AI Transforms Rare Disease Diagnosis as Awareness Day Highlights Patient Challenges

Artificial intelligence is revolutionizing the diagnosis and treatment of rare diseases, cutting diagnostic timelines from years to weeks, as health experts mark Rare Disease Day on February 28. More than 300 million people globally are affected by over 7,000 distinct rare diseases, according to the World Health Organization.

AI algorithms can process medical records, genetic information, and images to identify patterns that may escape human detection, allowing for earlier and more accurate diagnoses. Rare diseases often take years to diagnose because of the lack of data, overlapping symptoms, and unclear diagnosis.

In a February paper published in the journal Nature, researchers from the Shanghai Jiao Tong University in China presented an AI system called DeepRare, a multi-agent system for rare disease differential diagnosis decision support, powered by large language models, integrating more than 40 specialized tools. Using the specialized tools and knowledge sources, the agentic AI system generates ranked diagnostic hypotheses for rare diseases. Each of the tools was also accompanied by reasoning that links the conclusions to verifiable medical evidence.

Harvard University-based researchers in a paper also published in the Nature journal in 2025 described an AI tool called PopEVE, which can identify genetic variants most likely to cause severe disease or death. The model was able to identify more than 100 novel alterations responsible for undiagnosed, rare genetic diseases.

By integrating deep phenotyping with AI, clinicians can decipher unknown genetic variants and accelerate the development of both repurposed and next-generation precision-therapies. AI is also instrumental in accelerating drug development by processing enormous amounts of biomedical data to discover new targets for drugs and repurpose existing ones. Machine learning algorithms enable the prediction of patient responses to particular treatments, making it possible to develop personalized treatment strategies.

AI-enabled infrastructure is helping to achieve geographic neutrality, delivering the same caliber of healthcare screening to rural villages and Tier-3 cities as to Tier-1 hubs, ensuring a high-quality healthcare gateway that is no longer defined by where the patient lives. AI-based platforms bring together researchers from across the globe, making it easier to share data.

The World Health Organization defines a rare disease as an often debilitating, chronic, or degenerative condition affecting 1 or fewer per 1,000 population. According to the WHO ICD-11 (International Classification of Diseases), there are over 5,500 rare diseases. It also assigns unique identifiers (URIs) to them for better tracking. The Center for Disease Control and Prevention defines a rare disease as a condition that affects fewer than 200,000 people.

For families living with rare diseases, the experience can feel isolating and overwhelming. Families often spend years seeking a diagnosis or treatment plan that addresses the full scope of their child's needs. Rare diseases are genetic and require coordinated, multidisciplinary care.

Specialized care centers have developed expertise in conditions that many health systems encounter infrequently. These facilities provide care for a broad range of rare diseases, including Spina Bifida, a congenital condition affecting the spine and nervous system that requires ongoing, multidisciplinary management; Duchenne Muscular Dystrophy (DMD), a progressive neuromuscular disorder that affects muscle strength and function; Rett Syndrome, a genetic neurological disorder that primarily affects girls; and Spinal Muscular Atrophy (SMA), a genetic condition that impacts motor nerve cells and muscle control.

Specialists across neurology, orthopedics, rehabilitation, complex care pediatrics, genetics and other disciplines work together to develop coordinated care plans using collaborative, team-based models. Research partnerships aimed at improving diagnostic accuracy, advancing treatment options and expanding knowledge about rare conditions continue to advance the field.