# Age-Related Changes in the Characteristics of the Elderly Females Using the Signal Features of an Earlobe Photoplethysmogram

^{1}

^{2}

^{3}

^{4}

^{5}

^{6}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Methods

#### 2.1. Data Collection

#### 2.2. Data Analysis

#### 2.2.1. Multiple Linear Regression

_{ab}, T

_{ac}, T

_{ad}, and T

_{ae}(the time intervals between the waves) and b/a, c/a, d/a, and e/a (the a, b, c, d, and e-wave ratios of SDPPG); aging index (AI: (b-c-d-e)/a); vascular aging variable [8]; and sum of TDPPG (∑TDPPG(×10

^{4})), which is the sum of the root mean square values of TDPPG. The regression analysis involved a stepwise method, and a model was developed with a dataset containing the information of 68 randomly selected subjects (approximately 80% of the 84 subjects). The test set comprised data from the remaining 16 subjects (corresponding to the remaining 20%), which were randomly extracted to verify the model.

#### 2.2.2. Fitting a Function with a Neural Network

_{ac}, AI, ∑TDPPG(×10

^{4}), four variables act as inputs to the neural network, and the real age is used as the target. A single hidden layer feed forward neural network (FFNN), with a comprising 15 neurons, was used for the neural network to learn age prediction (Figure 3). FFNN is a type of neural network that does not include connection loops between units. Information moves only in one direction from the input to the hidden layer; thus, no cycles or loops exist in the network. The error was calculated via back-propagation using the Levenberg–Marquardt weight. The FFNN model learned to minimize the error of the output value by iteratively updating the weights.

## 3. Results

_{ac}, AI, and sum of the TDPPG variables were significantly correlated with age (Table 2). Figure 4 demonstrates the first-order linear fitting of the four correlated variables with age, c/a, T

_{ac}, AI, and the sum of TDPPG.

^{2}value was 0.46 (p = 0.00). Table 3 presents the overall results of the model. The model developed using the predicted value is expressed using Equation (1): Figure 5 demonstrates the first-order linear fitting of the test set.

## 4. Discussion

^{2}value indicated a model accuracy of 0.46. The results obtained using the randomly extracted test set showed that r was slightly lowered to 0.556 (r

^{2}= 0.309), as shown in Figure 3. This means that AI and the sum of the TDPPG variables are improved when used independently.

