Anisha Halder Roy

Bio: Anisha Halder Roy is an academic researcher from University of Calcutta. The author has contributed to research in topics: Support vector machine & Feature extraction. The author has an hindex of 1, co-authored 7 publications receiving 3 citations.

Detection of Stress in Human Brain

Abstract: The essence of the paper is to develop a stress detection mechanism and a stress level indicator circuit for measuring the stress level of human brain using the Electro-encephalogram (EEG) Signal. Signals coming from the frontal lobe of human brain have been used for the measurement of stress. The brain signals of the thirty subjects are recorded while they are solving five mathematical question sets with increasing complexity. We assume that the subjects undergo through five different stress levels i.e. ‘Relaxed’, ‘Less stressed’, ‘Moderately Stressed’, ‘High Stressed’ and ‘Alarmingly Stressed’ while solving these question sets. After that recorded EEG data is processed and features are extracted. We design a feed forward neural network for classifying the stress level in human brain. We prepare a new question set consisting of easy as well as complex numerical questions for testing purpose. We record the EEG data of a subject while solving this question set. We extract six feature values from the processed EEG data of the subject. These data is fed to the designed feed forward neural network. The neural network predicts the stress level and the predicted stress level is indicated in the ‘Stress Indicating’ circuit.

Detection of Cardiac Arrhythmia based on feature fusion and Machine Learning algorithms

Abstract: Automated analysis of Electrocardiogram(ECG) signal has a great importance for early diagnosis of cardiovascular arrhythmia. This work deals with the governing process for the detection of arrhythmic and normal ECG signals which are collected from MIT-BIH database. Collected raw ECG signals are used for preprocessing to remove the noise from the signal. From the preprocessed ECG signal different morphological and nonlinear parameters are extracted as features for the detection process. Adaptive weight factor of individual feature is evaluated for the development of fused feature sets for better classification accuracy. Machine learning algorithms like Decision tree, Random Forest, Support Vector Machine have been employed for the classification process. The best classification result with 99.08% accuracy has been obtained using the Random Forest classifier.

Detecting Different Emotional States of Human Brain Using Bio-potential Signals

Abstract: The essence of this paper is to design a mechanism for detecting emotional state of person using different bio-potential signals like electroencephalogram (EEG) signals of frontal lobe, pulse rate and SpO2. We record EEG signals of Fp1, Fp2, F3 and F4 electrodes, pulse rate and SpO2 of thirty subjects and extract twenty-two features from the recorded bio-potential signals. We design a k-nearest neighbor (KNN) classifier model for predicting the emotional state of a person. The designed KNN classifier model is trained with the extracted feature values of the thirty subjects. We again record the same bio-potential signals of ten new subjects and extract features. These extracted feature values are used for validating the performance of the trained KNN classifier model. The obtained overall efficiency of our designed emotion detection mechanism is 95.4%.

Design of a Computationally Economical Image Classifier using Generic Features

Abstract: In this paper, we propose an image classification technique which uses a simple autoencoder with a regularizer. Nowadays, Convolutional Neural Networks (CNN) are primarily used for image classification. Our method can be used for image classification with much reduced requirement of computational capability than a complex CNN which has a huge number of degrees of freedom. Here, the terms simple and complex, respectively, correspond to the simplicity and the complexity of a network in terms of the number of learnable parameters (degrees of freedom) and the number of hidden layers. This technique uses features extracted from a pretrained CNN, trained on a completely different dataset. Genetic algorithm solves for the optimal hyperparameters of the pretrained CNN. It is observed that these features serve as important and robust parameters for the training of the autoencoder, as a final average image classification accuracy improvement of about 17.45% is observed with the inclusion of these features. We use a pretrained CNN on MNIST dataset and classify images of several other benchmark datasets. We utilize different classifiers for image classification based on features extracted from the autoencoder and repeat each of the experiments a number of times with different random initialization of the classifier and the weight matrix of the autoencoder. We also perform experiments by pretraining the CNN with different datasets. Our results show a notable image classification accuracy and a significant reduction of training time with respect to a complex CNN.

