Keralam: CUSAT Researchers Develop AI Model To Aid Early Detection Of Brain Aneurysms
The technology is still at the research stage and will require further validation and clinical studies before it can be integrated into routine healthcare systems.
Published : September 2, 2026 at 3:58 PM IST
Ernakulam: Researchers at the Cochin University of Science and Technology (CUSAT) have developed an Artificial Intelligence-based technology that could help detect cerebral aneurysms — abnormal ballooning or swelling in blood vessels in the brain — at an early stage with high precision.
The researchers say the anatomical-model-based AI system could assist doctors and radiologists in identifying subtle changes in brain blood vessels that may be difficult to detect through conventional examination methods.
However, the technology is still at the research stage and will require further validation and clinical studies before it can be integrated into routine healthcare systems.
The research was carried out at the Artificial Intelligence and Computer Vision Lab under the Department of Computer Science at CUSAT.
Researchers Keerthi A S Pillai and Preena K P developed the proposed algorithm under the guidance of Prof. Dr Madhu S. Nair.
Designed to Detect Subtle Changes in Brain Arteries
Prof Dr Madhu S Nair told ETV Bharat that detecting small or early-stage abnormalities in brain arteries can sometimes be challenging because of the complex structure and branching pattern of cerebral blood vessels.
According to him, the newly developed AI-based model is designed to analyse these complex vascular structures and identify even subtle abnormalities that could indicate the presence of a cerebral aneurysm.
The technology uses data obtained through Magnetic Resonance Angiography (MRA), an imaging technique used to visualise blood vessels.
The researchers said the system focuses on the anatomical structure and interconnected network of blood vessels in the brain to identify abnormal changes.
What Is A Cerebral Aneurysm?
A cerebral aneurysm is a bulging or balloon-like swelling that develops in a weakened area of a blood vessel in the brain.
While some aneurysms may not cause symptoms, a ruptured aneurysm can result in bleeding in or around the brain and can lead to a life-threatening medical emergency.
Researchers said the ability to identify such abnormalities at an early stage could potentially help doctors assess patients sooner and plan appropriate medical intervention.
Powered By Graph Neural Networks
The CUSAT research team used an advanced deep-learning technique known as a Graph Neural Network (GNN). Unlike conventional AI models that analyse information primarily as isolated images or data points, GNNs are designed to understand complex relationships and connections within structured data.
This makes the technology particularly suitable for analysing the branching and interconnected network of blood vessels in the brain.
The researchers incorporated the natural anatomical characteristics of cerebral blood vessels into the AI model.
According to the research team, this approach enables the algorithm to analyse vascular variations, branching patterns and structural differences that may be associated with cerebral aneurysms.
International Dataset Used for Research
The study used datasets associated with the internationally recognised Medical Image Computing and Computer Assisted Intervention (MICCAI) research community. The use of specialised medical imaging datasets allowed the researchers to train and evaluate the AI model for identifying abnormalities in cerebral vascular structures.
The researchers believe anatomical information could play an important role in improving the ability of AI systems to analyse complex medical imaging data.
Research Published In IEEE Journal
The research has received international academic recognition with the publication of the study in IEEE Transactions on Artificial Intelligence, a multidisciplinary peer-reviewed journal published by the Institute of Electrical and Electronics Engineers (IEEE) from Piscataway, New Jersey in the United States.
The publication marks an important academic milestone for the research team and the Artificial Intelligence and Computer Vision Lab at CUSAT.
Potential Role in Preventive Healthcare
Prof Madhu said earlier identification of cerebral aneurysms could potentially help doctors initiate timely monitoring and treatment, reducing the risk of serious complications, including stroke caused by bleeding in the brain.
The researchers believe AI-assisted systems could eventually help make medical image analysis faster and support doctors in clinical decision-making.
However, they emphasised that the technology is intended to assist medical professionals and cannot currently replace clinical examination or the judgement of doctors and radiologists.
Further Clinical Validation Needed
Despite the promising results, the researchers said extensive additional research and clinical validation will be necessary before the technology can be deployed in hospitals and routine diagnostic systems.
The AI model will need to be tested across larger and more diverse patient populations and evaluated in real-world clinical settings.
If successfully validated through further studies and clinical trials, the technology could eventually become a valuable decision-support tool for radiologists and doctors in the early detection and assessment of cerebral aneurysms.
The CUSAT study represents another step in the growing use of Artificial Intelligence in medical imaging, where researchers are increasingly exploring ways in which AI can assist doctors in detecting diseases earlier and improving the speed and accuracy of diagnosis.
Also read
