loading page

Machine Learning with environmental predictors to forecast Hospital Attendances and In-hospital Mortality: a Systematic Review
  • +1
  • Levi Monteiro Martins ,
  • Elsa Coz ,
  • Delphine Maucort-Boulch ,
  • Mohand-Saïd Hacid
Levi Monteiro Martins
Université Claude Bernard Lyon 1

Corresponding Author:[email protected]

Author Profile
Delphine Maucort-Boulch
Author Profile
Mohand-Saïd Hacid
Author Profile

Abstract

Many studies have demonstrated a correlation between environmental predictors and hospital attendance, particularly air pollution, a significant risk to health. One important factor that affects air pollution concentration is meteorology. Both factors contribute to an increase in the healthcare demand, which can lead to the global problem of Emergency Department Crowding. To address this problem, forecasting models are essential in resources allocation and general management to improve patient outcomes. Machine Learning (ML), especially Deep Learning (DL), techniques offer considerable promise for patient volume forecasting of hospital visits and in-hospital mortality. In this work, we present a systematic review of the use of ML to predict patient volume hospital attendance and in-hospital mortality with the use of environmental predictors. We focus on answering how have ML been applied and what are the major environmental predictors. Using the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA), 21 studies were retained  from three databases PubMed, Embase, and IEEE Explore. The search included studies from 2012 to 2022. Data  extraction and Risk of Bias assessment were done by using standardized tools Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies (CHARMS), and Prediction model Risk Of Bias ASsessment Tool (PROBAST). The study protocol for this review was registered on PROSPERO, assigned registration code CRD42023390400.