Diagnosis of Pneumonia using Image Recognition Techniques

• Published in: 2023 7th International Conference on Intelligent Computing and Control Systems (ICICCS) pp. 1332 - 1337
• Main Authors: Sangeetha, S., Suruthika, S., Keerthika, S., Vinitha, S., Sugunadevi, M.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 17-05-2023
• Subjects: Convolutional Neural Network (CNN) - ReLU; Deep learning; Forward and Backward propagation; Max-pooling; Medical treatment; Microorganisms; Neural networks; Pulmonary diseases; Sepsis; Snake algorithm; Support vector machines
• ISSN: 2768-5330
• DOI: 10.1109/ICICCS56967.2023.10142892

Pneumonia is a severe and fatal disease caused by viral or bacterial sepsis in human lungs. Early detection of pneumonia is critical for successful treatment. Deep learning algorithms that use many layers to interpret a boundary of data have obtained the best results in a variety of disciplines, particularly in the documenting and sorting of various diseases. As a result, there is an increasing need to design and develop a pneumonia detection system by using deep learning methods with an ability to evaluate chest CT-scan imageries and aid in the pneumonia exposure process. The proposed framework contains three phases: feature extraction, image classification, and image pre-processing. As a consequence, the projected Compared to other deep-learning models that have already been trained, such as (ResNet 50) and SVM, pneumonia diagnosis using the CNN model performed better in terms of accuracy and reliability. As a result, the efficient performance of the proposed Convolutional Neural Network centred pneumonia perception model in every parameter can provide valuable assistance to patients' care and reduce the human error rates.