Abstract

In recent years, MRI imaging has been used more widely for examination and detection of brain tumors than other medical imaging techniques. Detecting and segmenting the brain tumor area from these images is a time-consuming and boring task, and automated systems are required to overcome it. Thus, the present study aims to obtain areas of brain tumors with acceptable and reliable accuracy by using multi-stage morphological methods. To obtain reliable and valid accuracy, three datasets are used here. The accuracy of differentiation of brain tumor areas was obtained with an accuracy of 98% and 95.65% for the first and second groups, respectively. The results show that the proposed method is reliable for segmenting brain tumor area for this data group and can be generalized to other data sets.