Detection of Kidney Disease in X-Ray Images Using Machine Learning on MobileNetV3 CNN Model
Kanwarpartap Singh Gill, Vatsala Anand, Rahul Chauhan, Prabhdeep Singh, Rupesh Gupta
Abstract
Classifying kidney pictures or patient data according to various disease categories, such as stages of chronic kidney disease (CKD), PKD, kidney stones, renal tumours, and other kidney-related illnesses, is known as kidney disease classification. These circumstances may be automatically recognised and categorised by machine learning algorithms that have been trained on labelled datasets. Renal cell carcinoma (RCC), angiomyolipoma (AML), oncocytoma, and various kinds of benign and malignant tumours are only few of the tumours or lesions that are the primary focus of Kidney Disease Classification. Kidney tumours may be easier diagnosed and treated with the help of this categorization. Kidney transplant recipients who are suffering rejection of their donated kidneys might have their biopsy samples classified according to the Kidney Disease Classification system. Acute cellular rejection, acute antibody-mediated rejection, chronic rejection, and no rejection are all possible categories. Social research in this field helps enhance quality of life by allowing the creation of new characteristics or feature combinations that may increase categorization accuracy. Creating a deep learning-based X-ray classifier that can detect renal disease is important to this study's overarching goal of better patient care. Classification tests showed that our MobileNetV3 model was <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$97{{\% }}$</tex> accurate in detecting renal illness.