Early Detection of Lung Cancer Using Deep Learning-Based Analysis of CT Scans

The advanced stage at which most occurrences of lung cancer (LC) are detected results in a dismal prognosis and few available therapeutic options for the disease. The ability to diagnose lung cancer at an early stage is an important clinical and scientific emphasis since it can greatly improve patient outcomes. Recent advances in deep learning have made it possible to use these methods effectively on medical images, which might improve the speed and precision with which lung cancer is diagnosed. Using deep learning to analyze computed tomography (CT) scans, this study introduces a unique method for the early diagnosis of lung cancer. The proposed method leverages a combination of two deep neural network architectures, UNET and LeNet, to achieve robust and accurate lung cancer detection. The UNET architecture is employed for semantic segmentation of lung regions from CT scans, allowing for precise localization and isolation of potential cancerous lesions. Subsequently, the LeNet architecture is applied to classify the segmented regions as either benign or malignant. This two-stage approach enables the model to identify lung cancer with high accuracy while minimizing false positives. Based on the results of this study, combining UNET and LeNet for CT scan analysis may result in an accuracy of 97.89%, providing doctors with a useful tool to enhance patient outcomes through prompt intervention.