A 13.34 μW Event-Driven Patient-Specific ANN Cardiac Arrhythmia Classifier for Wearable ECG Sensors

A Novel Instruction Driven 1-D CNN Processor for ECG Classification

Electrocardiography (ECG) has emerged as a ubiquitous diagnostic tool for the identification and characterization of diverse cardiovascular pathologies. Wearable health monitoring devices, equipped with on-device biomedical artificial intelligence (AI) processors, have revolutionized the acquisition, analysis, and interpretation of ECG data. However, these systems necessitate AI processors that exhibit flexible configuration, facilitate portability, and demonstrate optimal performance in terms of power consumption and latency for the realization of various functionalities. To address these challenges, this study proposes an instruction-driven convolutional neural network (CNN) processor. This processor incorporates three key features: (1) An instruction-driven CNN processor to support versatile ECG-based application. (2) A Processing element (PE) array design that simultaneously considers parallelism and data reuse. (3) An activation unit based on the CORDIC algorithm, supporting both Tanh and Sigmoid computations. The design has been implemented using 110 nm CMOS process technology, occupying a die area of 1.35 mm2 with 12.94 µW power consumption. It has been demonstrated with two typical ECG AI applications, including two-class (i.e., normal/abnormal) classification and five-class classification. The proposed 1-D CNN algorithm performs with a 97.95% accuracy for the two-class classification and 97.9% for the five-class classification, respectively.

A 36-nW Electrocardiogram Anomaly Detector Based on a 1.5-bit Non-Feedback Delta Quantizer for Always-on Cardiac Monitoring

An always-on electrocardiogram (ECG) anomaly detector (EAD) with ultra-low power (ULP) consumption is proposed for continuous cardiac monitoring applications. The detector is featured with a 1.5-bit non-feedback delta quantizer (DQ) based feature extractor, followed by a multiplier-less convolutional neural network (CNN) engine, which eliminates the traditional high-resolution analog-to-digital converter (ADC) in conventional signal processing systems. The DQ uses a computing-in-capacitor (CIC) subtractor to quantize the sample-to-sample difference of ECG signal into 1.5-bit ternary codes, which is insensitive to low-frequency baseline wandering. The subsequent event-driven classifier is composed of a low-complexity coarse detector and a systolic-array-based CNN engine for ECG anomaly detection. The DQ and the digital CNN are fabricated in 65-nm and 180-nm CMOS technology, respectively, and the two chips are integrated on board through wire bonding. The measured detection accuracy is 90.6% ∼ 91.3% when tested on the MIT-BIH arrhythmia database, identifying three different ECG anomalies. Operating at 1 V and 1.4 V power supplies for the DQ and the digital CNN, respectively, the measured long-term average power consumption of the core circuits is 36 nW, which makes the detector among those state-of-the-art always-on cardiac anomaly detection devices with the lowest power consumption.

An Ultra-Energy-Efficient and High Accuracy ECG Classification Processor With SNN Inference Assisted by On-Chip ANN Learning

The ECG classification processor is a key component in wearable intelligent ECG monitoring devices which monitor the ECG signals in real time and detect the abnormality automatically. The state-of-the-art ECG classification processors for wearable intelligent ECG monitoring devices are faced with two challenges, including ultra-low energy consumption demand and high classification accuracy demand against patient-to-patient variability. To address the above two challenges, in this work, an ultra-energy-efficient ECG classification processor with high classification accuracy is proposed. Several design techniques have been proposed, including a reconfigurable SNN/ANN inference architecture for reducing energy consumption while maintaining classification accuracy, a reconfigurable on-chip learning architecture for improving the classification accuracy against patent-to-patient variability, and a dual-purpose binary encoding scheme of ECG heartbeats for further reducing the energy consumption. Fabricated with a 28nm CMOS technology, the proposed design consumes extremely low classification energy (0.3μJ) while achieving high classification accuracy (97.36%) against patient-to-patient variability, outperforming several state-of-the-art designs.

A 2.66 µW Clinician-Like Cardiac Arrhythmia Watchdog Based on P-QRS-T for Wearable Applications

A wearable electrocardiogram (ECG) device is an effective tool for managing cardiovascular diseases. This paper presents a low power clinician-like cardiac arrhythmia watchdog (CAW) for wearable ECG devices. The CAW is based on a novel P-QRS-T detection algorithm that makes use of clinical features to identify abnormalities. Implemented in 0.18 μm CMOS process, the CAW consumes 2.66 µW for 80 bpm heart rate at 1.2 V supply with an area of 0.578 mm². Verified on QT database, the average sensitivity/positive predictivity for P-wave, QRS complex and T-wave are over 93.39%/88.55%, 99.69%/99.48%, and 97.13%/93.18% respectively, across over 190000 beats. It shows over 99.8% arrhythmia detection accuracy for 43 subjects evaluated on MIT-BIH database.

