MétaCan
Menu
Back to cohort
Record W4387007762 · doi:10.1101/2023.09.24.23295960

Moving Biosurveillance Beyond Coded Data: AI for Symptom Detection from Physician Notes

2023· preprint· en· W4387007762 on OpenAlexaff
Andrew McMurry, Amy R. Zipursky, Alon Geva, Karen L. Olson, James Jones, Vladimir Ignatov, Timothy A. Miller, Kenneth D. Mandl

Bibliographic record

VenuemedRxiv · 2023
Typepreprint
Languageen
FieldMedicine
TopicCOVID-19 diagnosis using AI
Canadian institutionsHospital for Sick Children
FundersNational Institute of Child Health and Human DevelopmentCenters for Disease Control and PreventionU.S. Department of Health and Human Services
KeywordsRecallEmergency departmentMedicineGold standard (test)Artificial intelligenceElectronic health recordHealth recordsPopulationRetrospective cohort studyCoronavirus disease 2019 (COVID-19)CohortNatural language processingPediatricsComputer scienceHealth carePsychologyDiseasePsychiatryInternal medicineInfectious disease (medical specialty)

Abstract

fetched live from OpenAlex

Abstract Background Real-time surveillance of emerging infectious diseases necessitates a dynamically evolving, computable case definition, which frequently incorporates symptom-related criteria. For symptom detection, both population health monitoring platforms and research initiatives primarily depend on structured data extracted from electronic health records. Objective To validate and test an artificial intelligence (AI) based Natural Language Processing (NLP) pipeline for detecting COVID-19 symptoms from physician notes. Methods Subjects in this retrospective cohort study are patients 21 years old and younger, who presented to a pediatric emergency department (ED) at a large academic children’s hospital between March 1, 2020 and May 31, 2022. ED notes for all patients were processed with an NLP pipeline tuned to detect the mention of 11 COVID-19 symptoms based on CDC criteria. For a gold standard, 3 subject matter experts labeled 226 ED notes and had strong agreement (F1=98.6; PPV=97.2; Recall=100.0). F1, PPV, and recall were used to compare the performance of both NLP and ICD-10 to the gold standard chart review. As a formative use case, variations in symptom patterns were measured across SARS-Cov2 variant eras. Results There were 85,678 ED encounters during the study period, 4.0% with patients with COVID-19. NLP was more accurate at identifying encounters with patients that had any of the COVID-19 symptoms (F1=79.6) than ICD-10 codes (F1=45.1%). NLP accuracy was higher for positive symptoms (recall=93%) than ICD-10 (recall=30%). However, ICD-10 accuracy was higher for negative symptoms (specificity=99.4%) than NLP (specificity=91.7%). Congestion or runny nose showed the highest accuracy difference: NLP F1=82.8%, ICD-10 F1=4.2%. Prevalence of NLP symptoms among patients with COVID-19 differed across variant eras. And patients with COVID-19 were more likely to have each symptom than patients without this disease. Effect sizes (odds ratios) varied across pandemic eras. Conclusions This study establishes the value of AI based NLP as a highly effective tool for real-time COVID-19 symptom detection in pediatric patients, outperforming traditional ICD-10 methods. It also reveals the evolving nature of symptom prevalence across different virus variants, underscoring the need for dynamic, technology-driven approaches in infectious disease surveillance.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.006
metaresearch head score (Gemma)0.033
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.009
Threshold uncertainty score0.032

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0060.033
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0040.002
Science and technology studies0.0010.001
Scholarly communication0.0030.002
Open science0.0010.002
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0020.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.073
GPT teacher head0.354
Teacher spread0.281 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations1
Published2023
Admission routes1
Has abstractyes

Explore more

Same venuemedRxivSame topicCOVID-19 diagnosis using AIFrench-language works237,207