Enhancing Performance in Detection Of Pneumonia with Deep Learning

CH.Vasundhara1, Vishnu Surya Gedala2

1Assistant Professor, 2 MCA Final Semester,

Master of Computer Applications, Sanketika Vidya Parishad Engineering College, Vishakhapatnam,

Andhra Pradesh, India

vishnusurya132@gmail.com

ABSTRACT:

Pneumonia is a dangerous infection that affects one or both lungs. It fills the tiny air sacs (alveoli) with fluid or pus, making it hard to breathe. According to the World Health Organization (WHO), pneumonia causes 1 in every 3 deaths in India. By 2030, it may lead to the deaths of around 11 million children under 5 if not treated properly. Doctors usually check for pneumonia using chest X-rays, but reading them needs trained experts. In many places, this leads to delays or mistakes in diagnosis. In recent years, the field of Artificial Intelligence (AI) and Machine Learning has shown great promise in automating medical image analysis.  But, the Machine Learning techniques did not meet the accuracy as needed. To solve this, the project uses a deep learning technique called Convolutional Neural Network (CNN) to check a chest X-ray and tell if the person has pneumonia. The system is build using free software tools like Python and OpenCV and Google Colab. This project results an accuracy of over 91%, confirm the robustness and reliability of the proposed model. Additionally, the proposed CNN model helps in reducing the workload of radiologists by automatically analyzing chest X-ray images with high efficiency. The system can provide faster diagnosis, enabling timely treatment and improving patient outcomes. Data preprocessing and image enhancement techniques are applied to improve the quality of X-ray images before classification. The developed model is cost-effective, user-friendly, and can be deployed in healthcare centers with limited medical resources.