Tapan Kumar Gandhi

A comparative study of wavelet families for EEG signal classification

Over the past two decades, wavelet theory has been used for the processing of biomedical signals for feature extraction, compression and de-noising applications. However the question as to which wavelet family is the most suitable for analysis of non-stationary bio-signals is still prevalent among researchers. This paper attempts to find the most useful wavelet function among the existing members of the wavelet families for electroencephalogram signal (EEG) analysis. The EEGs considered for this study belong to both normal as well as abnormal signals like epileptic EEG. Important features such as energy, entropy and standard deviation at different sub-bands were computed using the wavelet functions-Haar, Daubechies (orders 2-10), Coiflets (orders 1-10), and Biorthogonal (orders 1.1, 2.4, 3.5, and 4.4). Feature vectors were used to model and train the Probabilistic Neural Network (PNN) and the classification accuracies were evaluated for each case. The results obtained from PNN classifier were compared with Support Vector Machine (SVM) classifier. From the statistical analysis, it was found that Coiflets 1 is the most suitable candidate among the wavelet families considered in this study for accurate classification of the EEG signals. In this work, we have attempted to improve the computing efficiency as it selects the most suitable wavelet function that can be used for EEG signal processing efficiently and accurately with lesser computational time.

Development of an expert multitask gadget controlled by voluntary eye movements

Design of assistive technology using advanced soft computing techniques on proper hardware platform has been an important issue of research for the last two decades. In the present study, a novel scheme is presented to develop a multitask gadget controlled by eye movements for the disabled, especially for individuals with spinal injury disorders. Electro-oculogram (EOG) signals generated by horizontal, vertical and diagonal eye movements and blinks were measured using a pair of surface electrodes with respect to a reference electrode placed on forehead. After preprocessing, the acquired signals were amplified with AC-coupling in order to reduce unnecessary drifts. Classifier based on DFA (Deterministic Finite Automata) was developed by using VHDL to discriminate 128 different EOG states from processed horizontal and vertical eye signals based on threshold settings specific to individuals. Later, online viability of the system was established by conducting some experiments on normal as well as disabled subjects. The utility of the proposed method was enhanced by implementing a robust algorithm for signal classification and training both the subjects and the device. It was found that with the proposed scheme, the accuracy of the detection and control of the specified gadget is 95.33%, with sensitivity and specificity as 95.6% and 95%, respectively. The proposed model can be used for designing smart houses for the disabled and elderly.

Multi-sequential MR brain image classification for tumor detection

Magnetic Resonance Imaging (MRI) is a diagnostic tool of remarkable potential in the area of neuroscience and clinical neuroimaging. The diagnostic accuracy can be limited by incompetence of the operating personnel, which can be supplemented by machine learning algorithms for classification of physiology and pathology. This paper uses effective information feature extraction, principal component analysis (PCA) for feature reduction and support vector machine (SVM) for classification of multi-sequence MR images of 7 patients. All axial slices of the brain are classified into normal and abnormal images. Various methods for feature extraction were tested among which effective information yielded the highest accuracy of 80.8% in a set of 677 images used for training and testing. The sensitivity and specificity were 80% and 81.06%, respectively. Different grid sizes were tested, and the highest accuracy was reported for 2 × 2 which indicates that the feature extraction must be taken over a small grid to ensure detection of minor variation from normal. The image sequences tested considered in the study are T1 weighted, T2 weighted, Fluid-attenuated inversion recovery (FLAIR), and post contrast T1 weighted. T2 weighted images were best classified with the maximum accuracy of 95.97%. This method proved to be effective to classify the images of all four sequences with accuracy ranging from 92–96%. The method was also tested with out of sample data and the accuracy obtained was 72.4%. The novelty of this work lies in the classification of multi-sequential images using all the different slices of the patient which includes the top of the skull as well as the mandible. The slices differ significantly as the spread of the tumor varies with each slice. The slices are taken at 5mm gap and the tumor can have a thickness less or more than the slice gap considered for the scan.

Enhancement of inter-hemispheric brain waves synchronisation after Pranayama practice

Pranayama practices are being used since ancient times as a holistic approach by saints and yogis to improve and control the subtle phenomenon of the brain and hence to bring mind and body in synchrony using breath as a link. This work is an attempt to understand this subtle phenomenon of the brain and the process of attaining higher state of cognition followed by consciousness. From the Electroencephalogram (EEG) signal analysis, it was observed that with the increase in Pranayama practices, the frequency of oscillation shifts from lower-frequency range to higher-frequency range with the significant rise of gamma power (>40 Hz) in frontal, central and also some part of temporal region of the brain. This frequency shifts count for attaining higher cognitive states leading to consciousness of human brain.

Dynamical neural activation in human brain during face recognition

This paper presents an experimental model and a set of stipulations for understanding neural progression in human brain while distinguishing familiar faces, and relationship between recognition and other aspects of face processing. Dynamical imagery stimuli of familiar and unfamiliar faces were shown to healthy individuals and were asked to recognise them as quickly and accurately as possible. Results obtained from the non-parametric analysis of the recorded multivariate data indicate that process of structural decoding of unfamiliar faces occurring inside the brain is delayed in comparison with familiar face probably due to few distinctive information that we derive from seen faces appear to influence the processing performance of the brain during the task.