Precision Tumor Identification in Lung CT Scans: Leveraging Convolutional Neural Networks for Accurate Segmentation

Lung cancer ranks as one of the most fatal types of cancer in the world, owing primarily to late detection. Machine learning advancements provide intriguing options for early diagnosis through automated classification of radiographic images. This study presents a Convolutional Neural Network (CNN) model for categorizing lung tumor CT scans into benign, malignant, & normal categories. We use a dataset with a variety of image sizes and qualities to provide full preprocessing, including resizing and Gaussian blurring, and solve class imbalance using the Synthetic Minority Over- sampling Technique (SMOTE). The model's efficacy is measured against a variety of cutting-edge approaches, with a particular emphasis on precision, recall, and accuracy as key performance metrics. The suggested CNN outperformed prior models, reaching 99% accuracy on the validation set and nearly flawless precision and recall across all classes. The findings also highlight possible clinical uses and open up new avenues for future study into integration with clinical workflows and larger, more diverse datasets.