Chennai: Researchers at IIT Madras and Christian Medical College (CMC), Vellore, have developed artificial intelligence-based tools to help doctors detect and assess kidney diseases at an early stage, potentially enabling faster diagnosis and better treatment planning.

The researchers have developed three complementary technologies, including a machine learning model to predict the risk of chronic kidney disease, a deep learning system to analyse CT scans, and a 3D imaging platform to assess kidney tumours.

The CT scan-based system can classify kidneys into four categories -- normal kidney, kidney cyst, kidney stone and kidney tumour. It has been trained using more than 12,000 images, the researchers said.

The 3D imaging platform recreates kidneys from CT scans and helps assess tumour volume and the extent of kidney involvement, providing doctors with patient-specific information for treatment decisions.

Kidney diseases often remain asymptomatic in their early stages and are detected only after significant damage has occurred. The AI-based tools are aimed at supporting physicians with faster and more consistent assessments.

The research was led by Prof G L Samuel of the Department of Mechanical Engineering at IIT Madras and research scholar Jennifer Delighta, in collaboration with Prof Santosh Varughese of the Department of Nephrology at CMC Vellore.

Samuel said the team sought to develop intelligent systems that could help clinicians make quicker and more informed decisions by combining machine learning with clinical knowledge.

The chronic kidney disease prediction model has also been developed into a user-friendly prototype to facilitate its possible clinical use.

The researchers said the work could be a step towards developing a kidney “Digital Twin”, combining AI-assisted image analysis with patient-specific 3D anatomical models for personalised clinical decision-making.

The research received institutional support from IIT Madras and the SPARC project.

The team plans to validate the models using larger patient datasets and work with healthcare institutions for their eventual deployment in real-world clinical settings.