Neural Networks Based Smart E-Health Application for the Prediction of Tuberculosis Using Serverless Computing

Area of Science:

• Artificial Intelligence in Healthcare
• Medical Informatics
• Computational Biology

Background:

• The integration of Internet of Things (IoT) and electronic health records (e-health) is transforming healthcare diagnostics and treatment.
• Predicting diseases like Tuberculosis (TB) early is crucial for effective management and improved patient outcomes.

Purpose of the Study:

• To propose a neural network-based smart e-health application for Tuberculosis (TB) prediction using serverless computing.
• To evaluate the performance of various Convolutional Neural Network (CNN) architectures for TB detection in lung images.
• To compare server and serverless deployment strategies for machine learning models in healthcare.

Main Methods:

• Trained and validated Densenet-201, VGG-19, and Mobilenet-V3-Small CNN architectures using transfer learning.
• Evaluated model performance using metrics like accuracy, loss, intersection over union, precision, recall, and F1 score.
• Assessed deployment performance using JMeter to measure response rate, throughput, and error rate in server and serverless environments.

Main Results:

• The VGG-19 architecture achieved the best performance among the evaluated CNN models for TB prediction.
• Serverless deployment strategies showed promising results for integrating ML models into e-health systems.
• Performance metrics indicated the feasibility of using CNNs for early TB detection within IoT-enabled healthcare frameworks.

Conclusions:

• Machine learning models, particularly VGG-19, hold significant promise for enhancing IoT and e-health systems in predicting Tuberculosis.
• Serverless computing offers an efficient deployment option for AI-driven healthcare applications.
• This research provides a foundation for developing data-driven, smart healthcare solutions for improved disease diagnosis and treatment.