Nizhny Novgorod Scientists Develop Hybrid Neural Network to Detect Brain Tumors
The system achieves accuracy of up to 98%.
Scientists at Lobachevsky State University of Nizhny Novgorod have developed a hybrid quantum-classical neural network that can detect brain tumors in MRI scans with accuracy of up to 98%.
The system combines a conventional neural network with a quantum layer. The latter processes a compressed representation of features extracted from the scans. This approach significantly reduces the amount of data needed to train the conventional neural network.
One of the key advantages of the new system is its resistance to image distortions. Even when processing distorted images, it achieves accuracy above 96%.
So far, the system has been tested on images from open datasets rather than in a clinical setting. The next step will be to test the method on new medical data and Russian quantum computers.
