Recognition of Abnormal Chest Compression Depth Using One-Dimensional Convolutional Neural Networks

When the displacement of an object is evaluated using sensor data, its movement back to the starting point can be used to correct the measurement error of the sensor. In medicine, the movements of chest compressions also involve a reciprocating movement back to the starting point. The traditional method of evaluating the effects of chest compression depth (CCD) is to use an acceleration sensor or gyroscope to obtain chest compression movement data; from these data, the displacement value can be calculated and the CCD effect evaluated. However, this evaluation procedure suffers from sensor errors and environmental interference, limiting its applicability. Our objective is to reduce the auxiliary computing devices employed for CCD effectiveness evaluation and improve the accuracy of the evaluation results. To this end, we propose a one-dimensional convolutional neural network (1D-CNN) classification method. First, we use the chest compression evaluation criterion to classify the pre-collected sensor signal data, from which the proposed 1D-CNN model learns classification features. After training, the model is used to classify and evaluate sensor signal data instead of distance measurements; this effectively avoids the influence of pressure occlusion and electromagnetic waves. We collect and label 937 valid CCD results from an emergency care simulator. In addition, the proposed 1D-CNN structure is experimentally evaluated and compared against other CNN models and support vector machines. The results show that after sufficient training, the proposed 1D-CNN model can recognize the CCD results with an accuracy rate of more than 95%. The execution time suggests that the model balances accuracy and hardware requirements and can be embedded in portable devices.

[1]  Pu Wang,et al.  LSTM-Based Auto-Encoder Model for ECG Arrhythmias Classification , 2020, IEEE Transactions on Instrumentation and Measurement.

[2]  Gavin D Perkins,et al.  Compression feedback devices over estimate chest compression depth when performed on a bed. , 2009, Resuscitation.

[3]  Benoit Gosselin,et al.  A Fully Embedded Adaptive Real-Time Hand Gesture Classifier Leveraging HD-sEMG and Deep Learning , 2019, IEEE Transactions on Biomedical Circuits and Systems.

[4]  Yuejin Zhao,et al.  Feature extraction and classification of heart sound using 1D convolutional neural networks , 2019, EURASIP Journal on Advances in Signal Processing.

[5]  Behrooz Lotfi,et al.  An approach for velocity and position estimation through acceleration measurements , 2016 .

[6]  Guillaume-Alexandre Bilodeau,et al.  Deep 1D-Convnet for accurate Parkinson disease detection and severity prediction from gait , 2019, Expert Syst. Appl..

[7]  Patrick Cardinal,et al.  End-to-End Environmental Sound Classification using a 1D Convolutional Neural Network , 2019, Expert Syst. Appl..

[8]  Paavo Alku,et al.  Analysis and Detection of Pathological Voice Using Glottal Source Features , 2020, IEEE Journal of Selected Topics in Signal Processing.

[9]  Yongqi Chang,et al.  A Fault Diagnosis Method of Rotating Machinery Based on One-Dimensional, Self-Normalizing Convolutional Neural Networks , 2020, Sensors.

[10]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[11]  Ahmet Kondoz,et al.  Radio Frequency Traffic Classification Over WLAN , 2017, IEEE/ACM Transactions on Networking.

[12]  Rodney G. Vaughan,et al.  Deep Learning Radar Design for Breathing and Fall Detection , 2020, IEEE Sensors Journal.

[13]  C Scheffer,et al.  A modeling approach to the effects of force guided versus depth guided compression during cardiopulmonary resuscitation on different chests and back support surfaces. , 2013, Resuscitation.

[14]  Mubarak Shah,et al.  Norm-Preservation: Why Residual Networks Can Become Extremely Deep? , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Pierre Jean Arnoux,et al.  Using an inertial navigation algorithm and accelerometer to monitor chest compression depth during cardiopulmonary resuscitation. , 2016, Medical engineering & physics.

[16]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[17]  Yongfeng Huang,et al.  CSFL: A novel unsupervised convolution neural network approach for visual pattern classification , 2017, AI Communications.

[18]  Ichinosuke Maeda,et al.  Real-time feedback of chest compressions using a flexible pressure sensor. , 2016, Resuscitation.

[19]  Stan Szpakowicz,et al.  Beyond Accuracy, F-Score and ROC: A Family of Discriminant Measures for Performance Evaluation , 2006, Australian Conference on Artificial Intelligence.

[20]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[21]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[22]  W. Haverkamp,et al.  Performance of chest compressions with the use of a new audio-visual feedback device: a randomized manikin study in health care professionals. , 2015, Resuscitation.

[23]  Paavo Alku,et al.  Glottal Source Information for Pathological Voice Detection , 2020, IEEE Access.

[24]  Min Hong,et al.  Deep Learning in Physiological Signal Data: A Survey , 2020, Sensors.

[25]  Yeongtak Song,et al.  A new chest compression depth feedback algorithm for high-quality CPR based on smartphone. , 2015, Telemedicine journal and e-health : the official journal of the American Telemedicine Association.

[26]  Tae Wook Kim,et al.  Novel Chest Compression Depth Measurement Sensor Using IR-UWB for Improving Quality of Cardiopulmonary Resuscitation , 2017, IEEE Sensors Journal.