Artificial intelligence serving medical research in Hepatology. A study carried out by Milano-Bicocca University’s Autoimmune Liver Disease Centre at the San Gerardo Hospital in Monza, alongside the Rulex Data Science Team in Genoa, has identified four new Primary Biliary Cholangitis (PBC) subtypes using clinical data from over 12,000 individuals from all over the world. The new algorithm can be added to existing prognostic scores to improve prognostic assessment for patients when they are first diagnosed.
“This is a really important study for us as patients, given the large number of Italian patients included and the potential for innovation offered by artificial intelligence,” explains Davide Salvioni, president of AMAF Onlus, the Italian association of patients with autoimmune liver disease. “A better understanding of these disorders will certainly have a positive impact on doctors’ abilities to manage them more effectively.”
PBC is a liver disease that, although rare, still affects over 10,000 people in Italy, primarily women aged over 40. The past decade has seen gradual improvements in severity stratification for patients with PBC, due in part to the development of new scores and calculators.
In recent years, artificial intelligence and machine learning have been successfully used in the study of common diseases, from infections to cardiovascular disease, breast cancer to colorectal tumors. But until now, experimental evidence for these new technologies and their applications for rare diseases, particularly PBC, was lacking.
The team at Monza’s Autoimmune Liver Disease Centre led by Professor Pietro Invernizzi used Rulex, an innovative data analysis tool. It utilizes a sophisticated artificial intelligence algorithm developed by the Rulex research and development team, coordinated by CEO Marco Muselli, and is based on a theoretical model devised by the CNR Institute for Electronics, Information Technology and Telecommunications in Genoa.
Published in Liver International journal, the study gathered the largest international cohort of PBC patients ever considered, including people from Europe, Japan and North America (DOI: 10.1111/liv.15141). The aim of the research was to draw on the huge amount of data to improve risk stratification for this rare disease. Four subgroups for the disease were identified, in order of increasing clinical severity, based on just three laboratory values: albumin, bilirubin and alkaline phosphatase.
“The Rulex team led by Damiano Verda grouped patients with PBC using a brand- new approach, creating very simple rules that can be applied by clinicians to classify new patients when they are first diagnosed,” explains Dr. Alessio Gerussi, study leader and researcher at Monza’s Autoimmune Liver Disease Centre.
“But our work doesn’t finish here: in the future, targeted studies will integrate clinical data with data from genetic sequencing, radiology imaging and digital scans of histological slides,” Gerussi goes on. “The objective is to illustrate the disease’s heterogeneity using a much more sophisticated approach, so we can offer patients customized treatment, the ultimate aim of Precision Medicine.”
Read the press release in Italian, Ricerca medica, nuovi risultati in Epatologia con l’intelligenza artificiale.
Learn more about Rulex for medical research.