Search

Recent emergence in deep learning resulted in significant improvement in the segmentation accuracy of sub cortical brain structures like hippocampus and amygdala. The traditional methods of segmentation cannot produce an ideal segmentation result that exhibits issues like redundant computations, inconsistencies, coefficient variations and motion artifacts. Therefore, in this paper, an improved 3D Flatteringly Amplified Neural Network model for biomedical imaging is efficiently proposed, which can make full use of the 3D spatial information of MRI image itself to overcome the inconsistency of segmented images along with equalizing the coefficient variation of tiny region of brain image segmentation. Also while equalizing the coefficient, certain significant minute details are lost due to motion artifacts hence, the robust Amyg-Hippo Seg algorithm has been introducing that extracts the features through deep learning, and achieve high-precision segmentation, it reduced the computational complexity without neglecting minute features. In addition, the Daytona dropout function provides uncertainty information and reduces over-fitting problems. The outcome of the proposed work efficiently segments the most significant regions of hippocampus and amygdala with 97.4% accuracy. [ABSTRACT FROM AUTHOR]

The most necessary part of the living things which standardizes and manages other organs is the brain. The brain may get affected through any disease if the patient is not in a normal condition. Therefore it is significant to examine the condition of the brain. In the region of brain MRI image deformity fragmentation, various research works were made. However these research efforts presentations are needed in the image pre-analysis. During pre-analysis brain via MRI brain images for identifying the deformity, it is essential to examine the acquired patient's image in detail. An error treatment will be specified to the influenced patient if the study may have any error. So there is a necessity to develop precision in the deformity segmentation by achieving the fundamental pre-analysis in the MRI images. A combined approach with MRI brain image abnormality segmentation and denoising process is proposed in this paper. The proposed technique comprised of five stages namely, (i) Preprocessing, (ii) Feature Extraction, (iii) Image Classification, (iv) Segmentation and (v) Tissues Classification. Initially the database images are given to the preprocessing stage, for removing the noise. In preprocessing, the denoising process is performed it increases the segmentation and feature extraction accuracy. After the preprocessing, the image features are extracted to classify the images in the image database into normal and abnormal. After the image classification, the abnormal MRI images abnormal tissues like stroke, trauma and tumor are segmented. For this, the features are extracted from the segmented abnormal tissues. In the proposed technique, three features such as modified entropy, energy and innovative feature are extracted in the feature extraction stage. By using these extracted features, the abnormal tissues are classified by using a well known classification technique called Feed Forward Back Propagation Neural Network (FFBNN). The implementation results show the effectiveness of proposed MRI abnormality tissues segmentation technique in segmenting and classifying the MRI images and the achieved improvement in the segmentation and classification result. Furthermore, the performance of the proposed technique is evaluated by comparing with the existing MRI image segmentation techniques. [ABSTRACT FROM AUTHOR]