A Multidisciplinary AI Board for Multimodal Dementia Characterization and Risk Assessment

Dimensions overall score 8.0

GitHub Code Pulse

No public code linked for this paper yet.

• Cerebra achieved an AUROC of 0.86 for dementia diagnosis, outperforming baseline models.result 95%
• Cerebra achieved AUROCs up to 0.80 for dementia risk prediction, compared to 0.74 for the strongest single-modality model.result 95%
• Cerebra uses a multi-agent AI system to process and integrate EHR, clinical notes, and medical imaging data.method 90%
• In a reader study with experienced physicians, Cerebra improved expert accuracy by 17.5 percentage points in prospective dementia risk estimation.result 90%
• Cerebra remains robust when modalities are incomplete.technical 85%
• Cerebra was evaluated using a dataset spanning 3 million patients from four independent healthcare systems.method 90%
• Ensuring robust performance in diverse clinical settings and potential biases in model training data could be challenges.limitation 80%
• Cerebra achieved a C-index of 0.81 for survival prediction.result 90%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card