COVID-19 diagnosis system by deep learning approaches.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computer-Aided Diagnosis

Background:

• The COVID-19 pandemic necessitates rapid and effective diagnostic tools.
• Current diagnostic methods for COVID-19 are limited in speed and accessibility.
• Computerized analysis of chest X-rays and CT scans offers a potential solution for immediate detection.

Purpose of the Study:

• To develop and evaluate a deep learning model for detecting COVID-19 from chest X-rays.
• To implement automated segmentation of infected lung regions.
• To classify patients as COVID-19 positive or negative using deep learning.

Main Methods:

• Utilized regional deep learning approaches for lung infection detection.
• Employed a deep Convolutional Neural Network (CNN) with a full-resolution convolutional network (FrCN) for segmentation.
• Trained and tested the model on a COVID-19 patient dataset using fourfold cross-validation.

Main Results:

• The model demonstrated effectiveness in detection, segmentation, and classification of COVID-19 cases.
• Performance was evaluated using metrics like Sensitivity, Specificity, Jaccard index, Dice coefficient, MCC, and Overall accuracy.
• Comparative analysis showed classification accuracy with and without mass segmentation.

Conclusions:

• The proposed deep learning model shows promise for rapid and accurate COVID-19 diagnosis using chest X-rays.
• Automated segmentation and classification can significantly aid in managing the COVID-19 pandemic.
• This approach offers a viable alternative to traditional diagnostic techniques for mass screening.