Novel Blood Pressure Waveform Reconstruction from Photoplethysmography using Cycle Generative Adversarial Networks

Mehrabadi, Milad Asgari; Aqajari, Seyed Amir Hossein; Zargari, Amir Hosein Afandizadeh; Dutt, Nikil; Rahmani, Amir M.

doi:10.1109/embc48229.2022.9871962

Cited by 22 publications

(6 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Generative Adversarial Networks (GANs) (Goodfellow et al 2014) have been widely used to translate ABP waveform from PPG signal (Mehrabadi et al 2022). Recent studies step forward by using more advanced backbones for adversarial training.…”

Section: Generative Waveform Transformationmentioning

confidence: 99%

“…Owing to the rapid development of deep learning techniques (Pan et al 2022;Bian et al 2021Bian et al , 2020Bi et al 2022), cutting-edge research in the field can be summarized into two categories, namely, estimation from blood pressure features (Miao et al 2019;Wu, Pang, and Kwong 2014) and waveform transformation (Mehrabadi et al 2022;Vardhan et al 2021;Ibtehaz et al 2022). Estimation from blood pressure features intends to predict systolic and diastolic blood pressure, but it inevitably discards the ABP waveform and results in the loss of cardiovascular disease details.…”

Section: Introductionmentioning

confidence: 99%

“…To comprehensively perceive the cardiac process (Mukkamala, Stergiou, and Avolio 2022) and measure the blood pressure from PPG, a precise mapping from the peripheral blood volume changes to the arterial blood pressure waveform is critical. Several studies attempted to address this bottleneck by applying deep learning-based techniques (e.g., convolutional networks (Brophy et al 2021), variational autoencoders (VAE) (Shimazaki et al 2018), Transformers (Chu et al 2023), and generative adversarial net- works (Mehrabadi et al 2022;Qin, Huang, and Zhang 2021)), as they have shown stronger representation learning capability than conventional analytical approaches (Shaltis, Reisner, and Asada 2006a). However, these methods highly rely on the consistency of waveform transformation.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Constraint Latent Space Matters: An Anti-anomalous Waveform Transformation Solution from Photoplethysmography to Arterial Blood Pressure

Bian,

Li,

et al. 2024

AAAI

View full text Add to dashboard Cite

Arterial blood pressure (ABP) holds substantial promise for proactive cardiovascular health management. Notwithstanding its potential, the invasive nature of ABP measurements confines their utility primarily to clinical environments, limiting their applicability for continuous monitoring beyond medical facilities. The conversion of photoplethysmography (PPG) signals into ABP equivalents has garnered significant attention due to its potential in revolutionizing cardiovascular disease management. Recent strides in PPG-to-ABP prediction encompass the integration of generative and discriminative models. Despite these advances, the efficacy of these models is curtailed by the latent space shift predicament, stemming from alterations in PPG data distribution across disparate hardware and individuals, potentially leading to distorted ABP waveforms. To tackle this problem, we present an innovative solution named the Latent Space Constraint Transformer (LSCT), leveraging a quantized codebook to yield robust latent spaces by employing multiple discretizing bases. To facilitate improved reconstruction, the Correlation-boosted Attention Module (CAM) is introduced to systematically query pertinent bases on a global scale. Furthermore, to enhance expressive capacity, we propose the Multi-Spectrum Enhancement Knowledge (MSEK), which fosters local information flow within the channels of latent code and provides additional embedding for reconstruction. Through comprehensive experimentation on both publicly available datasets and a private downstream task dataset, the proposed approach demonstrates noteworthy performance enhancements compared to existing methods. Extensive ablation studies further substantiate the effectiveness of each introduced module.

show abstract

Section: Generative Waveform Transformationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Constraint Latent Space Matters: An Anti-anomalous Waveform Transformation Solution from Photoplethysmography to Arterial Blood Pressure

Bian,

Li,

et al. 2024

AAAI

View full text Add to dashboard Cite

show abstract

“…Many scholars have tried to measure continuous BP waveform by PPG or ECG based on machine learning methods. [8][9][10] Landry et al 8 utilized nonlinear autoregressive models with exogenous input (NARX), took the PPG and ECG signals as the model inputs and obtained the corresponding BP waveform. Chen et al 9 adopted the structure of multiple gated recurrent unit (GRU) embedded in SENet, took PPG signals as the model inputs, and obtained the BP waveform of subjects.…”

Section: Introductionmentioning

confidence: 99%

“…Many scholars have tried to measure continuous BP waveform by PPG or ECG based on machine learning methods 8–10 . Landry et al 8 utilized nonlinear autoregressive models with exogenous input (NARX), took the PPG and ECG signals as the model inputs and obtained the corresponding BP waveform.…”

Section: Introductionmentioning

confidence: 99%

Non‐invasive measurement of blood pressure waveform based on spatiotemporal multi‐dimensional pulse waves

Lin,

Wang,

Zhang

et al. 2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Continuous monitoring of blood pressure (BP) waveform is essential for the prevention and diagnosis of cardiovascular diseases because it contains rich information. A vision‐based non‐invasive measurement method of BP waveform was proposed in this work. A binocular vision pulse acquisition system was utilized to collect continuous pulse images in a predetermined area of radial artery, and multi‐dimensional pulse waves were extracted from image sequences. By using Bi‐directional Long‐Short Term Memory (BiLSTM) networks, a model suitable for the inputs of multitype pulse waves under multi‐length windows has been established. The model can measure BP wave with high accuracy more than 0.94 (correlation coefficient). Measurement errors are less than (4.65 ± 5.80) mmHg for systolic blood pressure and (6.45 ± 7.40) mmHg for diastolic blood pressure. This study breaks through the limitation of obtaining BP from single‐dimensional pulse wave in previous studies and provides a new idea for non‐invasive measurement of human BP waveform.

show abstract

Comparison of Different Methods for Estimation of Arterial Blood Pressure Using PPG Signals

Mladenovska,

Ackovska,

Kostoska

et al. 2024

EAI/Springer Innovations in Communication and Computing

View full text Add to dashboard Cite

Novel Blood Pressure Waveform Reconstruction from Photoplethysmography using Cycle Generative Adversarial Networks

Cited by 22 publications

References 21 publications

Constraint Latent Space Matters: An Anti-anomalous Waveform Transformation Solution from Photoplethysmography to Arterial Blood Pressure

Constraint Latent Space Matters: An Anti-anomalous Waveform Transformation Solution from Photoplethysmography to Arterial Blood Pressure

Non‐invasive measurement of blood pressure waveform based on spatiotemporal multi‐dimensional pulse waves

Comparison of Different Methods for Estimation of Arterial Blood Pressure Using PPG Signals

Contact Info

Product

Resources

About