Falls in! older adults are a major cause of morbidity and mortality and are a key class of preventable injuries. This paper presents a patient-specific (PS) fall prediction and detection prototype system that utilizes a single tri-axial accelerometer attached to the patient’s thigh to distinguish between activities of daily living (ADL) and fall events. The proposed system consists of two modes of operation: 1) fast mode for fall predication (FMFP) predicting a fall event (300msec-700msec) before occurring, 2) slow mode for fall detection (SMFD) with a 1-sec latency for detecting a fall event. The nonlinear Support Vector Machine Classifier (NLSVM)-based FMFP algorithm extracts 7 discriminating features for the pre-fall case to identify a fall risk event and alarm the patient. The proposed SMFD algorithm utilizes a Three-cascaded 1-sec sliding frames classification architecture with a linear regression-based offline training to identify a single and optimal threshold for each patient. Fall incidence will trigger an alarming notice to the concern healthcare providers via the internet. Experiments are performed with 20 different subjects (age above 65 years) and a total number of 100 associated falls and ADL recordings indoors and outdoors. The accuracy of the proposed algorithms is furthermore validated via MobiFall Dataset. FMFP achieves sensitivity and specificity of 97.8% and 99.1%, respectively, while SMFD achieves sensitivity and specificity of 98.6% and 99.3%, respectively, for a total number of 600 measured falls and ADL cases from 77 subjects.
The past few decades have seen ongoing development of continuous glucose monitoring (CGM) systems that are noninvasive and accurately measure blood glucose levels. The conventional finger-prick method, though accurate, is not feasible for use multiple times a day, as it is painful and test strips are expensive. Although minimally invasive and noninvasive CGM systems have been introduced into the market, they are expensive and require finger-prick calibrations. As the diabetes trend is high in low- and middle-income countries, a cost-effective and easy-to-use noninvasive glucose monitoring device is the need of the hour. This review paper briefly discusses the noninvasive glucose measuring technologies and their related research work. The technologies discussed are optical, transdermal, and enzymatic. The paper focuses on Near Infrared (NIR) technology and NIR Photoplethysmography (PPG) for blood glucose prediction. Feature extraction from PPG signals and glucose prediction with machine learning methods are discussed. The review concludes with key points and insights for future development of PPG NIR-based blood glucose monitoring systems.
9 March 2020 in IEEE Transactions on Biomedical Circuits and Systems
Conventional glucose monitoring methods for the growing number of diabetic patients around the world are invasive, painful, costly and, time-consuming. Complications aroused due to the abnormal blood sugar level in diabetic patients have created the necessity for continuous noninvasive glucose monitoring. This paper presents a wearable system for glucose monitoring based on a single wavelength near-infrared (NIR) Photoplethysmography (PPG) combined with machine-learning regression (MLR). The PPG readout circuit consists of a switched capacitor Transimpedance amplifier with 1MΩ gain and a 10-Hz switched capacitor LPF. It allows a DC bias current rejection up to 20μA with an input-referred current noise of ${7.3pA/\sqrt{Hz}}$ . The proposed digital processor eliminates motion artifacts, and baseline drifts from PPG signal extract six distinct features and finally predicts the blood glucose level using (Support Vector Regression with Fine Gaussian kernel) (FGSVR) MLR. A novel piece-wise linear (PWL) approach for the exponential function is proposed to realize the FGSVR on-chip. The overall system is implemented using 180nm CMOS process with a chip area of 4.0mm2 while consuming 1.62mW. The glucose measurement is performed for 200 subjects with R2 of 0.937. The proposed system accurately predicts the sugar level with a mean absolute relative difference (mARD) of 7.62%.
10 June 2019 in IEEE Transactions on Biomedical Circuits and Systems
Accurate monitoring of the depth of anesthesia (DoA) is essential for intraoperative and postoperative patient's health. Commercially available electroencephalograph (EEG)-based DoA monitors are recommended only for certain anesthetic drugs and specific age-group patients. This paper presents a machine learning classification processor for accurate DoA estimation irrespective of the patient's age and anesthetic drug. The classification is solely based on six features extracted from EEG signal, i.e. spectral edge frequency (SEF), beta ratio, and four bands of spectral energy (FBSE). A machine learning fine decision tree classifier is adopted to achieve a four-class DoA classification (deep, moderate, and light DoA versus awake state). The feature selection and the classification processor are optimized to achieve the highest classification accuracy for the state of moderate anesthesia required for the surgical operations. The proposed 256-point Fast Fourier Transform (FFT) accelerator is implemented to realize SEF, beta ratio and FBSE that enables minimal latency and high accuracy feature extraction. The proposed DoA processor is implemented using a 65nm CMOS technology and experimentally verified using FPGA based on the EEG recordings of 75 patients undergoing elective surgery with different types of anesthetic agents. The processor achieves an average accuracy of 92.2% for all DoA states, with a latency of 1 sec. The 0.09mm $^2$ DoA processor consumes 140nJ/classification.
Recent trends in research and the market show high demand for frequent monitoring of glucose to estimate the blood sugar level non-invasively which can replace the conventional finger-prick glucometer for daily use. This paper presents a high precision near-infrared Photoplethysmography (PPG) based noninvasive glucose monitoring System on Chip (SoC). The proposed system implements a fully differential Analog Frontend (AFE) with nonlinear medium Gaussian support-vector-regression (NMG-SVR) for glucose estimation. The AFE design incorporates chopping which enables the reduction of the integrated input-referred current noise to 9.4pArms thus achieving a dynamic range of 115dB. The glucose prediction processor (GPP) removes noise from the PPG signal, extracts ten unique features, and estimates the blood glucose level using a trained customized NMG-SVR model that minimizes the hardware cost by 25%. The extracted features are carefully designed and implemented to ensure inter-feature dependency, which helps to reduce the overall area by more than 40%. Moreover, GPP is implemented using power and clock gating techniques to minimize both static and dynamic power consumption. The proposed SoC is realized with $0.18~\mu \text{m}$ CMOS technology and occupies an area of 6 mm 2 . It dissipates a power of $186~\mu \text{W}$ and achieves a mean absolute relative difference (mARD) of 6.9% verified on 200 subjects.
23 February 2021 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Most surgical procedures are not possible without general anesthesia which necessitates continuous and accurate monitoring of the patients' level of hypnosis (LoH). Currently, the LoH is monitored using the conventional methods of either observing the patient's physiological parameters or using electroencephalogram (EEG)-based monitors. To overcome the limitations of the conventional methods, this work implements an accurate EEG-based LoH monitoring processor using a bagged tree machine-learning (BTML) classifier. It is based on 12 temporal and spectral features to incorporate robustness against age variation and achieve high classification accuracy. Spectral features are computed using discrete wavelet transform (DWT) that uses time-multiplexed filter (TMF) architecture. The TMF DWT consumes 110.6-nJ/feature vector for a 100-tap filter while reducing the area by 11% compared with the conventional method. Moreover, the BTML is implemented using a pipelined approach which enables an efficient on-chip implementation to reduce the hardware cost by 15x compared with the parallel approach. The proposed processor is implemented using a 180-nm CMOS process with an active area of 0.9 mm² while consuming 1.6 mW. The accuracy of the proposed hypnotic state monitor is verified using two EEG databases with a total of 95 patients and achieves a sensitivity and specificity of 95.4% and 97.7%, respectively.