Improved Classification for Corona Virus Disease using XceptionNet in X-Ray Images

Diagnosis is necessary to confirm SARS-CoV-2 infection. Imaging methods are vital in aiding the diagnosis, assessing disease severity, directing treatment, identifying complications, and monitoring the response to treatment. The lungs are the foremost organ impacted, and chest X-rays are the preferred initial imaging exam due to their accessibility and cost-effectiveness, whether obtained at conventional X-ray centers or using portable equipment. This research study has used a combination of virus-infected people as a dataset. The classification of diseases will help doctors to treat the patients and save their life. For classifying Normal, COVID-19 and Viral Pneumonia classes’ the architecture of the pre-trained network XceptionNet is modified. XceptionNet is the method to use for the classification of diseases. The proposed model has resulted in a 98.80% accuracy for classifying the diseases.