# Electronic Health Record

Natural-language data requests in clinical research used to require staff to translate them into computable rules, pick data sources and write institution-specific queries. A large-language-model multi-agent pipeline decomposes each investigator request into source-specific tasks, generates and executes SQL, then combines structured records with clinical-note adjudication. Of 50 pediatric orthopedic requests, 43 completed manual review, with adjusted median positive predictive value 0.93, sensitivity within the observed union 0.92 and F1 0.90.

Natural-language data requests in clinical research used to require staff to translate them into computable rules, pick data sources and write institution-specific queries. A large-language-model multi-agent pipeline decomposes each investigator request into source-specific tasks, generates and executes SQL, then combines structured records with clinical-note adjudication. Of 50 pediatric orthopedic requests, 43 completed manual review, with adjusted median positive predictive value 0.93, sensitivity within the observed union 0.92 and F1 0.90.

Most ICU sepsis warning models answer only whether sepsis will occur, leaving clinicians unsure of the onset context and early course once an alarm sounds. SepsisFlow places hourly pre-onset risk, the patient state at onset and the 72-hour organ-support trajectory on one clock advancing hourly from ICU admission: before onset it uses only bedside-visible records without antimicrobial or culture orders, reaching a 24-hour AUROC of 0.822 on the MIMIC-IV internal test set, and its onset states form three contexts with 28-day in-hospital mortality of 17.1%, 15.1% and 2.2%.

Most ICU sepsis warning models answer only whether sepsis will occur, leaving clinicians unsure of the onset context and early course once an alarm sounds. SepsisFlow places hourly pre-onset risk, the patient state at onset and the 72-hour organ-support trajectory on one clock advancing hourly from ICU admission: before onset it uses only bedside-visible records without antimicrobial or culture orders, reaching a 24-hour AUROC of 0.822 on the MIMIC-IV internal test set, and its onset states form three contexts with 28-day in-hospital mortality of 17.1%, 15.1% and 2.2%.

FERMAT, a 65-million-parameter clinical foundation model pretrained on records from 3.77 million patients, was compared against itself on the same patients: patient representations predicted five-year kidney, fatty liver, and diabetes risk better, while generated trajectories predicted the diagnoses, prescriptions, and procedures recorded together on the next clinical date better.

FERMAT, a 65-million-parameter clinical foundation model pretrained on records from 3.77 million patients, was compared against itself on the same patients: patient representations predicted five-year kidney, fatty liver, and diabetes risk better, while generated trajectories predicted the diagnoses, prescriptions, and procedures recorded together on the next clinical date better.

A multi-agent workflow under investigator direction carried a large electronic health record study from cohort construction and executable statistics through data-quality checks, literature retrieval and reporting, using sequential sacituzumab govitecan (SG) and trastuzumab deruxtecan (T-DXd) orders in breast cancer as the clinical case. From 4.95 million patients, 106 had orders for both drugs, 73 with T-DXd first and 33 with SG first; median second-drug record spans were 63 and 49 days, a median difference of −14 days (95% CI, −41 to 21), descriptive and unadjusted.

A multi-agent workflow under investigator direction carried a large electronic health record study from cohort construction and executable statistics through data-quality checks, literature retrieval and reporting, using sequential sacituzumab govitecan (SG) and trastuzumab deruxtecan (T-DXd) orders in breast cancer as the clinical case. From 4.95 million patients, 106 had orders for both drugs, 73 with T-DXd first and 33 with SG first; median second-drug record spans were 63 and 49 days, a median difference of −14 days (95% CI, −41 to 21), descriptive and unadjusted.

In multi-source learning the deployment population's source composition is unknown, and optimizing against the worst source mixture can be overly conservative. PILLAR regularizes adversarial source weights toward a user-specified reference distribution while learning a shared low-rank matrix, solving the resulting nonconvex-concave problem by bifactored gradient descent with mirror ascent, with linear convergence and an error bound that separates low-rank estimation from adversarial reweighting. Simulations and an electronic health record study of suicide-attempt risk show improved worst-source and transported predictive accuracy.

