The kidney plays a vital role in filtering waste and fluids but is vulnerable to abnormalities such as cysts, stones, and tumors, necessitating accurate and timely detection for effective medical intervention. This study introduces a systematic approach leveraging Deep Convolutional Neural Networks (DCNNs) to detect and classify kidney abnormalities in CT radiography. Advanced models, including ResNet-50, MobileNet, InceptionV3, and EfficientNetB0, are employed and evaluated on a robust dataset of 12,446 images categorized into cyst, stone, tumor, and normal conditions. The dataset undergoes meticulous preprocessing and is divided into training, validation, and testing sets to ensure reliable performance assessment. Among the models, InceptionV3 demonstrates superior accuracy at 99.759%, alongside remarkable precision, recall, and F1-score metrics, surpassing prior methodologies. This research highlights the efficacy of tailored CNN architectures for kidney pathology detection, offering a significant advancement in diagnostic accuracy and potential integration into clinical workflows. These findings underscore the importance of AI-driven medical imaging solutions in enhancing patient care and aligning with global health priorities.