Feature-versus deep learning-based approaches for the automated detection of brain tumor with magnetic resonance images: A comparative study

The public health is significantly affected by development of brain tumors in human patients. Glioblastoma (GBM) is a relatively common, malignant form of brain tumor, which is currently challenging to treat and cure. In contrast, Lower Grade Gliomas (LGGs) originate from glial cells and can mostly be treated and cured in the initial stages if they are detected early. A computer-aided diagnostics (CAD) tool may help to test for the presence and extent of any such tumor, and thus can be assistive in the clinical diagnostic process. Herein, we compare handcrafted versus non-handcrafted features-based CAD to characterize GBM and LGG. Our machine learning-based handcrafted model uses quantitative techniques of enhanced elongated quinary patterns and entropies analysis. We have also developed a non-handcrafted deep learning model using Visual Geometry Group-16 architecture for segregating GBM and LGG subjects results in 94.25% accuracy using k-nearest neighbor classifier.