What does this research mean for the field?

Deep neural network models, specifically ResNet, can accurately classify four types of ECG waveforms measured via IoT prototypes, achieving an accuracy of 0.97 and a kappa statistic of 0.96. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

June 8, 2021BiosensorsOpen Access

Integrating ECG Monitoring and Classification via IoT and Deep Neural Networks

Q: What is the clinical evidence from this study?

Study design: Other. Population: Arrhythmia (n=48). Intervention: Deep Neural Network (ResNet) via IoT vs. AlexNet and SqueezeNet. Primary outcome: Accuracy of ECG waveform classification.

View Full Paper

Ask AI

Bookmark

Key result

ResNet deep neural network classifies ECG waveforms with ~97% accuracy across four categories.

n=48

Why the study?

Interpreting ECG signals during anesthesia assessment is difficult, and even experienced physicians can misjudge them.

Can deep neural network models combined with an IoT ECG prototype accurately classify ECG signals into four categories?

Comparison

ResNet vs AlexNet vs SqueezeNet deep neural network models

Authors

LYLi-Ren YehE-Da Hospital WCWei-Chin ChenI-Shou University HCHua-Yan ChanI-Shou University

Discussion

Member takes

Implication

Supports IoT-ECG AI development for anesthesia monitoring; leaves open need for prospective validation before clinical adoption.

Structured PICO

Can deep neural network models combined with an IoT ECG prototype accurately classify ECG signals into four categories?

Population

ECG signals from 48 recordings in the MIT-BIH Arrhythmia Database were used to train and test deep neural network models for classifying four ECG patterns.

Intervention

Deep neural network models (ResNet, AlexNet, SqueezeNet) combined with an IoT-based ECG measurement prototype.

Comparator

Comparison between different deep neural network models (ResNet, AlexNet, SqueezeNet) and clinical ECG (Philips IntelliVue MP70) for the prototype.

Outcome

Accuracy and kappa statistics of ResNet, AlexNet, and SqueezeNet in ECG waveform classificationsurrogate

Main Result

Absolute Event Rate: 0.97% vs 0.96%

An IoT-based ECG monitoring system combined with a ResNet deep learning model can accurately classify ECG signals into four categories with 97% accuracy, potentially assisting in real-time anesthesia assessment.

Limitations

The presented device can be applied to a patient with a resting heart rate of 60-100 bpm, but it may not be accurate with arrhythmia or tachycardia.
The prototype may not replace current laboratory or hospital equipment.

Cite This Study

Yeh et al. (2021) studied Arrhythmia (n=48). Deep Neural Network (ResNet) via IoT vs. AlexNet and SqueezeNet was evaluated on Accuracy of ECG waveform classification. The ResNet deep neural network model achieved an accuracy of 0.97 and a kappa statistic of 0.96 in classifying ECG waveforms into four categories.

synapsesocial.com/papers/6a1908c6a6ebdba58ad31913 https://doi.org/10.3390/bios11060188

View Full Paper

Ask AI

Bookmark

Integrating ECG Monitoring and Classification via IoT and Deep Neural Networks

Discussion

Member takes

Implication

Structured PICO

Main Result

Limitations

Cite This Study

Integrating ECG Monitoring and Classification via IoT and Deep Neural Networks

Discussion

Member takes

Implication

Structured PICO

Main Result

Limitations

Cite This Study

Also Consider

Also Consider