Other affiliations: Indian Institutes of Technology
Bio: S. Ramakrishnan is an academic researcher from Indian Institute of Technology Madras. The author has contributed to research in topics: Muscle fatigue & Electromyography. The author has an hindex of 9, co-authored 37 publications receiving 378 citations. Previous affiliations of S. Ramakrishnan include Indian Institutes of Technology.
TL;DR: The k-nearest neighbour algorithm is found to be the most accurate in classifying the features, with a maximum accuracy of 93% with the features selected using information gain ranking.
Abstract: In this work, an attempt has been made to differentiate surface electromyography (sEMG) signals under muscle fatigue and non-fatigue conditions with multiple time window (MTW) features. sEMG signals are recorded from biceps brachii muscles of 50 volunteers. Eleven MTW features are extracted from the acquired signals using four window functions, namely rectangular windows, Hamming windows, trapezoidal windows, and Slepian windows. Prominent features are selected using genetic algorithm and information gain based ranking. Four different classification algorithms, namely naive Bayes, support vector machines, k-nearest neighbour, and linear discriminant analysis, are used for the study. Classifier performances with the MTW features are compared with the currently used time- and frequency-domain features. The results show a reduction in mean and median frequencies of the signals under fatigue. Mean and variance of the features differ by an order of magnitude between the two cases considered. The number of features is reduced by 45% with the genetic algorithm and 36% with information gain based ranking. The k-nearest neighbour algorithm is found to be the most accurate in classifying the features, with a maximum accuracy of 93% with the features selected using information gain ranking.
TL;DR: An approach that uses discrete wavelet transform to decompose signals and singular value decomposition (SVD) to embed the secret information into the decomposed ECG signal and the observations validate that HH is the ideal sub-band to hide data.
Abstract: ECG Steganography provides secured transmission of secret information such as patient personal information through ECG signals. This paper proposes an approach that uses discrete wavelet transform to decompose signals and singular value decomposition (SVD) to embed the secret information into the decomposed ECG signal. The novelty of the proposed method is to embed the watermark using SVD into the two dimensional (2D) ECG image. The embedding of secret information in a selected sub band of the decomposed ECG is achieved by replacing the singular values of the decomposed cover image by the singular values of the secret data. The performance assessment of the proposed approach allows understanding the suitable sub-band to hide secret data and the signal degradation that will affect diagnosability. Performance is measured using metrics like Kullback---Leibler divergence (KL), percentage residual difference (PRD), peak signal to noise ratio (PSNR) and bit error rate (BER). A dynamic location selection approach for embedding the singular values is also discussed. The proposed approach is demonstrated on a MIT-BIH database and the observations validate that HH is the ideal sub-band to hide data. It is also observed that the signal degradation (less than 0.6 %) is very less in the proposed approach even with the secret data being as large as the sub band size. So, it does not affect the diagnosability and is reliable to transmit patient information.
TL;DR: MBD based time–frequency spectrum is able to provide the instantaneous variations of frequency components associated with fatiguing contractions and it is found that the values of IMDF, IMNF and InstSPR in LFB region have lowest variability across different subjects in comparison with other two features.
20 Dec 2012
TL;DR: In this work, particle swarm optimization algorithm based multilevel thresholding is adopted for detecting the vasculature structures in retinal fundus images and the optimal multi-threshold selection using particle Swarm optimization seems to provide better results.
Abstract: Retinal vasculature of the human circulatory system which can be visualized directly provides a number of systemic conditions and can be diagnosed by the detection of lesions. Changes in these structures are found to be correlated with pathological conditions and provide information on severity or state of various diseases. In this work, particle swarm optimization algorithm based multilevel thresholding is adopted for detecting the vasculature structures in retinal fundus images. Initially, adaptive histogram equalization is used for pre-processing of the original images. Tsallis multilevel thresholding is used for the segmentation of the blood vessels. Further, similarity measures are used to quantify the similarity between the segmented result and the corresponding ground truth. The optimal multi-threshold selection using particle swarm optimization seems to provide better results. Similarity measures analysis using dendrogram and box plot provide validation of the segmentation procedure attempted.
TL;DR: Results show that there is a progressive increase in cyclostationary during the progression of muscle fatigue, and these SCD features could be useful in the automated analysis of sEMG signals for different neuromuscular conditions.
Abstract: Analysis of neuromuscular fatigue finds various applications ranging from clinical studies to biomechanics. Surface electromyography (sEMG) signals are widely used for these studies due to its non-invasiveness. During cyclic dynamic contractions, these signals are nonstationary and cyclostationary. In recent years, several nonstationary methods have been employed for the muscle fatigue analysis. However, cyclostationary based approach is not well established for the assessment of muscle fatigue. In this work, cyclostationarity associated with the biceps brachii muscle fatigue progression is analyzed using sEMG signals and Spectral Correlation Density (SCD) functions. Signals are recorded from fifty healthy adult volunteers during dynamic contractions under a prescribed protocol. These signals are preprocessed and are divided into three segments, namely, non-fatigue, first muscle discomfort and fatigue zones. Then SCD is estimated using fast Fourier transform accumulation method. Further, Cyclic Frequency Spectral Density (CFSD) is calculated from the SCD spectrum. Two features, namely, cyclic frequency spectral area (CFSA) and cyclic frequency spectral entropy (CFSE) are proposed to study the progression of muscle fatigue. Additionally, degree of cyclostationarity (DCS) is computed to quantify the amount of cyclostationarity present in the signals. Results show that there is a progressive increase in cyclostationary during the progression of muscle fatigue. CFSA shows an increasing trend in muscle fatiguing contraction. However, CFSE shows a decreasing trend. It is observed that when the muscle progresses from non-fatigue to fatigue condition, the mean DCS of fifty subjects increases from 0.016 to 0.99. All the extracted features found to be distinct and statistically significant in the three zones of muscle contraction (p?0.05). It appears that these SCD features could be useful in the automated analysis of sEMG signals for different neuromuscular conditions.
