rare-disease

Montana's new right-to-try law offers hope to families like Kris DeVault's, whose son Brody has a rare genetic condition that an experimental, unapproved drug might help treat.

Anthropic announces grants of up to $50,000 in Claude usage credits for researchers working on rare diseases under its AI for Science program.

Introduces RareDxR1, an end-to-end reasoning-centric large language model for open-domain rare disease diagnosis from unstructured clinical notes, using a progressive training framework and reflection-enhanced reasoning sampling, achieving state-of-the-art accuracy.

This paper presents AgentODE, a framework that uses an LLM to propose ODE structures and a tool-augmented agent to refine parameter distributions from aggregate summary statistics alone, enabling mechanistic modeling of rare diseases under data scarcity and privacy constraints.

Talos is an open-source tool for automated, iterative genomic reanalysis that resolves the bottleneck of human review time, recovering 90% of diagnoses while surfacing only 1.3 candidate variants per patient.

This paper presents RaDaR, a 32B open-source reasoning LLM trained on public and synthetic rare disease cases, which outperforms larger models like DeepSeek-R1 in diagnosis benchmarks and improves physician accuracy by 21.44 percentage points in a randomized trial.

OpenAI highlights how o3 Deep Research can aid rare disease diagnosis by integrating clinical features, inheritance patterns, variant evidence, and scientific literature into actionable hypotheses for specialists.

MedLatentDx proposes a latent multi-agent communication framework for cross-hospital rare-disease diagnosis, using latent KV blocks to share diagnostic evidence without exposing clinical text, and introduces the CrossRare-Bench benchmark.

Boston Children's Hospital has integrated AI across its clinical and operational infrastructure, using a secure ChatGPT environment to diagnose over 40 rare conditions, reduce operational costs, and improve care delivery.

OpenAI collaborated with Boston Children's Hospital and Harvard's Manton Center on a study covering 376 cases, using AI workflows to assist in diagnosing rare diseases and leading to 18 confirmed diagnoses. The core approach was to have AI perform literature retrieval, hypothesis ranking, and evidence summarization, which were then reviewed by human geneticists, rather than directly providing conclusions.

Anthropic announces a focused call for rare disease research grants under their AI for Science program, offering up to $50,000 in Claude credits to researchers and early-stage biotechs.