Deep neural network correlation learning mechanism for CT brain tumor detection

Abstract Modern medical clinics support examinations with computer systems which use Computational Intelligence on the way to detect potential health problems in more efficient way. One of most important applications is evaluation CT brain scans, where precise results come from deep learning approaches. In this article, we propose a novel correlation mechanism (CLM) for neural network architectures that combines convolutional (CNN) classic architecture. The helps CNN find adequate filers pooling and convolution layers. As result, main classifier learns faster reaches higher efficiency. Results show our CLM model able reach about 96% accuracy, 95% precision recall. We have described proposed discussed numerical draw conclusions future works.