ANN Assisted-IoT Enabled COVID-19 Patient Monitoring.

Area of Science:

• Healthcare Informatics
• Computational Biology
• Machine Learning Applications

Background:

• COVID-19's high infectivity necessitates rapid identification and clinical support.
• Machine Learning (ML) and IoT are explored for categorizing COVID-19 patients.
• Artificial Neural Networks (ANN) are recognized for their utility in healthcare decision-making.

Purpose of the Study:

• To illustrate the applicability and suitability of ANN for classifying COVID-19 patient health states.
• To categorize patients into four distinct health statuses: infected (IN), uninfected (UI), exposed (EP), and susceptible (ST).

Main Methods:

• Utilized Bayesian and back propagation algorithms for generating results within the ANN framework.
• Employed the Viterbi algorithm to enhance the accuracy of the proposed classification system.
• Validated the ANN mechanism against conventional methods like Random Tree (RT), Fuzzy C Means (FCM), and REPTree (RPT) using various performance metrics.

Main Results:

• The proposed ANN system demonstrated effectiveness in categorizing COVID-19 patient health statuses.
• The Viterbi algorithm integration led to improved accuracy in the classification outcomes.
• Comparative analysis showed the ANN approach's performance against established ML algorithms.

Conclusions:

• ANN provides a viable and accurate method for the classification of COVID-19 patient health statuses.
• The integration of algorithms like Viterbi can further optimize the diagnostic capabilities of ML models in healthcare.
• This research supports the use of advanced computational techniques for immediate and effective COVID-19 patient management.