Effective diagnosis and treatment planning depend on the precise segmentation and volume calculation of kidney cancers in computed tomography (CT) images. A deep learning-based framework created especially for the automated segmentation of kidney tumors and the ensuing volume estimation is examined in this paper. Convolutional neural networks (CNNs) with numerous layers tuned for picture segmentation tasks are used in the suggested model. The design incorporates a number of preprocessing methods to improve image quality, including normalization and noise reduction, which help the model better identify tumor boundaries. After segmentation, the tumor volume is calculated using a three-dimensional reconstruction methodology, which aids in assessing the tumor’s growth and response to therapy. Through a comparison with existing techniques, this study demonstrates the advantages of deep learning in achieving higher segmentation accuracy. After segmentation, the tumor volume is calculated using a three-dimensional reconstruction methodology, which aids in assessing the tumor’s growth and response to therapy. Through comparison with existing segmentation techniques, this study demonstrates the advantages of deep learning in achieving higher accuracy and robustness. By offering a reliable, automated technique for evaluating kidney cancers on CT images, the results demonstrate the model’s potential in clinical settings.