Deep Learning-Based Hierarchical Delineation of Kidneys, Tumors, and Cysts in CT Images

Andrew Heschl, Hosein Beheshtifard, Phuong Thao Nguyen, Tapotosh Ghosh, Katie L. Ovens, Farhad Maleki

Abstract: Kidney cancer is among the most prevalent forms of cancer. Recent studies have investigated the use of machine learning, especially deep learning, in detecting tumors from kidney CT scans. This paper introduces a hierarchical pipeline for segmenting kidneys, tumors, and cysts in CT images. Our method demonstrates promising outcomes in the segmentation of kidneys, renal cysts, and tumors, showcasing its potential to accurately identify and localize tumors and cysts in CT scans. This can significantly aid early diagnosis and treatment planning.