Exploring Agentic AI and its Implementation into MS Research

WATCH TIME: 5 minutes | Captions are auto-generated and may contain errors.

Agentic artificial intelligence (AI) refers to AI systems designed to autonomously plan, reason, and execute multistep tasks in pursuit of defined goals, adapting dynamically based on new inputs rather than responding to single, isolated prompts. Unlike traditional machine learning models that perform narrowly scoped analyses, agentic AI can coordinate complex workflows, such as integrating electronic health records, imaging, genomic data, and other real-world data sources, while iteratively refining outputs with limited human intervention.¹

In neurology, these systems are being explored for applications including MRI pattern recognition, disease phenotyping, predictive modeling, and large-scale real-world data interrogation, with the potential to accelerate hypothesis generation and clinical insight.² Although promising, implementation requires careful consideration of model transparency, data governance, regulatory oversight, and ethical safeguards to ensure safe and equitable use in clinical and research settings.³

A reoccurring theme throughout the Americas Committee for Treatment and Research in Multiple Sclerosis (ACTRIMS) Forum 2026, held February 5 to 7, 2026, in San Diego, California, was the continued integration of AI into the field of MS research. Among the clinicians at the forefront of this research is Gianmarco Bellucci, MD, clinician and researcher at the University of California, San Francisco. Recognized at the conference as a Young Investigator, Bellucci presented “Agentic AI for Multiple Sclerosis Research: Real-World Applications of the MedCP Framework,” highlighting practical use cases and implementation strategies.

To gain further insight, NeurologyLive^® spoke with Bellucci, who presented new research on integrating agentic artificial intelligence into the analysis of real-world data in multiple sclerosis (MS). In the discussion, he described the development of a clinician-facing application designed to simplify complex big data analysis, enabling researchers to test hypotheses more efficiently without requiring advanced technical expertise. Bellucci also addressed how AI is reshaping MS research, from MRI pattern recognition and genetic disease detection to multimodal data integration, and highlighted the potential of federated learning models to bridge gaps between institutions while accelerating clinically meaningful discoveries.

For more ACTRIMS 2026 coverage, click here.

REFERENCES
1. Russell S, Norvig P. Artificial Intelligence: A Modern Approach. 4th ed. Pearson; 2021.
2. Esteva A, Robicquet A, Ramsundar B, et al. A guide to deep learning in healthcare. Nat Med. 2019;25(1):24–29.
3. Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019;25(1):44–56.