HYBRID DEEP LEARNING MODEL WITH ENHANCED ACCURACY FOR PNEUMONIA

Computer Engineering

Pneumonia, a contagious respiratory illness, is often caused by a bacterial infection in the tiny air sacs known as alveoli within the lungs. When these lung tissues become infected, they accumulate pus. Medical professionals typically diagnose pneumonia through physical examinations and diagnostic tests such as Chest X-rays, ultrasounds, or lung biopsies. Misdiagnosis, incorrect treatment, or overlooking the disease can have severe consequences for the patient's health and quality of life. Recent advancements in deep learning have significantly aided healthcare professionals in the diagnostic process for such illnesses. This approach utilizes a flexible and efficient deep learning technique, specifically Convolutional Neural Networks (CNN), to predict and detect whether a patient is affected by the disease based on their chest X-ray images. In this study, a dataset containing 20,000 images with a resolution of 224x224 pixels and a batch size of 32 was used to evaluate the performance of the CNN model. During the training phase, the model achieved an impressive accuracy rate of 95%. The results of this experiment demonstrate that deep learning, particularly the CNN model, can effectively detect and predict various respiratory illnesses, including COVID-19, bacterial pneumonia, and viral pneumonia, based on chest X-ray images. This advancement in medical technology holds great promise for improving the accuracy and speed of diagnosis, ultimately benefiting patients and healthcare providers alike.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.