Functional MRI Based Brain Mapping in Occipital Gyrus using Face Stimuli

Functional magnetic resonance imaging (fMRI) is one of the most powerful neuroimaging modalities due to its high spatio-temporal resolution characteristics. This known modality is applied on mapping the temporal, occipital, frontal cortices of the brain for localizing the neural activities generated due to any visual, physical or mental task or brain diseases or brain disorders. The occipital cortex is composed of middle, left, right, interior and exterior occipital gyrus and is responsible for visional function of human brain. The occipital gyrus reflects the neural image generated in the brain due to any visual activity. In this research paper, four different visual stimuli images of faces, scrambled, scenes and objects along with gap of blank space, forming a long sequence of stimuli observed by two female subjects, are experimented to examine and localize the most contrasting neural image generated in occipital gyrus of the brain. The visual fMRI brain data received from the two subjects is processed through fMRI-SPM12 toolbox based on Matlab software. In order to demonstrate the results statistically, two regressions such as T-contrast and F-contrast vectors are applied on fMRI images to highlight, and to localize the most active neural stimuli activities generated in the occipital gyrus of brain. In the results, it is demonstrated that maximum neural response can be mapped only for face stimulus in the bilateral occipital gyrus of the brain by applying T-contrast vectors regressions as when compared to other stimuli conditions and F-contrast vectors regressions. Further, it is also investigated that, the response of the face stimulus in F-contrast regressions achieved is somehow dispersed and unclear due to the large variances and interlinked communication of other stimuli or induced neural noises generated in entire volume of the brain.  Further from the given images, it is also investigated that the most reflecting and contrast area for any visual stimuli (such as face stimulus in this case) is either the middle or bilateral part of occipital gyrus of the human brain as identified through application of  T-contrast vectors regressions.

Efficient Computer Diagnostic System for Early Detection of Cancer

Breast cancer is considered to be one of the most fatal types of cancer among women. The early detection of breast cancer increases the survival probability. Computer Aided Diagnosis (CAD) system for detection and classifying of masses in mammograms is essential. In this thesis, new pre-processing techniques for mammograms have been proposed. First, an improved segmentation technique using image pre-processing and modified reaction diffusion based level set method is developed. The mammogram is passed through morphological operations, contrast enhancement and interpolation algorithm. The pre-processed image is then passed through a modified reaction diffusion based level set algorithm for accurate breast boundary segmentation. In Second pre-processing technique, pectoral muscle is suppressed by the proposed technique based on geometry and contrast variance between the breast tissues and pectoral muscles. Simulation results verify the significance of proposed schemes visually and quantitatively as compared to state of art existing schemes. Three class classification based on deep learning techniques i.e. CNN and RBMs is proposed. In the CNN-DW method, enhanced mammogram images are decomposed as its four subbands by means of two-dimensional discrete wavelet transform (2D-DWT), while in the second method discrete curvelet transform (DCT) is used. Proposed methods have been compared with existing methods in terms of accuracy rate, error rate, and various validation assessment measures. Simulation results clearly validate the significance and impact of our proposed model as compared to other well-known existing techniques. Last but not the least, a novel three-class classification technique for a large dataset of mammograms using a deep-learning method of restricted Boltzmann machine (RBM) is proposed. The augmented dataset is generated using mammogram patches. The dataset is filtered using a non-local means (NLM) filter to enhance the contrast of patches. These patches are decomposed using (2D-DWT) and (CT). The proposed method is compared with existing methods in terms of ROC curve, accuracy rate and various validation assessment measures. The simulation results clearly demonstrate the significance and impact of our proposed model compared to other well-known existing techniques.