India has become the most populous nation with 142.86 crore people since April 2023, surpassing China. The prevalence of diabetes in India has increased significantly, with the nationwide prevalence now exceeding 9%, and as high as 20% in the relatively prosperous southern cities. By 2030, it is predicted that India will have 100 million people with diabetes. Chronic kidney disease (CKD) can lead to kidney failure, heart disease, and stroke, resulting in early death and other health problems if left untreated. CKD is diagnosed by measuring the estimated glomerular filtration rate (eGFR), which is a calculation based on the level of creatinine in the blood. Renal biopsy is recommended for CKD patients with impaired renal function to make pathological diagnoses, especially for those with uncertain diagnoses. Renal biopsy is indicated for patients with diabetes under the suspicion of the presence of nephropathies other than diabetic nephropathy. Renal biopsy plays a crucial role in evaluating the nature and severity of renal injury in patients with diabetic kidney disease (DKD). In this study, the authors processed renal biopsy images with normal and sclerotic tissue sample artifacts classification histopathological images with whole slide imaging (WSI) contributed by the European project AIDPATH. The proposed deep neural network or deep convolutional neural network (CNN) for image classification used the Parametric Rectified Linear Unit (PReLU) activation function to classify sclerosed biopsy images and perform testing the dataset. The model’s accuracy was reported as 0.9465, indicating a high level of accuracy. The recall, which measures the model’s ability to correctly identify positive cases, was reported as 0.9358, showing that the model performs well in identifying positive cases. The F1-score, which combines precision and recall, was reported as 0.9375, indicating the effectiveness of the proposed method in diagnosing DKD. The authors have collected an MRI dataset from two hospitals, namely the Shri Ram Murti Smarak Institute of Medical Sciences (SRMS IMS) and the Bareilly MRI & CT Scan Centre, both located in Bareilly, Uttar Pradesh, India, for further study. In upcoming research, this dataset will be analyzed for renal biopsy. The dataset was collected to classify kidney diseases using transfer learning 1, to predict diabetic nephropathy, and to diagnose acute kidney injury. The Radiodiagnosis College and Labs at SRMS IMS Bareilly provide diagnostic and interventional support round the clock. The dataset will be useful in centers where CT/MRI is hard to develop. The authors’ research will contribute to the early diagnosis and treatment of kidney diseases, which can lead to better patient outcomes.