PPGMotion: Model-based detection of motion artifacts in photoplethysmography signals

Maity, Akash Kumar; Veeraraghavan, Ashok; Sabharwal, Ashutosh

doi:10.1016/j.bspc.2022.103632

Cited by 12 publications

(8 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MAs are probably the most common issue in PPG analysis [ 1 ]. With PPG sensors being extremely sensitive to even the slightest movement, and the MA spectrum overlapping with the HR spectrum, MA detection and removal has been a field of research in recent years, unfortunately with no satisfactory results [ 7 , 12 , 16 , 17 , 18 , 19 , 21 ]. This fact significantly complicates the task of maintaining consistently clean PPG signals, especially concerning endeavors such as estimating blood pressure from PPG data, which is a popular area of investigation [ 35 ].…”

Section: Discussionmentioning

confidence: 99%

“…Even in the case of MA detection without the use of additional signals, algorithms are limited either from the noise type that they can reliably detect or from the MAs’ usability. A recent study has been able to detect MAs without the use of additional signals or complex methodology [ 7 ]. Despite the fact that performance was very high for aperiodic signals, the use is limited to only detecting the artifacts, and there is no information of how the corrupted signal can be utilized.…”

Section: Discussionmentioning

confidence: 99%

“…PPG devices are compact and non-invasive, while one of their biggest advantages is allowing freedom of movement during data collection, hence being ideal for remote monitoring [ 1 , 2 , 5 , 6 ]. Key health parameters frequently monitored with PPG include heart-rate (HR), HR variability (HRV), oxyhemoglobin saturation, blood pressure, and respiratory rate analysis [ 4 , 5 , 7 , 8 ].…”

Section: Introductionmentioning

confidence: 99%

“…Notwithstanding, wavelet decomposition or comb filtering might be more efficient preprocessing methods leading to a higher signal-to-noise ratio (SNR) [ 14 , 15 ]. It is worth noting that the frequency spectrum of MAs can overlap with HR frequencies, necessitating additional processing for their detection and removal, which remains a challenging task [ 4 , 5 , 7 , 12 , 13 , 16 ].…”

Section: Introductionmentioning

confidence: 99%

“…The final category is actually a medley of variable methods other than the ones described above [ 12 ]. This includes classic spectral analysis, when the frequency spectrum of the noisy source does not overlap with the HR frequency, an analysis of statistical or morphological parameters of PPG signals, or machine learning techniques [ 7 , 12 , 21 , 22 , 23 ]. Besides the unsuitability of some of those methods when MAs are the main source of noise, they often require additional signals as references [ 21 , 22 , 23 ].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Novel Signal Restoration Method of Noisy Photoplethysmograms for Uninterrupted Health Monitoring

Vraka,

Zangróniz,

Quesada

et al. 2023

Sensors

View full text Add to dashboard Cite

Health-tracking from photoplethysmography (PPG) signals is significantly hindered by motion artifacts (MAs). Although many algorithms exist to detect MAs, the corrupted signal often remains unexploited. This work introduces a novel method able to reconstruct noisy PPGs and facilitate uninterrupted health monitoring. The algorithm starts with spectral-based MA detection, followed by signal reconstruction by using the morphological and heart-rate variability information from the clean segments adjacent to noise. The algorithm was tested on (a) 30 noisy PPGs of a maximum 20 s noise duration and (b) 28 originally clean PPGs, after noise addition (2–120 s) (1) with and (2) without cancellation of the corresponding clean segment. Sampling frequency was 250 Hz after resampling. Noise detection was evaluated by means of accuracy, sensitivity, and specificity. For the evaluation of signal reconstruction, the heart-rate (HR) was compared via Pearson correlation (PC) and absolute error (a) between ECGs and reconstructed PPGs and (b) between original and reconstructed PPGs. Bland-Altman (BA) analysis for the differences in HR estimation on original and reconstructed segments of (b) was also performed. Noise detection accuracy was 90.91% for (a) and 99.38–100% for (b). For the PPG reconstruction, HR showed 99.31% correlation in (a) and >90% for all noise lengths in (b). Mean absolute error was 1.59 bpm for (a) and 1.26–1.82 bpm for (b). BA analysis indicated that, in most cases, 90% or more of the recordings fall within the confidence interval, regardless of the noise length. Optimal performance is achieved even for signals of noise up to 2 min, allowing for the utilization and further analysis of recordings that would otherwise be discarded. Thereby, the algorithm can be implemented in monitoring devices, assisting in uninterrupted health-tracking.

show abstract