ABSTRACT
In this work, a novel fractional conformable edge detector for medical image structure feature extraction is proposed. The cases of study are detection and analysis of cerebral arteriovenous malformations, meningiomas and medulloblastomas by using computerised tomograph (CT) scans and magnetic resonance imaging (MRI) scans, which is motivated by the fact that these kinds of medical images are the most commonly used to carry out clinical studies to diagnose different diseases or pathologies for establishing treatment planning. In this sense, for a given medical image, the aim of the present work is to take advantage of the frequency characteristics of the fractional conformable derivative for extracting more structure feature details of these kinds of images, thus giving patients a timely diagnosis and adequate medical treatment. The experimental evaluation has demonstrated that the proposed operator gives superior performance compared to those classic operators, the Gabor method, and a fuzzy operator because it is able to detect more edge detail features of the medical images. Also, the proposed operator is more robust to noise. For the first time, in this work, novel fractional conformable masks based on fractional conformable derivative with power law for image edge detection are presented.
