Finding scars in the cerebral cortex through the analysis of intensities in T2/MRI sequences

Nowadays, the detection of scars in the cerebral cortex usually involves a manual process performed by neurologists and radiologists. It is a difficult task to carry out, since multiple troubles have to be overcome, from bad calibration of the equipment used to get the cerebral cortex images to several errors, such as spacial and geometrical distortions of MRI (Magnetic Resonance Imaging) sequences. These problems present serious complications in activities such as radiosurgery, which requires high spacial accuracy. Through the implementation of algorithms capable of analyzing MRI sequences, it is possible to automatically detect scars in the cerebral cortex in an easy and successful manner. In addition, the automatic detection of scars decreases the subjectivity of human interpretations and serves as a tool to support diagnoses of diseases. In this paper, a new methodology to automatically detect scars in the cerebral cortex is proposed. The main goal of this methodology is to facilitate the analysis of intensities in T2/MRI sequences by using the region growing and thresholds, as well as artificial neural networks.