Beat-to-Beat Continuous Blood Pressure Estimation Using Bidirectional Long Short-Term Memory Network

Dongseok Lee; Hyunbin Kwon; Dongyeon Son; Heesang Eom; Cheolsoo Park; Yonggyu Lim; Chulhun Seo; Kwangsuk Park

doi:10.3390/s21010096

December 25, 2020SensorsOpen Access

Beat-to-Beat Continuous Blood Pressure Estimation Using Bidirectional Long Short-Term Memory Network

View Full Paper

Ask AI

Bookmark

Share

Why the study?

Continuous cuff-based blood pressure measurement can be cumbersome for hypertensive patients, requiring cuffless deep-learning estimation models that generalize well and maintain reproducibility across multi-day scenarios.

Does a bidirectional long short-term memory network using ECG, PPG, and BCG features accurately estimate continuous blood pressure?

Structured PICO

Does a bidirectional long short-term memory network using ECG, PPG, and BCG features accurately estimate continuous blood pressure?

P

Population

Subjects for continuous blood pressure estimation (specific demographics and sample size not provided in abstract)

I

Intervention

Cuffless blood pressure estimation model using a bidirectional long short-term memory network with features extracted from electrocardiogram, photoplethysmogram, and ballistocardiogram

O

Outcome

Mean absolute error (MAE) for systolic and diastolic blood pressuresurrogate

A bidirectional long short-term memory network using ECG, PPG, and BCG features provides accurate cuffless blood pressure estimation, with acceptable error rates in both one-day and multi-day scenarios.

Implication

Suggests viable cuffless BP estimation; leaves open prospective validation before clinical use.

Abstract

Continuous blood pressure (BP) monitoring is important for patients with hypertension. However, BP measurement with a cuff may be cumbersome for the patient. To overcome this limitation, various studies have suggested cuffless BP estimation models using deep learning algorithms. A generalized model should be considered to decrease the training time, and the model reproducibility should be taken into account in multi-day scenarios. In this study, a BP estimation model with a bidirectional long short-term memory network is proposed. The features are extracted from the electrocardiogram, photoplethysmogram, and ballistocardiogram. The leave-one-subject-out (LOSO) method is incorporated to generalize the model and fine-tuning is applied. The model was evaluated using one-day and multi-day tests. The proposed model achieved a mean absolute error (MAE) of 2.56 and 2.05 mmHg for the systolic and diastolic BP (SBP and DBP), respectively, in the one-day test. Moreover, the results demonstrated that the LOSO method with fine-tuning was more compatible in the multi-day test. The MAE values of the model were 5.82 and 5.24 mmHg for the SBP and DBP, respectively.