Identification of Anomalies in the Lungs by Developing a System Using Deep Learning

In the field of healthcare, early and accurate detection of lung abnormalities is essential for the diagnosis and effective treatment of various respiratory diseases. This study presents the development of a system for identifying abnormalities in the lungs based on deep learning techniques. Using a diversified dataset of radiological images, a deep convolutional neural network (CNN) was trained to recognize subtle patterns and anomalous features on patients’ Tomographys and computed tomography (CT) scans. The proposed system demonstrates a high accuracy with a Mean Average Precision (MAP) of 97.6 percent in detecting lung abnormalities, including nodules, infiltrates, masses, and other common pathologies. Additionally, a CNN interpretation functionality was implemented, allowing clinicians to visualize and understand the regions highlighted by the model, thereby improving confidence in clinical decisions.