Saranya, R and Kavitha, R and Dhivya, S (2026) A Review of Machine Learning Approaches for Respiratory Health Risk Prediction. A Review of Machine Learning Approaches for Respiratory Health Risk Prediction, 12 (9). pp. 210-216. ISSN 2349-6002

[thumbnail of DrR.Saranya.pdf] Text
DrR.Saranya.pdf - Published Version
Download (311kB)

Abstract

respiratory diseases such as asthma, chronic
obstructive pulmonary disease, and acute respiratory
distress syndrome represent a major global health
burden and require timely risk assessment to prevent
adverse outcomes. The increasing availability of
electronic health records, medical imaging, wearable
sensor data, and environmental information has
accelerated the application of machine learning
techniques for respiratory health risk prediction.
Machine learning models offer the ability to analyze
complex, high-dimensional data and capture nonlinear
relationships that are difficult to model using traditional
statistical approaches. This paper presents a
comprehensive survey of machine learning approaches
applied to respiratory health risk prediction. The review
systematically examines commonly used data sources,
feature engineering and selection strategies, predictive
modeling techniques, evaluation metrics, and validation
methodologies. Both traditional machine learning
models and advanced deep learning architectures are
analyzed with respect to their suitability for structured
clinical data, imaging modalities, and longitudinal
physiological signals. A comparative discussion
highlights the strengths and limitations of ensemble
learning and deep learning approaches, emphasizing the
trade-offs between predictive performance,
interpretability, and clinical usability. In addition, this
survey identifies key challenges related to data quality,
model generalizability, interpretability, and clinical
integration that currently limit real-world deployment.
Finally, emerging research directions, including
explainable artificial intelligence, multimodal data
fusion, federated learning, and prospective validation,
are discussed to guide future developments. This survey
aims to provide researchers and practitioners with a
consolidated understanding of current progress and open
research issues in machine learning-based respiratory
health risk prediction

Item Type: Article
Uncontrolled Keywords: Machine Learning, Respiratory Health Risk Prediction, Clinical Decision Support Systems, Deep Learning, Predictive Analytics
Divisions: PSG College of Arts and Science > Department of Computer Science
Depositing User: Dr. B Sivakumar
Date Deposited: 16 Apr 2026 07:47
Last Modified: 16 Apr 2026 07:47
URI: https://ir.psgcas.ac.in/id/eprint/2794

Actions (login required)

View Item
View Item