(ANSA) – ROME, 02 SEPTEMBER – Machine learning helps distinguish subtypes of rheumatoid arthritis and could potentially improve diagnosis and treatment for those who suffer from it. The results of the new application based on artificial intelligence were published in the journal Nature Communications.
There are several studies measuring the effectiveness of machine learning in oncology. Now researchers led by Fei Wang, founding director of the Institute of AI for Digital Health at Weill Cornell Medicine (New York City), have worked to expand the use of this technology to the subtyping process of rheumatoid arthritis.
Distinguishing between the three subtypes of the disease, in fact, can help doctors choose which therapy is most likely to be effective for a given patient.
Currently, pathologists manually classify arthritis subtypes, a slow process that increases research costs and can lead to inconsistencies. The team first trained their new algorithm on tissue samples from a set of mice with the disease and then validated the tool on tissue biopsies from patients with rheumatoid arthritis, showing that it could effectively type human clinical samples. “If you can create an algorithm that identifies a patient’s subtype, you’ll be able to provide the treatments that people need more quickly,” says Richard Bell, a computational pathology analyst in the Molecular Histopathology Core Laboratory at the Hospital for Special Surgery.
“The findings demonstrate the growing impact of AI in advancing personalized medicine,” said Rainu Kaushal, chair of the Department of Population Health Sciences at Weill Cornell Medicine. The tool could also increase the efficiency of clinical trials testing treatments for patients with different subtypes. (ANSA).
2024-09-02 11:50:28