| 研究実績の概要 |
The photoplethysmogram (PPG) extracted from the video (imaging PPG, iPPG) has great potential for applications in the health monitoring area. However, the iPPG unavoidably contains a significant amount of noise due to the surrounding light interference, equipment noises, subject movements, etc., which limits iPPG signal applicability. This makes noise filtering an important issue. However, due to the complexity of the PPG dynamics, the choice of the appropriate filter is a challenging task. Therefore during the research period (FY2019), various PPG filtration techniques and its parameters were studied for the contact type green light PPG signals. Methods of nonlinear time series analysis, including the possible attraction reconstruction, recurrence plot and its quantitative analysis, predictability and the largest Lyapunov exponent estimation, were applied to the raw and filtered PPG data to investigate the effect of these filters on the PPG dynamics. A comparison of the results obtained from the analysis of differently processed time series demonstrated that filters conventionally applied to the PPG can significantly alter its dynamics. Thus qualitative and quantitative results have demonstrated shifts in the chaotic dynamics of original data towards regularity in filtered PPG signals. Based on the obtained dynamical characteristics filters with minimized impact on the PPG's chaotic dynamics were identified. This contributes to further applied studies on both PPG and iPPG signals.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
3: やや遅れている
理由
Effect of filtration on complex photoplethysmogram (PPG) dynamics was investigated on a large number of reference PPG datasets. Qualitative and quantitative changes in dynamical characteristics were successfully obtained and analyzed. However, due to the extended hospitalization of the principal investigator, processing and analysis of data collected by the video camera was slightly delayed.
|
