Deep Learning Approach for Detecting Kidney Abnormalities from Medical Imaging

View at publisher Summarise with VIRAT
Authors: Jackulin T, Harini K R, Gopika L
Cited by: 0
References: 28

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Medical imaging is computed tomography (CT), is the instrumental in detection and diagnosis of the kidney disease. Manual analysis of CT scans, however, is, in most cases, it consumes large amount of time and exposed to inconsistency based on error done by the human. In an effort to replace this manual process, we created a deep learning–based classification system that would automatically analyze the CT kidney images. A publicly accessible dataset was utilized, and appropriate preprocessing measures including balancing, resizing, and augmentation were conducted to provide stable training. The model proposed relies on InceptionV3 architecture using transfer learning and fine-tuning so that the network can leverage pretrained ImageNet features for medical imaging tasks. The system was trained for classifying the CT scans into four classes: Normal, Tumor, Cyst and Stone. Performance based on test , set validated that such model could classify between these classes efficiently and effectively, its ability as a supportive tool for computer- aided diagnosis thus demonstrated. This research is an addition to the utilization of the Deep Learning in the medical imaging and attests to the utility of the same in enhancing diagnostic efficiency in kidney diseases.

Fields of Study & Subjects

AI in cancer detection Radiomics and Machine Learning in Medical Imaging Medical Image Segmentation Techniques

Publication details

Journal / source
International Conference on Image Processing and Capsule Networks (ICIPCN)
Pages
746–752
Publication date
27 January 2026
Language
EN