paperarXivTrust 82 · PrimaryPublished 8d agoLive · 7d ago

Multi-Large Language Model Orchestrated Severity Assessment of Clinical Records (MOSAIC)

Background: Disease severity is a multidimensional construct difficult to capture with rule-based approaches in Electronic Healthcare Records (EHR). Agentic large language model (LLM) systems could synthesise clinical evidence and reason over EHRs, but remain unevaluated for this task. Methods: MOSAIC is a two-phase agentic LLM framework for severity phenotyping, using type 2 diabetes (T2D) as a proof-of-concept. MOSAIC was evaluated on a synthetic cohort (SyntheticMass; open-weight N = 4,886; closed-weight N = 200) against three algorithmic ground truths (DCSI, DiSSCo, Cooper) and against all

Lineage graph

Paper → model → repo connections mined from source citations (Tier-1 exact match).

Why these links exist

Linked via arxiv authorManuela Del Castillo Suero →
Multi-Large Language Model Orchestrated Severity Assessment of Clinical Records (MOSAIC)
Linked via arxiv authorArnault-Quentin Vermillet →
Multi-Large Language Model Orchestrated Severity Assessment of Clinical Records (MOSAIC)
Linked via arxiv authorNicole Sonne Heckmann →
Multi-Large Language Model Orchestrated Severity Assessment of Clinical Records (MOSAIC)
Linked via arxiv authorDarmendra Ramcharran →
Multi-Large Language Model Orchestrated Severity Assessment of Clinical Records (MOSAIC)
Linked via arxiv authorMaurizio Sessa →
Multi-Large Language Model Orchestrated Severity Assessment of Clinical Records (MOSAIC)

Covers

newsTowards AI-augmented decision making in psychiatry newsClinical drug report generation using multi-phase prompt large language models - Nature newsCo-pilot, Not Autopilot: A Practical Method for Using Large Language Models in Interventional Cardiology - EMJ

Implements

repotyang816/Awesome-TCM-LLM repoEmo-gml/Awesome-Mental-Health-LLMs

Covers (incoming)

newsLarge language models exhibit stigmatizing behaviour in contextual judgements of health conditions - Nature newsClinician Use of a General-Purpose Large Language Model in Hospital Medicine: A Mixed-Methods Pilot Study - Cureus newsBenchmarking large language models against practicing clinicians on psychopathological assessment - Nature newsSmall language models in medicine - Nature

authored (incoming)

personManuela Del Castillo Suero personArnault-Quentin Vermillet personNicole Sonne Heckmann personDarmendra Ramcharran personMaurizio Sessa

Related across the graph

newsClinician Use of a General-Purpose Large Language Model in Hospital Medicine: A Mixed-Methods Pilot Study - Cureus personManuela Del Castillo Suero personNicole Sonne Heckmann newsClinical drug report generation using multi-phase prompt large language models - Nature newsBenchmarking large language models against practicing clinicians on psychopathological assessment - Nature newsLarge language models exhibit stigmatizing behaviour in contextual judgements of health conditions - Nature newsTowards AI-augmented decision making in psychiatry personMaurizio Sessa repotyang816/Awesome-TCM-LLM repoEmo-gml/Awesome-Mental-Health-LLMs newsSmall language models in medicine - Nature personArnault-Quentin Vermillet personDarmendra Ramcharran newsCo-pilot, Not Autopilot: A Practical Method for Using Large Language Models in Interventional Cardiology - EMJ

Topics

cs.CL