Detecting Lung Nodule Malignancy Using CNN-Based Deep Learning Techniques

This study focuses on the development of a Convolutional Neural Network (CNN)-based deep learning model to detect cancerous lung nodules from computed-tomography (CT) scans. Timely identification of lung cancer is essential in improving survival rates, as it remains a leading cause of global cancer-related mortality. CNN models improve medical image analysis by automating the classification process to offer increased diagnostic accuracy. A custom-built CNN model is employed to distinguish between benign and malignant lung nodules using CT scans from the Kaggle's Data Science Bowl 2017 dataset. Our model reached a 92% accuracy in data classification. The findings show that this CNN-driven approach excels previous techniques in identifying lung nodule malignancy, allowing more precise diagnoses, and improved patient outcomes. Future research should explore using cascaded algorithms and leveraging multiple datasets to enhance prediction accuracy and generalization of malignancy, as well as reduce overfitting impact.