Automated Kidney Tumor Detection from CT Images Using CNN-Based Deep Learning

Kidney cancer detection using CT scan images plays a vital role in early diagnosis and treatment planning, but manual interpretation remains a slow and error-prone process. To address this, we propose a deep learning-based system that leverages Convolutional Neural Networks (CNNs) for automated classification of kidney CT images into positive and negative cases. The model is trained using augmented CT scan images to improve robustness and accuracy, especially in detecting small lesions. The CNN architecture includes input, hidden, and output layers designed to extract meaningful spatial features from 64x64 pixel images. By integrating morphological convolutional layers and structured image preprocessing, the system enhances sensitivity to subtle lesion patterns. Training and testing are conducted using labeled datasets, and the model's performance is evaluated using metrics like accuracy, precision, recall, and a confusion matrix. This approach demonstrates a significant improvement in classification performance and offers a reliable, efficient solution for computer-aided kidney cancer diagnosis using only CNNs.