Low Complexity Binarized 2D-CNN Classifier for Wearable Edge AI Devices

Wearable Artificial Intelligence-of-Things (AIoT) devices exhibit the need to be resource and energy-efficient. In this paper, we introduced a quantized multilayer perceptron (qMLP) for converting ECG signals to binary image, which can be combined with binary convolutional neural network (bCNN) for classification. We deploy our model into a low-power and low-resource field programmable gate array (FPGA) fabric. The model requires 5.8× lesser multiply and accumulate (MAC) operations than known wearable CNN models. Our model also achieves a classification accuracy of 98.5%, sensitivity of 85.4%, specificity of 99.5%, precision of 93.3%, and F1-score of 89.2%, along with dynamic power dissipation of 34.9 μW.

A Highly Energy-Efficient Hyperdimensional Computing Processor for Biosignal Classification

Hyperdimensional computing (HDC) is a brain-inspired computing paradigm that operates on pseudo-random hypervectors to perform high-accuracy classifications for biomedical applications. The energy efficiency of prior HDC processors for this computationally minimal algorithm is dominated by costly hypervector memory storage, which grows linearly with the number of sensors. To address this, the memory is replaced with a light-weight cellular automaton for on-the-fly hypervector generation. The use of this technique is explored in conjunction with vector folding for various real-time classification latencies in post-layout simulation on an emotion recognition dataset with 200 channels. The proposed architecture achieves 39.1 nJ/prediction; a 4.9× energy efficiency improvement, 9.5× per channel, over the state-of-the-art HDC processor. At maximum throughput, the architecture achieves a 10.7× improvement, 33.5× per channel. An optimized support vector machine (SVM) processor is designed in this work for the same use-case. HDC is 9.5× more energy-efficient than the SVM, paving the way for it to become the paradigm of choice for high-accuracy, on-board biosignal classification.

A Neuromorphic Processing System With Spike-Driven SNN Processor for Wearable ECG Classification

This paper presents a neuromorphic processing system with a spike-driven spiking neural network (SNN) processor design for always-on wearable electrocardiogram (ECG) classification. In the proposed system, the ECG signal is captured by level crossing (LC) sampling, achieving native temporal coding with single-bit data representation, which is directly fed into an SNN in an event-driven manner. A hardware-aware spatio-temporal backpropagation (STBP) is suggested as the training scheme to adapt to the LC-based data representation and to generate lightweight SNN models. Such a training scheme diminishes the firing rate of the network with little plenty of classification accuracy loss, thus reducing the switching activity of the circuits for low-power operation. A specialized SNN processor is designed with the spike-driven processing flow and hierarchical memory access scheme. Validated with field programmable gate arrays (FPGA) and evaluated in 40 nm CMOS technology for application-specific integrated circuit (ASIC) design, the SNN processor can achieve 98.22% classification accuracy on the MIT-BIH database for 5-category classification, with an energy efficiency of 0.75 μJ/classification.

A 746 nW ECG Processor ASIC Based on Ternary Neural Network

This paper presents an ultra-low power electrocardiography (ECG) processor application-specific integrated circuit (ASIC) for the real-time detection of abnormal cardiac rhythms (ACRs). The proposed ECG processor can support wearable or implantable ECG devices for long-term health monitoring. It adopts a derivative-based patient adaptive threshold approach to detect the R peaks in the PQRST complex of ECG signals. Two tiny machine learning classifiers are used for the accurate classification of ACRs. A 3-layer feed-forward ternary neural network (TNN) is designed, which classifies the QRS complex's shape, followed by the adaptive decision logics (DL). The proposed processor requires only 1 KB on-chip memory to store the parameters and ECG data required by the classifiers. The ECG processor has been implemented based on fully-customized near-threshold logic cells using thick-gate transistors in 65-nm CMOS technology. The ASIC core occupies a die area of 1.08 mm². The measured total power consumption is 746 nW, with 0.8 V power supply at 2.5 kHz real-time operating clock. It can detect 13 abnormal cardiac rhythms with a sensitivity and specificity of 99.10% and 99.5%. The number of detectable ACR types far exceeds the other low power designs in the literature.

