People around the globe rely on their blood samples for their glucose level measurement. There is a demand for non-invasive, precise and cost-effective solutions to monitor blood glucose level and control of diabetes. Serum glucose is an accurate blood glucose measurement method in comparison to capillary glucose measurement. Presently, the serum glucose is measured through laboratory setup with an invasive approach. The invasive method is painful and is not suitable for continuous glucose measurement. In this paper, we propose a novel wearable non-invasive consumer device (called iGLU 2.0) which can be used by consumers for accurate continuous blood glucose monitoring. This device uses a novel short near infrared (NIR) spectroscopy developed by us. It is incorporated with Internet-of-Medical-Things (IoMT) for smart healthcare where the healthcare data is stored on the cloud and is accessible to the users and caregivers. Analysis of the optimized regression model is performed and the system is calibrated and validated through healthy, prediabetic and diabetic patients. The robust regression models of serum glucose level is then deployed as the mechanism for precise measurement in iGLU 2.0. The performance of iGLU 2.0 is validated with the prediction of capillary blood glucose using Average Error (AvgE) and Mean Absolute Relative Difference (mARD) which are calculated as 6.09% and 6.07%, respectively, whereas for serum glucose, AvgE and mARD are estimated as 4.88% and 4.86%, respectively.
Electromyography (EMG) signals are gaining popularity to develop the prosthetics. In this paper, an efficient multi-channel EMG signal acquisition system has been proposed for upper limb prosthetic application. Various arm exercises have been performed to obtain EMG signals from five different arm muscles for the validation of developed hardware. The muscle's position has been selected by palpation method. Furthermore, the classification algorithms have been examined for seven different activities. Total 29 subjects have been chosen (25 intact and four Amputees) to acquire the EMG data by these activities. To classify the recorded EMG data set, nine time domain and seven frequency domain features have been extracted. A comparative analysis of different classifiers is presented for different muscle position of electrodes. The signal processing and classification algorithms have been processed in MATLAB 2016a. The accuracy of classification ranges for different classification algorithms from 57.69% to 99.92% for all subjects.
Diabetes is a long-term condition in which a person’s body cannot break down blood sugar adequately due to a shortage of insulin. The most crucial element of health care is continuously monitoring blood glucose (BG) levels. The main concern of effective glucose monitoring equipment is based on the blood-pricking technique. However, this may not be suggested for frequent glucose measurement. The paper presents various glucose-measuring technologies. The research discusses various non-invasive glucose measurement techniques and their management using advanced medical technologies. The configuration of the precise measuring device is essential to meet the blood glucose monitoring requirements that are not invasive systems. Non-invasive glucose monitoring devices solve the issue of frequently pricking patients for blood samples for clinical tests. For the goal of continuous health monitoring, a Smart Healthcare framework would be built on the Internet-of-Medical-Things (IoMT) and a Healthcare Cyber-Physical System (H-CPS) to estimate blood glucose. The study also discusses a few consumer devices and cutting-edge methods for measuring glucose. The paper also outlines the several difficulties and open challenges with glucose prediction.
A novel wearable consumer electronics device for detecting Major Depressive Disorder (MDD) has been developed using deep learning techniques for smart healthcare. Accurate identification of MDD through individual interviews or perceiving Electroencephalogram (EEG) signals is challenging. This study presents the concept of a novel wearable smart cap named DepCap for real-time detection of depression using EEG signals. First, spectrogram images are generated from the EEG signals of depressed and healthy patients using Short-Time Fourier Transform (STFT) to extract valuable features. Then, these spectrogram images obtained from STFT are used as input to the classification model. A deep analysis is done using various neural networks consisting of Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs). RNNs are used to extract temporal data from the EEG, while CNNs are used to retrieve spatial information. Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) are the two different kinds of RNNs evaluated in this work. The implemented combination models are (STFT+CNN), (STFT+CNN-LSTM) and (STFT+CNN-GRU). Four pre-trained models Inception, AlexNet, VGG16, and ResNet50 are also implemented along with the combination models. The dataset for this work is a publicly accessible dataset with 33 major depressive disorders and 30 healthy subjects. The evaluation results show that the STFT+CNN-LSTM has much better performance in terms of accuracy, sensitivity, specificity, and precision of 99.9%, 100%, 99.8%, and 99.4% respectively than other implemented models. The proposed wearable device DepCap is based on the STFT+CNN+LSTM model and is also integrated with the Internet of Medical Things (IoMT) framework for real-time depression detection.
16 October 2019 in IEEE Transactions on Industrial Electronics