Multitask Residual Shrinkage Convolutional Neural Network for Sleep Apnea Detection Based on Wearable Bracelet Photoplethysmography

Qi, Shanshan; Yang, Xin; Zou, Lang; Wei, Keming; Wang, Changhong; Liu, Guanzheng

doi:10.1109/jiot.2022.3195777

Cited by 18 publications

(6 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, PPG signals obtained from commercial bracelets are analyzed, along with simultaneously collected PSG signals as the gold standard [ 24 , 25 ]. The PPG bracelet is an optical device that is often used for pulse rate monitoring based on the detection of blood volume changes in the microvascular bed of tissue.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Information-Based Similarity of Ordinal Pattern Sequences as a Novel Descriptor in Obstructive Sleep Apnea Screening Based on Wearable Photoplethysmography Bracelets

Chen

et al. 2022

Biosensors

Self Cite

View full text Add to dashboard Cite

Obstructive sleep apnea (OSA) is a common respiratory disorder associated with autonomic nervous system (ANS) dysfunction, resulting in abnormal heart rate variability (HRV). Capable of acquiring heart rate (HR) information with more convenience, wearable photoplethysmography (PPG) bracelets are proven to be a potential surrogate for electrocardiogram (ECG)-based devices. Meanwhile, bracelet-type PPG has been heavily marketed and widely accepted. This study aims to investigate the algorithm that can identify OSA with wearable devices. The information-based similarity of ordinal pattern sequences (OP_IBS), which is a modified version of the information-based similarity (IBS), has been proposed as a novel index to detect OSA based on wearable PPG signals. A total of 92 PPG recordings (29 normal subjects, 39 mild–moderate OSA subjects and 24 severe OSA subjects) were included in this study. OP_IBS along with classical indices were calculated. For severe OSA detection, the accuracy of OP_IBS was 85.9%, much higher than that of the low-frequency power to high-frequency power ratio (70.7%). The combination of OP_IBS, IBS, CV and LF/HF can achieve 91.3% accuracy, 91.0% sensitivity and 91.5% specificity. The performance of OP_IBS is significantly improved compared with our previous study based on the same database with the IBS method. In the Physionet database, OP_IBS also performed exceptionally well with an accuracy of 91.7%. This research shows that the OP_IBS method can access the HR dynamics of OSA subjects and help diagnose OSA in clinical environments.

show abstract

Section: Methodsmentioning

confidence: 99%

“…In this study, the same data as [ 24 , 25 ] are used. The PPG recordings were collected from wearable bracelets for analysis.…”

Section: Methodsmentioning

confidence: 99%

Information-Based Similarity of Ordinal Pattern Sequences as a Novel Descriptor in Obstructive Sleep Apnea Screening Based on Wearable Photoplethysmography Bracelets

Chen

et al. 2022

Biosensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this section, MMARSCNN [31] model is proposed for the Tennis Teaching Based on Data Mining. MMARSCNN is used to training of tennis in data mining.…”

Section: E Tennis Teaching Based On Data Mining Using Multitask Multi...mentioning

confidence: 99%

Multitask Multi attention Residual Shrinkage Convolutional Neural Network Personalized Assessment Model of Tennis Teaching Based on Data Mining and Intelligent Recommendation