In multi-source learning the deployment population's source composition is unknown, and optimizing against the worst source mixture can be overly conservative. PILLAR regularizes adversarial source weights toward a user-specified reference distribution while learning a shared low-rank matrix, solving the resulting nonconvex-concave problem by bifactored gradient descent with mirror ascent, with linear convergence and an error bound that separates low-rank estimation from adversarial reweighting. Simulations and an electronic health record study of suicide-attempt risk show improved worst-source and transported predictive accuracy.

This study deployed the 3.8B-parameter open-weight small language model Phi4-mini (via Ollama) within an institutional firewall, combined with deterministic regular expression post-processing, to extract six uncoded urinary tract infection clinical features (fever, nitrites, leukocytes, lumbar pain, abdominal pain, and haematuria) from 15,498 Catalan/Spanish bilingual MEAP primary care narratives in the SIDIAP database in Catalonia, achieving 93.8% accuracy, 96.6% specificity, and 83.6% sensitivity in a double-blind clinician gold standard validation of 60 real patient records, and 87.2% accuracy, 99.5% specificity, and 99.2% positive predictive value in adversarial synthetic stress-testing of 720 notes, with no patient data leaving institutional servers.

This study deployed the 3.8B-parameter open-weight small language model Phi4-mini (via Ollama) within an institutional firewall, combined with deterministic regular expression post-processing, to extract six uncoded urinary tract infection clinical features (fever, nitrites, leukocytes, lumbar pain, abdominal pain, and haematuria) from 15,498 Catalan/Spanish bilingual MEAP primary care narratives in the SIDIAP database in Catalonia, achieving 93.8% accuracy, 96.6% specificity, and 83.6% sensitivity in a double-blind clinician gold standard validation of 60 real patient records, and 87.2% accuracy, 99.5% specificity, and 99.2% positive predictive value in adversarial synthetic stress-testing of 720 notes, with no patient data leaving institutional servers.

EviGen proposes a three-layer framework that first retrieves and ranks evidence from a patient's longitudinal EHR by predictive contribution rather than textual relevance using learnable queries trained on outcome labels, then has an LLM generate a citation-grounded clinical rationale within that evidence scaffold, and finally applies a process-supervised verifier to check each reasoning step and flag unreliable claims; across MIMIC-IV one-year mortality, autism, and ADHD tasks it improves prediction performance and rationale faithfulness over full-context and RAG baselines and is preferred by most reviewers in a clinical pilot.

EviGen proposes a three-layer framework that first retrieves and ranks evidence from a patient's longitudinal EHR by predictive contribution rather than textual relevance using learnable queries trained on outcome labels, then has an LLM generate a citation-grounded clinical rationale within that evidence scaffold, and finally applies a process-supervised verifier to check each reasoning step and flag unreliable claims; across MIMIC-IV one-year mortality, autism, and ADHD tasks it improves prediction performance and rationale faithfulness over full-context and RAG baselines and is preferred by most reviewers in a clinical pilot.

This systematic review searched PubMed, MEDLINE, Embase, PsycINFO, and Web of Science from inception to November 11, 2025, and included 29 studies that developed, validated, or evaluated multivariable prediction models using routinely collected electronic health record or administrative data to predict acute mental status deterioration during adult hospital admissions, all operationalized as delirium; using CHARMS and TRIPOD/TRIPOD-AI for data extraction and PROBAST for risk of bias, it found that the evidence clustered into four overlapping prediction tasks (admission or early-stay risk stratification, perioperative or postoperative prediction, dynamic intensive care unit prediction, and external validation or workflow evaluation of existing tools), that most studies were retrospective co

This systematic review searched PubMed, MEDLINE, Embase, PsycINFO, and Web of Science from inception to November 11, 2025, and included 29 studies that developed, validated, or evaluated multivariable prediction models using routinely collected electronic health record or administrative data to predict acute mental status deterioration during adult hospital admissions, all operationalized as delirium; using CHARMS and TRIPOD/TRIPOD-AI for data extraction and PROBAST for risk of bias, it found that the evidence clustered into four overlapping prediction tasks (admission or early-stay risk stratification, perioperative or postoperative prediction, dynamic intensive care unit prediction, and external validation or workflow evaluation of existing tools), that most studies were retrospective co