## 5. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Mohanalakshmi, S.; Sivasubramanian, A.; Swarnalatha, A. Predicting arterial stiffness from physiological characteristics of photoplethysmography signals quantified through second derivative. Indian J. Sci. Technol.
**2017**, 10, 104111. [Google Scholar] [CrossRef] - Moraes, J.L.; Rocha, M.X.; Vasconcelos, G.G.; Filho, J.E.; Albuquerque, V.H.; Alexandria, A.R. Advances in photopletysmography signal analysis for biomedical applications. Sensors
**2018**, 18, 1894. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Inoue, N.; Kawakami, H.; Yamamoto, H.; Ito, C.; Fujiwara, S.; Sasaki, H.; Kihara, Y. Second derivative of the finger photoplethysmogram and cardiovascular mortality in middle-aged and elderly Japanese women. Hypertens. Res.
**2017**, 40, 207–211. [Google Scholar] [CrossRef] [PubMed] - Takada, H.; Washino, K.; Harrell, J.S.; Iwata, H. Acceleration plethysmography to evaluate aging effect in cardiovascular system. Using new criteria of four wave patterns. Clin. Trial
**1996**, 21, 205–210. [Google Scholar] - Hartmann, V.; Liu, H.; Chen, F.; Qiu, Q.; Hughes, S.; Zheng, D. Quantitative comparison of photoplethysmography waveform characteristics: Effect of measurement site. Front. Physiol.
**2019**, 10, 198. [Google Scholar] [CrossRef] [PubMed] - Allen, J.; Murray, A. Age-related changes in the characteristics of the photoplethysmographic pulse shape at various body sites. Physiol. Meas.
**2003**, 24, 297–307. [Google Scholar] [CrossRef] [PubMed] - Kawada, T.; Otsuka, T. Factor structure of indices of the second derivative of the finger photoplethysmogram with metabolic components and other cardiovascular risk indicators. Diabetes Metab. J.
**2013**, 37, 40–45. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Takazawa, K.; Fujita, N.T.; Matsuoka, O.; Aikawa, T.S.; Tamura, S.; Ibukiyama, C. Assessment of vasoactive agents and vascular aging by the second derivative of photoplethysmogram waveform. Hypertension
**1998**, 32, 365–370. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Perpetuini, D.; Chiarelli, A.M.; Cardone, D.; Rinella, S.; Massimino, S.; Bianco, F.; Bucciarelli, V.; Vinciguerra, V.; Fallica, G.; Perciavalle, V.; et al. Photoplethysmographic Prediction of the Ankle-Brachial Pressure Index through a Machine Learning Approach. Appl. Sci.
**2020**, 10, 2137. [Google Scholar] [CrossRef] [Green Version] - Dall’Olio, L.; Curti, N.; Remondini, D.; Safi Harb, Y.; Asselbergs, F.W.; Castellani, G.; Uh, H.-W. Prediction of vascular aging based on smartphone acquired PPG signals. Sci. Rep.
**2020**, 10, 19756. [Google Scholar] [CrossRef] [PubMed] - Chiarelli, A.M.; Bianco, F.; Perpetuini, D.; Bucciarelli, V.; Filippini, C.; Cardone, D.; Zappasodi, F.; Gallina, S.; Merla, A. Data-driven assessment of cardiovascular ageing through multisite photoplethysmography and electrocardiography. Med. Eng. Phys.
**2019**, 73, 39–50. [Google Scholar] [CrossRef] [PubMed] - Hsiu, H.; Liu, J.-C.; Yang, C.-J.; Chen, H.-S.; Wu, M.-S.; Hao, W.-R.; Lee, K.-Y.; Hu, C.-J.; Wang, Y.-H.; Fang, Y.-A. Discrimination of vascular aging using the arterial pulse spectrum and machine-learning analysis. Microvasc. Res.
**2022**, 139, 104240. [Google Scholar] [CrossRef] [PubMed] - Rinkevičius, M.; Kontaxis, S.; Gil, E.; Bailón, R.; Lázaro, J.; Laguna, P.; Marozas, V. Photoplethysmogram Signal Morphology-Based Stress Assessment. In Proceedings of the 2019 Computing in Cardiology (CinC), Singapore, 8–11 September 2019. [Google Scholar]
- Elgendi, M. On the analysis of fingertip photoplethysmogram signals. Curr. Cardiol. Rev.
**2012**, 8, 14–25. [Google Scholar] [CrossRef] [PubMed] - Yousef, Q.; Reaz, M.B.; Ali, M.A. The analysis of PPG morphology: Investigating the effects of aging on arterial compliance. Meas. Sci. Rev.
**2012**, 12, 266–271. [Google Scholar] [CrossRef] [Green Version] - Özkan, İ.; Kuvvetli, Y.; Bayka, P.D.; Erol, R. Comparison of the neural network model and linear regression model for predicting the intermingled yarn breaking strength and elongation. J. Text. Inst.
**2014**, 105, 1203–1211. [Google Scholar] [CrossRef] - Charlton, P.H.; Celka, P.; Farukh, B.; Chowienczyk, P.; Alastruey, J. Assessing mental stress from the photoplethysmogram: A numerical study. J. Physiol. Meas.
**2018**, 15, 054001. [Google Scholar] [CrossRef] [PubMed] - Hafifah, A.H.; Anuar, N.N. Methods of extracting feature from photoplethysmogram waveform for non-invasive diagnostic applications. Int. J. Online Biomed. Eng.
**2020**, 16, 39–62. [Google Scholar] [CrossRef] - Zahedi, E.; Chellappan, K.; Ali, M.A.; Singh, H. Analysis of the effect of ageing on rising edge characteristics of the photoplethysmogram using a modified windkessel model. Cardiovasc. Eng.
**2007**, 7, 172–181. [Google Scholar] [CrossRef] [PubMed] - Pilt, K.; Ferenets, R.; Meigas, K.; Lindberg, L.; Temitski, K.; Viigimaa, M. New photoplethysmographic signal analysis algorithm for arterial stiffness estimation. Sci. World J.
**2013**, 2013, 169035. [Google Scholar] [CrossRef] [PubMed] - Yousef, K.Q.; Rubins, U.; Mafawez, A. Photoplethysmogram second derivative review: Analysis and applications. Sci. Res. Essays
**2015**, 10, 633–639. [Google Scholar] [CrossRef] [Green Version]