Yang Liu

2024

jes

View full text Add to dashboard Cite

Tennis is highly explosive, continuous, intense sport that involves numerous continuous short-term explosive actions. This type of training is short-term, high-intensity, and focuses on competitive skills. Athletes must be physically fit and have long-term endurance to excel in technical and tactical skills during competition. In this manuscript, Multitask Multi attention Residual Shrinkage Convolutional Neural Network Personalized Assessment Model of Tennis Teaching Based on Data Mining and Intelligent Recommendation (MMARSCNN-PAM-TT-DM-IR) is proposed. The input data are collected from training levels of technical – sportive actions of 12-14 years old tennis players’ dataset. Initially the input data is preprocessed using Generalized Multi-kernel Maximum Correntropy Kalman Filter for exploring the data. The pre-processed data is given into Gradient Descent Namib Beetle Optimization (GDNBO) for choosing features from the dataset. Then, selected features are given to Dual Tree Complex Discrete Wavelet Transform to extract gray-scale statistical features such as Homogeneity, Entropy, Energy and Smoothness. Finally, the extracted feature attributes are given to MMARSCNN is used to Tennis Teaching Based on Data Mining. In general, MMARSCNN does not express some adaption of optimization strategies for determining optimal parameters to assure accurate Teaching of Tennis. Therefore, Zebra Optimization Algorithm (ZOA) is proposed to enhance weight parameter of MMARSCNN, which precisely Teaching of tennis depend on Data Mining. The proposed model is implemented, its efficacy is assessed utilizing some performance metrics such as accuracy, precision, F1-score, MSE, AUC. The proposed MMARSCNN-PAM-TT-DM-IR method provides 22.55%, 24.72% and 29.63% higher accuracy; 32.66%, 34.97% and 29.57% higher precision; 25.18%, 21.52% and 28.68% higher F1-Score is compared with existing method such as analysis characteristics of Tennis singles matches depend on 5G with data mining technology (AOC-TSM-DM), Tennis online teaching information platform depend on android mobile intelligent terminal (AMIT-TTP-DM), optimization analysis of Tennis players’ physical fitness index depend on data mining with mobile computing (OA-TP-DM-MC) respectively.

show abstract

“…DNNs with fusion techniques are often used in signal classification, such as electroencephalography [39], electrocardiogram [40], [41], and photoplethysmography [42]. However, to address variability and degradation within the Doppler signatures (see definition in Section V-A) obtained experimentally, we proposed a novel multi-stream, feature resonated-fusion network, termed eMSFRNet.…”

Section: Proposed Network Designmentioning

confidence: 99%

An Indoor Fall Monitoring System: Robust, Multistatic Radar Sensing and Explainable, Feature-Resonated Deep Neural Network

Shen

Tsui

Nussbaum

et al. 2023

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Indoor fall monitoring is challenging for community-dwelling older adults due to the need for high accuracy and privacy concerns. Doppler radar is promising, given its low cost and contactless sensing mechanism. However, the line-of-sight restriction limits the application of radar sensing in practice, as the Doppler signature will vary when the sensing angle changes, and signal strength will be substantially degraded with large aspect angles. Additionally, the similarity of the Doppler signatures among different fall types makes it extremely challenging for classification. To address these problems, in this paper we first present a comprehensive experimental study to obtain Doppler radar signals under large and arbitrary aspect angles for diverse types of simulated falls and daily living activities. We then develop a novel, explainable, multi-stream, feature-resonated neural network (eMSFRNet) that achieves fall detection and a pioneering study of classifying seven fall types. eMSFRNet is robust to both radar sensing angles and subjects. It is also the first method that can resonate and enhance feature information from noisy/weak Doppler signatures. The multiple feature extractors -including partial pre-trained layers from ResNet, DenseNet, and VGGNet -extracts diverse feature information with various spatial abstractions from a pair of Doppler signals. The feature-resonated-fusion design translates the multi-stream features to a single salient feature that is critical to fall detection and classification. eMSFRNet achieved 99.3% accuracy detecting falls and 76.8% accuracy for classifying seven fall types. Our work is the first effective multistatic robust sensing system that overcomes the challenges associated with Doppler signatures under large and arbitrary aspect angles, via our comprehensible feature-resonated deep neural network. Our work also demonstrates the potential to accommodate different radar monitoring tasks that demand precise and robust sensing.

show abstract

Multitask Residual Shrinkage Convolutional Neural Network for Sleep Apnea Detection Based on Wearable Bracelet Photoplethysmography

Cited by 18 publications

References 67 publications

Information-Based Similarity of Ordinal Pattern Sequences as a Novel Descriptor in Obstructive Sleep Apnea Screening Based on Wearable Photoplethysmography Bracelets

Information-Based Similarity of Ordinal Pattern Sequences as a Novel Descriptor in Obstructive Sleep Apnea Screening Based on Wearable Photoplethysmography Bracelets

Multitask Multi attention Residual Shrinkage Convolutional Neural Network Personalized Assessment Model of Tennis Teaching Based on Data Mining and Intelligent Recommendation

An Indoor Fall Monitoring System: Robust, Multistatic Radar Sensing and Explainable, Feature-Resonated Deep Neural Network

Contact Info

Product

Resources

About