How AI can help advance liver disease research

Gilead Sciences, a Foster City, California-based biopharmaceutical company, is scheduled to present new AI-powered research related to nonalcoholic steatohepatitis (NASH) at The Liver Meeting 2019 in Boston. The research is the result of a collaboration with Boston-based PathAI.

“By combining data from across our NASH clinical development program with AI-based tools, we have the opportunity to better characterize this complex disease and understand how potential therapies can impact disease progression,” Mani Subramanian, MD, senior vice president of liver diseases for Gilead Sciences, said in a prepared statement. “Applying PathAI’s deep learning research platform for liver histology assessment will enable a more rigorous review of treatment response and has potential for the exploration of novel biology in patients with advanced fibrosis due to NASH.”

The two companies worked together to explore how machine learning can be used to improve the diagnosis and staging of NASH, comparing the performance of experienced pathologists to AI-powered techniques. PathAI’s research platform was trained on more than 68,000 annotations from a team of 75 pathologists before it was evaluated. The study’s results suggested that, yes, AI can achieve a performance comparable to that of trained pathologists.

“The evaluation of new therapies for NASH can be advanced with quantitative and reproducible assessment of liver pathology,” Andy Beck, MD, PhD, PathAI co-founder and CEO, said in the same statement. “We are thrilled to apply the PathAI research platform to support development of new treatment approaches.”

The Liver Meeting, an annual scientific conference hosted by the American Association for the Study of Liver Diseases (AASLD), occurs from Nov. 8 to Nov. 12, 2019. More information is available on the AASLD website.