TL;DR: This in vivo dMRI study demonstrates that N ODDI identifies potential tissue sources contributing to DTI indices and NODDI may provide greater specificity to pathology in Alzheimer's disease.
TL;DR: In the process of gesture recognition using sEMG signals generated by thumb, a method of redundant electrode determination based on variance theory is proposed and the best method of thumb motion pattern recognition is obtained.
Abstract: Human computer interaction plays an increasingly important role in our life. People need more intelligent, concise and efficient human-computer interaction. It is of great significance to optimize the process of human-computer interaction by using appropriate calculation methods. In order to eliminate the interference data of thumb recognition based on sEMG signal in the process of human-computer interaction, simplify the data processing, and improve the working efficiency of general equipment of sEMG signal. In the process of gesture recognition using sEMG signals generated by thumb, a method of redundant electrode determination based on variance theory is proposed. The redundancy of five groups of action signals is divided into 16 levels and visualized. By comparing the results of thumb motion recognition when different redundant channels are removed, the optimal channel combination in the process of thumb motion recognition is obtained. Finally, two kinds of classifiers suitable for sEMG signal field are selected, and the classification results are compared, and the best method of thumb motion pattern recognition is obtained.
TL;DR: Histogram based multilevel thresholding approach is proposed using Brownian distribution guided firefly algorithm (FA) and results show that BD guided FA provides better objective function, PSNR, and SSIM, whereas LF based FA provides faster convergence with relatively lower CPU time.
Abstract: Histogram based multilevel thresholding approach is proposed using Brownian distribution (BD) guided firefly algorithm (FA). A bounded search technique is also presented to improve the optimization accuracy with lesser search iterations. Otsu's between-class variance function is maximized to obtain optimal threshold level for gray scale images. The performances of the proposed algorithm are demonstrated by considering twelve benchmark images and are compared with the existing FA algorithms such as Levy flight (LF) guided FA and random operator guided FA. The performance assessment comparison between the proposed and existing firefly algorithms is carried using prevailing parameters such as objective function, standard deviation, peak-to-signal ratio (PSNR), structural similarity (SSIM) index, and search time of CPU. The results show that BD guided FA provides better objective function, PSNR, and SSIM, whereas LF based FA provides faster convergence with relatively lower CPU time.
TL;DR: The paper introduces elementary concepts of digital watermarking, characteristics and novel applications of watermark in detail, and various analysis and comparison of different notable water marking techniques are discussed in tabular format.
Abstract: Robustness, imperceptibility and embedding capacity are the preliminary requirements of any watermarking technique. However, research concluded that these requirements are difficult to achieve at same time. In this paper, we review various recent robust and imperceptible watermarking methods in spatial and transform domain. Further, the paper introduces elementary concepts of digital watermarking, characteristics and novel applications of watermark in detail. Furthermore, various analysis and comparison of different notable watermarking techniques are discussed in tabular format. We believe that our survey contribution will helpful for fledgling researchers to develop robust and imperceptible watermarking algorithms for various practical applications.
TL;DR: This paper proposes a semi-automated tool to investigate the medical MRI captured with contrast improved T1 modality (T1C), which considers the integration of Bat algorithm and Tsallis based thresholding along with region growing (RG) segmentation.
Abstract: In medical domain, diseases in critical internal organs are generally inspected using invasive/non-invasive imaging techniques. Magnetic resonance imaging (MRI) is one of the commonly considered imaging approaches to confirm the abnormality in various internal organs. After recording the MRI, an appropriate image processing exercise is to be implemented to investigate and infer the severity of the disease and its location. This paper proposes a semi-automated tool to investigate the medical MRI captured with contrast improved T1 modality (T1C). This technique considers the integration of Bat algorithm (BA) and Tsallis based thresholding along with region growing (RG) segmentation. Proposed approach is tested on RGB/gray scale images of brain and breast MRI recorded along with a contrast agent. After mining the infected region, its texture features are extracted with Haralick function to assess the surface details of abnormal section. Performance of RG is confirmed against other segmentation methods, such as level set (LS), principal component analysis (PCA) and watershed. The clinical significance of the proposed technique is also validated using the brain images of BRATS recorded using T1C modality. The experiment outcome confirms that, the implemented procedure provides better values of Jaccard (87.41%), Dice (90.36%), sensitivity (98.27%), specificity (97.72%), accuracy (97.53%) and precision (95.85%) for the considered BRATS brain MRI.