Can AI and Machine Learning Help Rare Patients Get Diagnosed and Treated Faster?

A research team from IPM.ai, a subsidiary of Swoop, and Insmed, a biotechnology company, recently collaborated on a case study that demonstrated how AI and machine learning could be used to help patients get the treatment they need in a more timely manner. The research focused on patients affected by Mycobacterium avium complex (MAC) lung disease, a rare infection that primarily affects people that are immunocompromised.

About MAC Lung Disease

MAC lung disease is a type of rare, infectious lung disease. It is caused by one of two species of bacteria: Mycobacterium avium or M. intracellulare. These widespread bacteria are almost never a threat to humans under normal circumstances; however, in patients with weakened immune systems, such as those living with HIV/AIDs, undergoing chemotherapy, or living with primary immunodeficiency, these bacteria present a risk. Symptoms of infection include cough, fever, night sweats, weight loss, sputum production, lethargy, diarrhea, and infection in other areas of the body. The treatment of MAC lung disease often involves the use of a combination of two or three antibacterial drugs. It can take some time for the infection to be eradicated, and treatment may last for a year or more. Outcomes can vary and often depend on the underlying condition of the patient. To learn more about MAC lung disease, click here.

The antibiotic ARIKAYCE was first approved in 2018 by the US Food and Drug Administration (FDA) as a treatment for MAC lung disease that is refractory, meaning that it is not longer being affected by other antibiotic regimens. Specifically, it is recommended for patients who do not have a bacteria-negative sputum culture after six months of treatment with other antibiotics. The goal of this study was to identify providers that were most likely to be treating patients with refractory disease so that Insmed, the company that developed the treatment, can reach out to them about the availability and impacts of it.

Harnessing AI

Research projections estimate that there could be around 17,000 patients in the US that could benefit from ARIKAYCE. IPM.ai and Insmed were working with a massive pool of data that included information from 300 million patients. This data covered a one year period. In order to identify the possible patients, applied aggregated learnings (using AI and machine learning) were utilized by the researchers and patients were linked to their providers based on how often they visited and how recently they visited.

Machine learning identified patterns that were primary based on their diagnoses, which procedures they had received, and what medicines they had been prescribed. By identifying these patients, commercial teams from Insmed now had a list of health providers to reach out about ARIKAYCE. While this obviously helps the company with making sales, it also means that patients will be able to access this valuable treatment more rapidly.

Machine learning and AI are advanced data calculation and research tools with tangible benefits, and they have the potential to help improve outcomes for rare disease patients.

Click here to learn more about this treatment.

Share this post

Share on facebook
Share on google
Share on twitter
Share on linkedin
Share on pinterest
Share on print
Share on email
Close Menu