IoT enabled depthwise separable convolution neural network with deep support vector machine for COVID-19 diagnosis...

Area of Science:

• Artificial Intelligence
• Medical Imaging
• Internet of Things

Background:

• The COVID-19 pandemic necessitates rapid and accurate diagnostic tools.
• Radiological imaging (X-ray, CT) is crucial for COVID-19 diagnosis.
• Existing diagnostic methods require enhancement for efficiency and accuracy.

Purpose of the Study:

• To develop and evaluate a novel intelligent model for COVID-19 diagnosis and classification.
• To integrate Internet of Things (IoT) for efficient data acquisition in healthcare.
• To leverage advanced deep learning and machine learning techniques for improved diagnostic performance.

Main Methods:

• Utilized an IoT-enabled Depthwise Separable Convolutional Neural Network (DWS-CNN) integrated with a Deep Support Vector Machine (DSVM).
• Implemented a workflow including data acquisition via IoT devices, Gaussian filtering (GF) for preprocessing, automatic feature extraction using DWS-CNN, and classification with DSVM.
• Tested the model on a Chest X-ray (CXR) image dataset for binary and multiclass classification.

Main Results:

• The proposed DWS-CNN with DSVM model achieved high accuracy in diagnosing COVID-19.
• Achieved 98.54% accuracy for binary classification and 99.06% accuracy for multiclass classification.
• Demonstrated superior performance compared to existing methods in classifying COVID-19 from CXR images.

Conclusions:

• The developed IoT-enabled DWS-CNN with DSVM model is effective for accurate COVID-19 diagnosis and classification.
• The integration of IoT and advanced AI techniques shows significant promise for future healthcare applications.
• This approach offers a robust solution for rapid and reliable disease detection in medical imaging.