DiabDeep: Pervasive Diabetes Diagnosis Based on Wearable Medical Sensors and Efficient Neural Networks

Abstract

Diabetes impacts the quality of life of millions of people around the globe. However, diabetes diagnosis is still an arduous process, given that this disease develops and gets treated outside the clinic. The emergence of wearable medical sensors (WMSs) and machine learning points to a potential way forward to address this challenge. WMSs enable a continuous, yet user-transparent, mechanism to collect and analyze physiological signals. However, disease diagnosis based on WMS data and its effective deployment on resource-constrained edge devices remain challenging due to inefficient feature extraction and vast computation cost. To address these problems, we propose a framework called DiabDeep that combines efficient neural networks (called DiabNNs) with off-the-shelf WMSs for pervasive diabetes diagnosis. DiabDeep bypasses the feature extraction stage and acts directly on WMS data. It enables both an (i) accurate inference on the server, e.g., a desktop, and (ii) efficient inference on an edge device, e.g., a smartphone, to obtain a balance between accuracy and efficiency based on varying resource budgets and design goals. On the resource-rich server, we stack sparsely connected layers to deliver high accuracy. On the resource-scarce edge device, we use a hidden-layer long short-term memory based recurrent layer to substantially cut down on computation and storage costs while incurring only a minor accuracy loss. At the core of our system lies a grow-and-prune training flow: it leverages gradient-based growth and magnitude-based pruning algorithms to enable DiabNNs to learn both weights and connections, while improving accuracy and efficiency. We demonstrate the effectiveness of DiabDeep through a detailed analysis of data collected from 52 participants. For server (edge) side inference, we achieve a 96.3% (95.3%) accuracy in classifying diabetics against healthy individuals, and a 95.7% (94.6%) accuracy in distinguishing among type-1 diabetic, type-2 diabetic, and healthy individuals. Against conventional baselines, such as support vector machines with linear and radial basis function kernels, k-nearest neighbor, random forest, and linear ridge classifiers, DiabNNs achieve higher accuracy, while reducing the model size (floating-point operations) by up to 454.5× (8.9×). Therefore, the system can be viewed as pervasive and efficient, yet very accurate.

CovidDeep: SARS-CoV-2/COVID-19 Test Based on Wearable Medical Sensors and Efficient Neural Networks

“…We continue to collect more data across various countries for further validation of the CovidDeep models. The CovidDeep framework, alongside previous studies on diabetes diagnosis with the help of such sensors [28], gives us confidence that in the future WMS-based disease detection is feasible for a large number of diseases [27].…”

“…Hence, they enable constant monitoring of the user's health status. Training AI algorithms with data produced by WMSs can enable pervasive health condition tracking and disease onset detection [28]. This approach exploits the knowledge distillation capability of machine learning algorithms to directly extract information from physiological signals.…”

“…Stress-Lysis [45] is a DNN-based approach for monitoring stress based on three sources of data, namely, body temperature, rate of motion, and sweat during physical activity. DiabDeep enables detection of Type I/II diabetes using WMSs and DNNs [28]. Another application of smart healthcare is in the design of edge devices to accurately detect seizure episodes based on electroencephalography (EEG) signals [46], [47], [48].…”

Systematic Review on Machine-Learning Algorithms Used in Wearable-Based eHealth Data Analysis

“…Support vector machines are used for analysing the ECG and EEG data for diabetes classification [19], Random Forest algorithms are used for prediction tasks like diabetes or hypertension prediction [20] [21]. Neural network algorithms have been used for the prediction task for disease diagnosis using data from wearable medical sensors [22], multilayer perceptrons are used for classification of different types of diabetes and for behavioural analysis for Parkinson's disease [23] [3]. Similarly, artificial neural networks (ANN) are used for the analysis of EEG, Electromyogram (EMG), voice signals [24] [25].…”

Intelligent COVID-19 Forecasting, Diagnoses and Monitoring Systems: A Survey

“…Benefits of COVID-19 Diagnosing Systems: Machine learning algorithms can assist in the clinical decision making and have been used successfully in diagnosis of a number of health problems such as pathological brain detection [57] [58], breast cancer [59], lung cancer [60] [61], colon cancer [62] [63], prostate cancer [63], Alzheimer's disease [64], diabetes [65] and flu [66]. This entails the use of intelligent approaches that can automatically extract useful insights from the chest X-rays those are characteristics of COVID-19.…”