Abstract
In addition to the acquisition of excessive and false positive rates, radiotherapists tend to neglect a considerable number of deformities. The identification of features from images, along with their processing by using a pattern detection algorithm, is required in image processing. This work developed a method on the basis of the adaptive fuzzy c-mean (AFCM) method and adaptive neuro-fuzzy inference system (ANFIS). The developed method considers selective parameters, which present a major challenge in differentiating hemorrhage and calcification. ANFIS can be considered an extension of the artificial neural network family and it exhibits excellent learning skills and estimation competencies. Thus, ANFIS is a highly productive tool that can effectively manage ambiguities in any system. In medical descriptions used by radiologists, the newly recommended AFCM is desirable to offer extensive information for the early determination of a cure via surgery and radiation treatment. Desirable outcomes and useful information are most likely to be provided by fuzzy clustering segmentation methods. Irregular tissues in hemorrhage and calcification can be clearly identified with the aid of AFCM techniques. Generally, the recently recommended AFCM segmentation technique is applied to medical image segmentation. By applying this unsupervised segmentation algorithm, radiotherapists can reduce the effects of noise resulting from low-resolution sensors or/and from the movement of assemblies during data collection. The medical discipline benefits from the proposed system, especially through the identification of hemorrhage and calcification.
Keyword(s)
Brain images, Calcification, Hemorrhage Fuzzy C-mean, Segmentation, ANFIS, classification, detection