An Energy Efficient ECG Ventricular Ectopic Beat Classifier Using Binarized CNN for Edge AI Devices

Wearable Artificial Intelligence-of-Things (AIoT) requires edge devices to be resource and energy-efficient. In this paper, we design and implement an efficient binary convolutional neural network (bCNN) algorithm utilizing function-merging and block-reuse techniques to classify between Ventricular and non-Ventricular Ectopic Beat images. We deploy our model into a low-resource low-power field programmable gate array (FPGA) fabric. Our model achieves a classification accuracy of 97.3%, sensitivity of 91.3%, specificity of 98.1%, precision of 86.7%, and F1-score of 88.9%, along with dynamic power dissipation of only 10.5-μW.

ANNet: A Lightweight Neural Network for ECG Anomaly Detection in IoT Edge Sensors

In this paper, we propose a lightweight neural network for real-time electrocardiogram (ECG) anomaly detection and system level power reduction of wearable Internet of Things (IoT) Edge sensors. The proposed network utilizes a novel hybrid architecture consisting of Long Short Term Memory (LSTM) cells and Multi-Layer Perceptrons (MLP). The LSTM block takes a sequence of coefficients representing the morphology of ECG beats while the MLP input layer is fed with features derived from instantaneous heart rate. Simultaneous training of the blocks pushes the overall network to learn distinct features complementing each other for making decisions. The network was evaluated in terms of accuracy, computational complexity, and power consumption using data from the MIT-BIH arrhythmia database. To address the class imbalance in the dataset, we augmented the dataset using SMOTE algorithm for network training. The network achieved an average classification accuracy of 97% across several records in the database. Further, the network was mapped to a fixed point model, retrained in a bit accurate fixed-point environment to compensate for the quantization error, and ported to an ARM Cortex M4 based embedded platform. In laboratory testing, the overall system was successfully demonstrated, and a significant saving of ≅ 50% power was achieved by gating the wireless transmission using the classifier. Wireless transmission was enabled only to transmit the beats deemed anomalous by the classifier. The proposed technique compares favourably with current methods in terms of computational complexity and has the advantage of stand-alone operation in the edge node, without the need for always-on wireless connectivity making it ideal for IoT wearable devices.

Machine Learning for Cardiovascular Outcomes From Wearable Data: Systematic Review From a Technology Readiness Level Point of View

BACKGROUND

Wearable technology has the potential to improve cardiovascular health monitoring by using machine learning. Such technology enables remote health monitoring and allows for the diagnosis and prevention of cardiovascular diseases. In addition to the detection of cardiovascular disease, it can exclude this diagnosis in symptomatic patients, thereby preventing unnecessary hospital visits. In addition, early warning systems can aid cardiologists in timely treatment and prevention.

OBJECTIVE

This study aims to systematically assess the literature on detecting and predicting outcomes of patients with cardiovascular diseases by using machine learning with data obtained from wearables to gain insights into the current state, challenges, and limitations of this technology.

METHODS

We searched PubMed, Scopus, and IEEE Xplore on September 26, 2020, with no restrictions on the publication date and by using keywords such as "wearables," "machine learning," and "cardiovascular disease." Methodologies were categorized and analyzed according to machine learning-based technology readiness levels (TRLs), which score studies on their potential to be deployed in an operational setting from 1 to 9 (most ready).

RESULTS

After the removal of duplicates, application of exclusion criteria, and full-text screening, 55 eligible studies were included in the analysis, covering a variety of cardiovascular diseases. We assessed the quality of the included studies and found that none of the studies were integrated into a health care system (TRL<6), prospective phase 2 and phase 3 trials were absent (TRL<7 and 8), and group cross-validation was rarely used. These issues limited these studies' ability to demonstrate the effectiveness of their methodologies. Furthermore, there seemed to be no agreement on the sample size needed to train these studies' models, the size of the observation window used to make predictions, how long participants should be observed, and the type of machine learning model that is suitable for predicting cardiovascular outcomes.

CONCLUSIONS

Although current studies show the potential of wearables to monitor cardiovascular events, their deployment as a diagnostic or prognostic cardiovascular clinical tool is hampered by the lack of a realistic data set and proper systematic and prospective evaluation.

Evaluation of Level-Crossing ADCs for Event-Driven ECG Classification

In this paper, a new methodology for choosing design parameters of level-crossing analog-to-digital converters (LC-ADCs) is presented that improves sampling accuracy and reduces the data stream rate. Using the MIT-BIH Arrhythmia dataset, several LC-ADC models are designed, simulated and then evaluated in terms of compression and signal-to-distortion ratio. A new one-dimensional convolutional neural network (1D-CNN) based classifier is presented. The 1D-CNN is used to evaluate the event-driven data from several LC-ADC models. With uniformly sampled data, the 1D-CNN has 99.49%, 92.4% and 94.78% overall accuracy, sensitivity and specificity, respectively. In comparison, a 7-bit LC-ADC with 2385 Hz clock frequency and 6-bit clock resolution offers 99.2%, 89.98% and 91.64% overall accuracy, sensitivity and specificity, respectively. It also offers 3x data compression while maintaining a signal-to-distortion ratio of 21.19 dB. Furthermore, it only requires 49% floating-point operations per second (FLOPS) for cardiac arrhythmia classification in comparison with the uniformly sampled ADC. Finally, an open-source event-driven arrhythmia database is presented.

