Search

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.

With advances in circuit design and sensing technology, the acquisition of data from a large number of Internet of Things (IoT) sensors simultaneously to enable more accurate inferences has become mainstream. In this work, we propose a novel convolutional neural network (CNN) model for the fusion of multimodal and multiresolution data obtained from several sensors. The proposed model enables the fusion of multiresolution sensor data, without having to resort to padding/ resampling to correct for frequency resolution differences even when carrying out temporal inferences like high-resolution event detection. The performance of the proposed model is evaluated for sleep apnea event detection, by fusing three different sensor signals obtained from UCD St. Vincent University Hospital's sleep apnea database. The proposed model is generalizable and this is demonstrated by incremental performance improvements, proportional to the number of sensors used for fusion. A selective dropout technique is used to prevent overfitting of the model to any specific high-resolution input, and increase the robustness of fusion to signal corruption from any sensor source. A fusion model with electrocardiogram (ECG), Peripheral oxygen saturation signal (SpO2), and abdominal movement signal achieved an accuracy of 99.72% and a sensitivity of 98.98%. Energy per classification of the proposed fusion model was estimated to be approximately 5.61 μJ for on-chip implementation. The feasibility of pruning to reduce the complexity of the fusion models was also studied.

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.

This paper presents a comparison of several artificial intelligence (AI) based binary classifiers for detecting the integrity of data obtained from Internet of Things (IoT) enabled wearable sensors. Detecting the integrity of data at the network edge facilitates the elimination of corrupted or unusable data, which translates to a lower amount of data stored and transmitted. This reduces the storage and power requirements of IoT devices without a reduction in functionality. In this work, we explore several machine learning-based classifiers to check the integrity of electrocardiogram (ECG) data. The feature vectors are derived from low complexity kurtosis and skewness based Signal Quality Indices (SQIs). From the experiments, it is found that a bagged ensemble of 3 neural networks achieves the highest detection accuracy of 99.47%. We also estimated the complexity and power consumed by the various classifier implementations and classifier fusion implementations. The energy consumed by the ensemble classifier was estimated to be around 0.039 nJ.

Stroke is widely considered as the second most common cause of mortality. The adverse consequences of stroke have led to global interest and work for improving the management and diagnosis of stroke. Various techniques for data mining have been used globally for accurate prediction of occurrence of stroke based on the risk factors that are associated with the electronic health care records (EHRs) of the patients. In particular, EHRs routinely contain several thousands of features and most of them are redundant and irrelevant that need to be discarded to enhance the prediction accuracy. The choice of feature-selection methods can help in improving the prediction accuracy of the model and efficient data management of the archived input features. In this paper, we systematically analyze the various features in EHR records for the detection of stroke. We propose a novel rough-set based technique for ranking the importance of the various EHR records in detecting stroke. Unlike the conventional rough-set techniques, our proposed technique can be applied on any dataset that comprises binary feature sets. We evaluated our proposed method in a publicly available dataset of EHR, and concluded that age, average glucose level, heart disease, and hypertension were the most essential attributes for detecting stroke in patients. Furthermore, we benchmarked the proposed technique with other popular feature-selection techniques. We obtained the best performance in ranking the importance of individual features in detecting stroke., Accepted in IEEE Access, 2020

This paper aims to reduce the power consumption of electrocardiography based wearable healthcare devices, by introducing power reduction approaches and considerations at system level design, where we have the highest potential to influence power. It focuses, in particular, on algorithm design and implementation, data acquisition, and transmission under constrained resources. A thorough investigation of the suitability of nine existing algorithms for on-sensor QRS feature detection is conducted, with respect to metrics such as sensitivity, positive predictivity, power consumption, parameter choice and time delay. Optimisation of data acquisition on CPU-based IoT systems is performed, and the current consumption is reduced by a factor of 3 using a combination of direct memory access (DMA) list approach and low-level register manipulations for task delegation. The acquisition data rate, sampling rate, buffer and batch size are also optimised. To reduce the power consumption by data transmission, the effect of on-sensor versus off-sensor processing is investigated. While focusing on CPU-based systems with experiments performed on a generic low-power wearable platform, the design optimisation and considerations proposed in this work could be extended to custom designs and allow further investigation into QRS detection algorithm optimisation for wearable devices.