Article citationsMore>>

Timotijevic, L., Hodgkins, C.E., Banks, A., Rusconi, P., Egan, B., Peacock, M., et al. (2020) Designing a m-Health Clinical Decision Support System for Parkinson’s Disease: A Theoretically Grounded User Needs Approach. BMC Medical Informatics and Decision Making, 20, Article No. 34.
https://doi.org/10.1186/s12911-020-1027-1

has been cited by the following article:

• TITLE: Analysis of Decision Support Systems (DSS) Integration with AI Using Mobile Health (m-Health) and Wearables

  AUTHORS: Aisha Alqarni

  KEYWORDS: Artificial Intelligence (AI), Decision Support Systems (DSS), Wearables, Mobile Health (m-Health), Real-Time Monitoring

  JOURNAL NAME: Intelligent Information Management, Vol.17 No.6, November 13, 2025

  ABSTRACT: Smart health refers to the integration of cutting-edge technologies into healthcare systems to improve patient care and apply intelligent clinical decision-making. The study investigates how artificial intelligence (AI) can be integrated with decision support systems (DSS) in healthcare using wearable technologies and mobile health (m-Health) platforms. This research focuses on applying artificial intelligence to clinical decision support systems to improve healthcare delivery in real-time. A proposed model aims to implement machine learning algorithms to analyze continuous health data retrieved from wearables and mobile health applications. The research will evaluate the proposed model using two datasets m-Health [1] and Indicators of Heart Disease [2], available online in Kaggle repository. Random Forest, Gradient Boosting Machines, and XGBoost are used to assess the predictive performance within structured sensor data. Only statistically significant variables are retained for feature selection based on their correlation coefficients. Random Forest reports 94.23% accuracy on the m-Health dataset, which affirms the effectiveness in sensor-driven activity performance evaluation. Gradient Boosting achieves 91.26% accuracy on the structured medical dataset, reaffirming its credibility in health risk assessment. It is noted that embedding AI models into decision support systems will act as a framework for responsive, data-driven decisions in medicine. The capability to recognize anomalies and forecast possible health threats facilitates proactive and continual care for patients, thus supporting real-time monitoring. Uniformed data preprocessing on other dataset enhances the experimental validity and reliability. The study demonstrates that AI-empowered CDSS have great capabilities in closing the gaps of delays in diagnosis and the need for timely, tailored interventions. Future research will focus in incorporating real-time wearable data with adaptive AI decision support systems to improve clinical applicability and scalability.