MétaCan
Menu
← Back to cohort
Record W7025445614

Virtual Medical Assistant Model Using Natural Language Processing in Healthcare System

2024· other· fr· W7025445614 on OpenAlexfundno aff

Bibliographic record

VenuePolyPublie (École Polytechnique de Montréal) · 2024
Typeother
Languagefr
Field
Topic
Canadian institutionsnot available
FundersPolytechnique Montréal
KeywordsDomain (mathematical analysis)Digital humanitiesContext (archaeology)Healthcare system
DOInot available

Abstract

fetched live from OpenAlex

RÉSUMÉ: L'importance des soins de santé dans nos vies est indéniable, et les progrès rapides continuent de transformer ce domaine quotidiennement. Un domaine clé dans cette évolution est l'intégration de l'intelligence artificielle et de l'apprentissage automatique pour améliorer les services de santé. De nos jours, les systèmes de santé peuvent utiliser ces modèles basés sur l'intelligence artificielle pour répondre aux besoins des patients, améliorer l'accessibilité à l'information médicale et accroître l'efficacité opérationnelle des prestataires de soins de santé. Cependant, en parallèle de ces avancées, de nouveaux défis sont apparus, notamment en matière de confidentialité des données, de précision des modèles, d’évolutivité et d'adaptabilité aux scénarios du monde réel. Dans ce mémoire, nous nous concentrerons sur le développement d'un système en deux parties pour améliorer la prestation des services de santé. Nous mettrons en oeuvre un modèle de mémoire à long terme bidirectionnelle (BiLSTM) pour la détection des symptômes, en utilisant l'analyse de texte pour fournir des recommandations de services médicaux précises, et un modèle KNN pour localiser les centres médicaux les plus proches à l'aide de données géographiques telles que les codes postaux et les villes. La performance des deux modèles sera évaluée à l'aide des courbes caractéristiques de fonctionnement du récepteur (ROC) afin de garantir une précision de prédiction élevée. Les modèles seront implémentés sur un MacBook Pro fonctionnant sous macOS Monterey. Les résultats devraient montrer des améliorations prometteuses en termes d'efficacité et de précision. Les résultats de notre modèle proposé démontrent une grande précision et efficacité. En conclusion, la mise en oeuvre réussie de ces modèles améliore considérablement l'efficacité de notre système de soutien de santé, validant son potentiel à améliorer l'accès aux services médicaux et à fournir une assistance rapide basée sur l'identification des symptômes. ABSTRACT: The importance of healthcare in our lives is undeniable, and rapid advancements continue to transform this field daily. A key area within this evolution is the integration of artificial intelligence and machine learning to enhance healthcare services. Today, healthcare systems can use these AIdriven models to address patient needs, improve accessibility to medical information, and increase the operational efficiency of healthcare providers. However, along-side these advancements, new challenges have arisen, including data privacy, model accuracy, scalability, and adaptability in realworld scenarios. In this dissertation, we will focus on developing a two-part system to improve healthcare service delivery. We will implement a Bidirectional Long Short-Term Memory (BiLSTM) model for symptom detection, using text analysis to provide accurate medical service recommendations, and a K-Nearest Neighbors (KNN) model to locate the nearest medical centers using geographical data such as postal codes and cities. The performance of both models will be evaluated using Receiveroperating characteristic curves (ROC), to ensure high predictive accuracy. The models will be implemented on a MacBook Pro running macOS Monterey. The results are expected to show promising improvements in terms of efficiency and precision. The results of our proposed model demonstrate high accuracy (i.e., 98.61% for the training set, 98.55% for the validation set and 98.50% for the test set). In conclusion, the successful implementation of these models significantly enhances the efficiency of our healthcare support system, validating its potential to improve access to medical services and provide timely assistance based on symptom identification.

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)0.001
metaresearch head score (Gemma)0.002
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: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.013
Threshold uncertainty score0.025

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.002
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.001
Bibliometrics0.0010.000
Science and technology studies0.0000.000
Scholarly communication0.0020.001
Open science0.0010.001
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0060.001

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 head0.013
GPT teacher head0.275
Teacher spread0.262 · 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
GenreEmpirical

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

Citations0
Published2024
Admission routes1
Has abstractyes

Explore more

Same venuePolyPublie (École Polytechnique de Montréal)→French-language works237,207→