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Abstract: Summary Background Since December, 2019, Wuhan, China, has experienced an outbreak of coronavirus disease 2019 (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Epidemiological and clinical characteristics of patients with COVID-19 have been reported but risk factors for mortality and a detailed clinical course of illness, including viral shedding, have not been well described. Methods In this retrospective, multicentre cohort study, we included all adult inpatients (≥18 years old) with laboratory-confirmed COVID-19 from Jinyintan Hospital and Wuhan Pulmonary Hospital (Wuhan, China) who had been discharged or had died by Jan 31, 2020. Demographic, clinical, treatment, and laboratory data, including serial samples for viral RNA detection, were extracted from electronic medical records and compared between survivors and non-survivors. We used univariable and multivariable logistic regression methods to explore the risk factors associated with in-hospital death. Findings 191 patients (135 from Jinyintan Hospital and 56 from Wuhan Pulmonary Hospital) were included in this study, of whom 137 were discharged and 54 died in hospital. 91 (48%) patients had a comorbidity, with hypertension being the most common (58 [30%] patients), followed by diabetes (36 [19%] patients) and coronary heart disease (15 [8%] patients). Multivariable regression showed increasing odds of in-hospital death associated with older age (odds ratio 1·10, 95% CI 1·03–1·17, per year increase; p=0·0043), higher Sequential Organ Failure Assessment (SOFA) score (5·65, 2·61–12·23; p Interpretation The potential risk factors of older age, high SOFA score, and d-dimer greater than 1 μg/mL could help clinicians to identify patients with poor prognosis at an early stage. Prolonged viral shedding provides the rationale for a strategy of isolation of infected patients and optimal antiviral interventions in the future. Funding Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences; National Science Grant for Distinguished Young Scholars; National Key Research and Development Program of China; The Beijing Science and Technology Project; and Major Projects of National Science and Technology on New Drug Creation and Development.

Golden Standard or Obsolete Method? Review of ECG Applications in Clinical and Experimental Context

Abstract: Cardiovascular system and its functions under both physiological and pathophysiological conditions have been studied for centuries. One of the most important steps in the cardiovascular research was the possibility to record cardiac electrical activity. Since then, numerous modifications and improvements have been introduced; however, an electrocardiogram still represents a golden standard in this field. This paper overviews possibilities of ECG recordings in research and clinical practice, deals with advantages and disadvantages of various approaches, and summarizes possibilities of advanced data analysis. Special emphasis is given to state-of-the-art deep learning techniques intensely expanded in a wide range of clinical applications and offering promising prospects in experimental branches. Since, according to the World Health Organization, cardiovascular diseases are the main cause of death worldwide, studying electrical activity of the heart is still of high importance for both experimental and clinical cardiology.

Prediction of atrial fibrillation based on nonlinear modeling of heart rate variability signal and SVM classifier

Abstract: Cardiac arrhythmia is one of the major causes of death worldwide. Atrial fibrillation (AF) is considered as the most prevalent sustained cardiac arrhythmia. It increases the risk of cardiac stroke and heart failure. This study aims to present an automated technique for AF detection by analyzing the ECG signal so that individual heart condition can be monitored accurately and an alarm system can be simulated if any serious cardiac abnormality occurs. The heart rate variability (HRV) signal reflects the fluctuation of heart in different time intervals. The proposed algorithm includes nonlinear methods for characterizing the dynamics of HRV signal to find diagnosis pattern for AF detection. The diagnostically relevant nonlinear parameters are extracted from HRV signal. The extracted features are subjected to decision tree and support vector machine (SVM) classifier to discriminate AF from normal heart condition. The experimental result is evaluated on 25 ECG data set of AF and 54 ECG data sets of normal subjects taken from Physionet database to illustrate the diagnostic ability of the classifiers. The tenfold cross-validation method is also applied for performance evaluation. The proposed algorithm has achieved an average accuracy of 99.11%, sensitivity, specificity, and F-score values of 98.92%, 99.25%, and 99.08%, respectively, using SVM classifier which is better than the result obtained from decision tree classifier having average accuracy of 96.41% and F-score of 95.71%. The proposed algorithm provides great potential for AF diagnosis with high accuracy. This work yields superior performance based on the comparative study with the existing scientific approaches to categorize AF from normal ones.