A contrast adaptive histogram equalization with neural learning quantization (CAHE-NLQ) for blood clot detection in brain

Abstract

Uncharacteristic development of cell within the brain cranium through tumor formed in Brain could reduces the function of brain. Recently, this malfunction could be resolved by identifying brain tumor with the help of enhancement in Machine Learning (ML) and Image Processing (IP). Also, on these days Computer Vision along with the ML techniques plays major role for identifying Brain Tumor. The most important analytical technique for imaging or prior identification is through MRI which have been deployed with anomalous alterations in tissues. Beside, it is not interfering in imaging method but Initially for the blood clot identification, initial symptoms of brain disease, the image has been taken through Computer Aided Diagnosis (CAD) using image processing techniques with deep learning. So first the image is preprocessed using contrast adaptive histogram equalization with neural learning quantization (CAHE-NLQ). The simulation results show blood clot presence in brain after processing the input images, also proposed method achieves 93% accuracy, Specificity is 94% and also 92% of precision better results when compared to existing technique.