ULECGNet: An Ultra-Lightweight End-to-End ECG Classification Neural Network

ECG classification is a key technology in intelligent ECG monitoring. In the past, traditional machine learning methods such as SVM and KNN have been used for ECG classification, but with limited classification accuracy. Recently, the end-to-end neural network has been used for the ECG classification and shows high classification accuracy. However, the end-to-end neural network has large computational complexity including a large number of parameters and operations. Although dedicated hardware such as FPGA and ASIC can be developed to accelerate the neural network, they result in large power consumption, large design cost, or limited flexibility. In this work, we have proposed an ultra-lightweight end-to-end ECG classification neural network which has extremely low computational complexity (~8.2k parameters & ~227k MUL/ADD operations) and can be squeezed into a low-cost MCU (i.e. microcontroller) while achieving 99.1% overall classification accuracy. This outperforms the state-of-the-art ECG classification neural network. Implemented on a low-cost MCU (i.e. MSP432), the proposed design consumes only 0.4 mJ and 3.1 mJ per heartbeat classification for normal and abnormal heartbeats respectively for real-time ECG classification.

A 1.2nW Analog Electrocardiogram Processor Achieving a 99.63% QRS Complex Detection Sensitivity

An energy-efficient electrocardiogram (ECG) processor for real-time QRS detection is presented. The proposed algorithm is based on the Pan-Tompkins algorithm and it is implemented in the analog domain leveraging ultra-low power analog electronics biased in subthreshold. Operational transconductance amplifiers with ∼100 mV linear range are used in almost all of the processing blocks, while squaring is performed on current signals. Additionally, instead of adaptive thresholding, a fixed-level thresholding is performed, thereby eliminating the need for additional blocks such as memory and threshold update. The processor is designed in 65 nm TSMC CMOS technology and has a footprint of 0.078 mm². When supplied by a 1 V supply, the processor consumes 1.2 nW. Using the recordings in the MIT-BIH database, the processor achieves an average QRS detection sensitivity of 99.63% and positive predictivity of 99.47%.

Development of Wearable Wireless Electrocardiogram Detection System using Bluetooth Low Energy

Wearable monitoring devices can provide patients and doctors with the capability to measure bio-signals on demand. These systems provide enormous benefits for people with acute symptoms of serious health conditions. In this paper, we propose a novel method for collecting ECG signals using two wireless wearable modules. The electric potential measured from a sub-module is transferred to the main module through Bluetooth Low Energy, and the collected values are simultaneously displayed in the form of a graph. This study describes the configuration and outcomes of the proposed system and discusses the important challenges associated with the functioning of the device. The proposed system had 84% signal similarity to that of other commercial products. As a band-type module was used on each wrist to check the signal, continuous observation of patients can be achieved without restricting their actions or causing discomfort.

Power Efficiency Comparison of Event-Driven and Fixed-Rate Signal Conversion and Compression for Biomedical Applications

Energy-constrained biomedical recording systems need power-efficient data converters and good signal compression in order to meet the stringent power consumption requirements of many applications. In literature today, typically a SAR ADC in combination with digital compression is used. Recently, alternative event-driven sampling techniques have been proposed that incorporate compression in the ADC, such as level-crossing A/D conversion. This paper describes the power efficiency analysis of such level-crossing ADC (LCADC) and the traditional fixed-rate SAR ADC with simple compression. A model for the power consumption of the LCADC is derived, which is then compared to the power consumption of the SAR ADC with zero-order hold (ZOH) compression for multiple biosignals (ECG, EMG, EEG, and EAP). The LCADC is more power efficient than the SAR ADC up to a cross-over point in quantizer resolution (for example 8 bits for an EEG signal). This cross-over point decreases with the ratio of the maximum to average slope in the signal of the application. It also changes with the technology and design techniques used. The LCADC is thus suited for low to medium resolution applications. In addition, the event-driven operation of an LCADC results in fewer data to be transmitted in a system application. The event-driven LCADC without timer and with single-bit quantizer achieves a reduction in power consumption at system level of two orders of magnitude, an order of magnitude better than the SAR ADC with ZOH compression. At system level, the LCADC thus offers a big advantage over the SAR ADC.