# An Interpretable Machine Learning Framework for Analyzing the Interaction Between Cardiorespiratory Diseases and Meteo-Pollutant Sensor Data

## TL;DR

This paper introduces an interpretable machine learning framework to study how environmental factors affect emergency room admissions for cardiorespiratory diseases.

## Contribution

The novel contribution is an interpretable ML framework that identifies critical environmental thresholds linked to increased hospital admissions for cardiorespiratory diseases.

## Key findings

- XGBoost achieved high predictive accuracy (R2 = 0.901; MAE = 0.047) in modeling ER admissions for cardiorespiratory diseases.
- SHAP analysis revealed that carbon monoxide, humidity, low pressure, and mild temperatures are key environmental factors linked to increased CRD cases.
- LIME identified critical thresholds for CO, atmospheric pressure, temperature, and humidity that significantly increase hospital admission risk.

## Abstract

This study presents an approach based on machine learning (ML) techniques to analyze the relationship between emergency room (ER) admissions for cardiorespiratory diseases (CRDs) and environmental factors. The aim of this study is the development and verification of an interpretable machine learning framework applied to environmental and health data to assess the relationship between environmental factors and daily emergency room admissions for cardiorespiratory diseases. The model’s predictive accuracy was evaluated by comparing simulated values with observed historical data, thereby identifying the most influential environmental variables and critical exposure thresholds. This approach supports public health surveillance and healthcare resource management optimization. The health and environmental data, collected through meteorological sensors and air quality monitoring stations, cover eleven years (2013–2023), including meteorological conditions and atmospheric pollutants. Four ML models were compared, with XGBoost showing the best predictive performance (R2 = 0.901; MAE = 0.047). A 10-fold cross-validation was applied to improve reliability. Global model interpretability was assessed using SHAP, which highlighted that high levels of carbon monoxide and relative humidity, low atmospheric pressure, and mild temperatures are associated with an increase in CRD cases. The local analysis was further refined using LIME, whose application—followed by experimental verification—allowed for the identification of the critical thresholds beyond which a significant increase in the risk of hospital admission (above the 95th percentile) was observed: CO > 0.84 mg/m3, P_atm ≤ 1006.81 hPa, Tavg ≤ 17.19 °C, and RH > 70.33%. The findings emphasize the potential of interpretable ML models as tools for both epidemiological analysis and prevention support, offering a valuable framework for integrating environmental surveillance with healthcare planning.