JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

A Deep Auto-Optimized Collaborative Learning (DACL) model for disease prognosis using AI-IoMT systems

  • 0Department of CSE, School of Computing, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Chennai, Tamil Nadu, 600062, India.
Scientific Reports +

|

May 4, 2024

|

May 4, 2024

View abstract on PubMed

Summary

This summary is machine-generated.

This study introduces a new AI-IoMT framework, the Deep Auto-Optimized Collaborative Learning (DACL) Model, for diagnosing chronic diseases like heart disease and diabetes using patient data. The framework enhances data accuracy and optimizes feature selection for precise disease identification.

Area Of Science

  • Medical Informatics
  • Artificial Intelligence in Healthcare
  • Internet of Medical Things (IoMT)

Background

  • Integrating Artificial Intelligence (AI) and the Internet of Medical Things (IoMT) offers significant advantages in modern healthcare for disease management.
  • Wearable sensors and interconnected networks are crucial for real-time health monitoring and disease control.
  • Accurate identification of chronic diseases from patient medical records remains a critical challenge.

Purpose Of The Study

  • To develop an advanced AI-IoMT framework for the accurate identification and classification of multiple chronic diseases.
  • To enhance the diagnostic process for conditions such as heart disease, diabetes, and stroke using patient medical records.
  • To create a robust system capable of handling missing data and optimizing feature selection for improved diagnostic accuracy.

Main Methods

  • Development of the Deep Auto-Optimized Collaborative Learning (DACL) Model, an novel AI-IoMT framework.
  • Utilization of a Deep Auto-Encoder Model (DAEM) for data imputation and preprocessing.
  • Application of the Golden Flower Search (GFS) algorithm for optimal feature selection.
  • Implementation of the Collaborative Bias Integrated GAN (ColBGaN) model for disease classification.
  • Optimization of the classification loss function using the Water Drop Optimization (WDO) technique.

Main Results

  • The proposed DACL framework demonstrates effectiveness and efficiency in identifying chronic diseases.
  • The integrated models (DAEM, GFS, ColBGaN, WDO) contribute to rapid and precise disease diagnosis.
  • Performance evaluation using benchmarking datasets and standard metrics validates the framework's capabilities.

Conclusions

  • The developed AI-IoMT framework, DACL, shows significant potential for improving the early detection and management of chronic diseases.
  • The study highlights the synergy between AI and IoMT in creating intelligent healthcare solutions.
  • Further validation and implementation of the DACL model could lead to enhanced patient outcomes in chronic disease care.

Keywords: