ACTRIMS 2026 | The use of real-world data in the MedCP framework for MS research
Gianmarco Bellucci, MD, PhD, University of California, San Francisco, CA, discusses the application of real-world data in the MedCP framework for multiple sclerosis (MS) research. Dr Bellucci highlights the importance of using real-world data, as opposed to trial data, and plans to incorporate unstructured clinical data, such as clinical notes and radiology reports, to further enrich the system. This interview is part of our coverage of the 11th Annual Americas Committee for Treatment and Research in Multiple Sclerosis (ACTRIMS) Forum, held in San Diego, CA.
These works are owned by Magdalen Medical Publishing (MMP) and are protected by copyright laws and treaties around the world. All rights are reserved.
Transcript
Our framework up to now is our research framework and the core advancement compared to other attempts that have been done in MS is the utilization of real-world data. And again, a curated source of biomedical information to contextualize the findings. There’ll be many attempts and there are many attempts of using AI systems to redefine MS cores, to identify new MS subtypes, to identify new therapeutic strategies or new therapeutic paths using the currently approved drugs...
Our framework up to now is our research framework and the core advancement compared to other attempts that have been done in MS is the utilization of real-world data. And again, a curated source of biomedical information to contextualize the findings. There’ll be many attempts and there are many attempts of using AI systems to redefine MS cores, to identify new MS subtypes, to identify new therapeutic strategies or new therapeutic paths using the currently approved drugs. But many of these have been tested on trial data, which are per se slightly different, substantially different from real-world data. So our system is supposed to play with real-world data, but is actually capable of using any source of data. And aside from this, so the next step for us, but in general, will be to embed also unstructured clinical data, meaning the data that comes from clinical notes, from radiology reports, and in order to further enrich the information that we feed, which we feed the system. So in the future, we foresee that all these tools of these resources, including ours, can actually, again, democratize research in MS and allow also for federated learning, federated analysis in parallel across several centers, so that eventually this information will be easier to bring into the clinics, into actual real-world applications. So I want to acknowledge my lab mate and colleague, Wanjun Gu, who is the technical mind behind MedCP. I’m a clinician and clinician scientist. So this is also a very crucial aspect. So we represent in a way the relationship that in the future clinicians will have to establish with technical people, but per se with AI tools. So having this assistance for, again, research applications, clinical decisions, and all in all to optimize and improve the care of people with multiple sclerosis.
This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.
Read more...
