Answers are grounded in this paper and cite their sources.
This article proposes a new method for identifying photoplethysmography (PPG) data, a non-invasive blood pressure (BP) measurement.Individual characteristic inconsistencies hinder PPG signal feature extraction.An innovative method for extracting PPG characteristics using continuous wavelet transform (CWT) and transfer learning is presented in this research.Pre-trained deep learning models like InceptionV3, VGG-16, and ResNet101 use robust characteristics to assess BP severity.The major purpose of this approach is to classify blood pressure values into average, pre-hypertension, and hypertension stages 1 and 2. The proposed method was trained on 219 individuals' PPG datasets.At the same time, InceptionV3, VGG-16, and ResNet101 had 99.5%, 22.5%, and 92.5% accuracy.Despite that, the study's findings reveal classifier behavior and how to improve the model's assessment metrics by utilizing category variances and modifying deep-learning network parameters.The proposed method for noninvasive blood pressure assessment utilizing PPG signals appears promising.Based on these findings, non-invasive blood pressure monitoring systems can be improved.
No takes yet. Share an insight, caveat, or question.
Fahoum et al. (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: