Corresponding Author

Nada M. Yakout

Subject Area

Biomedical Engineering

Article Type

Original Study


Classification of tumors among kidney masses (KM) through deep neural networks is one of the most important applications for detecting the disease at an early stage. It can increase patient survival rates, prevent mass growth, and prevent complications. Preoperative multi-phase abdominal Contrast Enhancement Computed Tomography (CE-CT) is widely used to detect lesions of kidney masses in order to avoid unneeded biopsy or surgery. However, decisions about treatment are difficult because of inter-observer variability caused by minute variations in the imaging characteristics of mass sub-types. In this study, we offer a comprehensive deep learning model using a Convolutional Neural Networks(CNN) for the differential diagnosis of five significant kidney masses with histologic subtypes, covering both benign and malignant tumours. The suggested model was trained and tested on 99 patients with overall testing accuracy, losses, F1-score, and AUC of 99.86%, 0.005, 99,86%, 1 by the given order in the provided results, which confirmed the excellent accuracy.


Kidney Masses(KM); Contrast Enhancement Computed Tomography(CE-CT); Convolution Neural Network(CNN); Deep Learning(DL).

Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.