Researchllmclinical nlpmass general brighamagents
Mass General Brigham Develops Autonomous Cognitive Screening
9.3
Mass General Brigham researchers develop a fully autonomous AI system to screen for cognitive impairment, publishing results in npj Digital Medicine and releasing the open-source Pythia tool. Tested on 3,300 clinical notes from 200 patients, the system achieved 98% specificity, 91% sensitivity in balanced tests and 62% sensitivity in real-world prevalence. The workflow runs locally with no external data transmission.
Key Points
- 1Develops autonomous LLM-based screening system achieving 98% specificity in real-world validation
- 2Targets underdiagnosed cognitive impairment using routine clinical notes, enabling earlier Alzheimer’s therapy eligibility
- 3Releases open-source Pythia for local autonomous prompt optimization; preserves patient data by avoiding cloud
Scoring Rationale
Peer-reviewed novelty with open-source, deployable agentic workflow drives high impact, limited by reduced real-world sensitivity and single-center data.
Sources
Public references used for this report.
Practice with real Health & Insurance data
90 SQL & Python problems · 15 industry datasets
Used by DS/ML engineers at top companies
250 free problems · No credit card
See all Health & Insurance problems
