BePCon: A Photoplethysmography-based Quality-aware Continuous Beat-to-Beat Blood Pressure Measurement Technique Using Deep Learning

• Published in: IEEE transactions on instrumentation and measurement Vol. 71; p. 1
• Main Authors: Roy, Monalisa Singha, Gupta, Rajarshi, Sharma, Kaushik Das
• Format: Journal Article
• Language: English
• Published: New York IEEE 2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Biomedical monitoring; Blood Pressure; Current measurement; Deep learning; Electrocardiography; Feature extraction; Hypertension; Measurement techniques; MIMICs; Monitoring; Photoplethysmogram (PPG); Pressure measurement; Quality assessment; Random access memory; Recursive Feature Elimination; Self organizing maps; Signal quality; Stack denoising autoencoder; Temporal Convolutional Neural Network; Time measurement; Unsupervised Learning; Waveforms
• ISSN: 0018-9456; 1557-9662
• DOI: 10.1109/TIM.2022.3212750

Research on noninvasive blood pressure (NIBP) measurement using electrocardiogram (ECG)/ photoplethysmogram (PPG) and their combinations have been most popular in ambulatory health monitoring. The real challenge is motion artifact (MA) corruption in the PPG, which makes the BP measurement unreliable. This paper presents BePCon, a deep learning-based model for beat-to-beat (BtB) BP measurement using a temporal convolutional network (TCN). Initially, the signal quality assessment (SQA) of PPG is done by a self-organizing map (SOM). Next, the time-domain, statistical, wavelet and stacked autoencoder features from current and previous good quality PPG cycles are extracted. A recursive feature elimination (RFE) selects optimum set of 20 features from each cycle before being fed to the TCN to predict the systolic (SBP) and diastolic blood pressure (DBP) of current beat. While evaluated over 150 data records from PhysioNet MIMIC-II/III waveform database, BePCon achieves standard deviation (SD), and mean absolute error (MAE) of 3.24 mmHg and 2.38 mmHg respectively for SBP, and 1.73 mmHg, and 1.23 mmHg respectively for the DBP respectively. An improvement of accuracy by a factor of 19.56% for SBP and 24.61% for DBP is obtained over without SQA. BePCon also complies with AAMI and BHS Grade A standard and improvement over published works on BtB BP measurement using MIMIC-II/III waveform database. A standalone implementation with a single core 1-GHz ARM v6 controller supported by 512 MB RAM shows low latency (~2.5 s per beat) and low memory requirement (~32.22 KB per beat). This establishes that BePCon has the potential for real-time ambulatory BtB BP measurement.