Kidney Disease X-Ray Image Classification Using a ResNet50V2 Model based Machine Learning Approach
Kanwarpartap Singh Gill, Vatsala Anand, Rahul Singh Chauhan, Ramesh Singh Rawat, Rupesh Gupta · 2023
The process of Kidney Disease Classification entails the categorization of kidney pictures or patient data into several disease classifications, including but not limited to chronic kidney disease (CKD) phases, polycystic kidney disease (PKD), kidney stones, renal tumours, and many other kidneyrelated illnesses. Machine learning algorithms have the capability to undergo training using datasets that are annotated with labels, enabling them to automatically detect and categorise certain circumstances. The primary objective of Kidney Disease Classification is to categorise kidney tumours or lesions into distinct categories, including renal cell carcinoma (RCC), angiomyolipoma (AML), oncocytoma, and several benign or malignant tumour subtypes. The use of this categorization system may provide valuable support in the process of diagnosing and formulating treatment strategies for individuals afflicted with kidney tumours. The process of Kidney Disease Classification entails the categorization of biopsy samples from patients who have had kidney transplantation, with the aim of ascertaining the presence or absence of rejection in the transplanted kidney. The categorization may include many classifications such as acute cellular rejection, acute antibody-mediated rejection, chronic rejection, or absence of rejection. The advancement of novel traits or combinations thereof, which have the potential to enhance the accuracy of categorization, is facilitated by social research in this domain, therefore contributing to the enhancement of quality of life. The objective of this research is to enhance patient health by using deep learning techniques to develop an X-ray classification system capable of detecting renal illness. The ResNet50V2 model demonstrated robust classification abilities in identifying renal disease, with a notable accuracy rate of 97%.