology artificial intelligence (AI) systems that lean on weakly or self-supervised paradigms popular in tumour biology, this is a fully (indeed, deeply) supervised model designed to think like a pathologist.2