Characteristic | Value |
---|---|

Age (year) | 71.19 ± 6.97 |

Weight (Kg) | 58.94 ± 9.33 |

Height (cm) | 153.73 ± 8.11 |

b/a | c/a | d/a | e/a | T_{ab} (ms) | T_{ac} (ms) | T_{ad} (ms) | AI | ∑TDPTG × 10^{4} | |
---|---|---|---|---|---|---|---|---|---|

Mean ± SD | −1.10 ± 0.07 | 0.14 ± 0.06 | 0.03 ± 0.06 | 0.21 ± 0.05 | 44.21 ± 2.57 | 88.21 ± 6.48 | 108.07 ± 5.03 | −1.49 ± 0.14 | 100.18 ± 0.05 |

r | 0.31 | −0.54 | 0.12 | 0.00 | −0.05 | 0.33 | 0.08 | 0.53 | −0.52 |

p | N.S | 0.00 * | N.S | N.S | N.S | 0.01 * | N.S | 0.00 * | 0.00 * |

Estimate | Standard Error (SE) | t-Statistics (Estimate/SE) | p | Adjusted-r^{2} | |
---|---|---|---|---|---|

Intercept | 109.65 | 6.56 | 16.73 | 0.00 | 0.46 |

AI | 22.16 | 4.46 | 4.97 | 0.00 | |

∑TDPTG × 10^{4} | −0.05 | 0.01 | −4.81 | 0.00 |

No. of Hidden Layer | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|

MSE | 38.98 | 27.25 | 47.91 | 18.67 | 20.11 | 19.61 | 31.48 | 47.44 | 28.77 | 21.42 | 153.92 | 13.13 | 40.98 | 30.01 | 16.02 | 45.51 | 52.75 | 39.33 | 60.17 | 70.98 |

Best epoch | 3 | 7 | 7 | 6 | 8 | 3 | 4 | 7 | 4 | 6 | 1 | 4 | 5 | 5 | 3 | 6 | 5 | 3 | 3 | 5 |

Training (r) | 0.67 | 0.77 | 0.75 | 0.85 | 0.88 | 0.79 | 0.83 | 0.89 | 0.83 | 0.86 | 0.31 | 0.84 | 0.94 | 0.78 | 0.86 | 0.89 | 0.87 | 0.75 | 0.92 | 0.59 |

Validation (r) | 0.55 | 0.71 | 0.74 | 0.75 | 0.41 | 0.76 | 0.51 | 0.65 | 0.68 | 0.79 | 0.39 | 0.91 | 0.78 | 0.43 | 0.71 | 0.64 | 0.23 | 0.58 | 0.58 | 0.21 |

Test (r) | 0.72 | 0.35 | 0.11 | 0.63 | 0.48 | 0.17 | 0.69 | 0.50 | 0.77 | 0.17 | 0.82 | 0.64 | 0.74 | 0.52 | 0.85 | 0.75 | 0.73 | 0.78 | 0.78 | 0.28 |

All (r) | 0.64 | 0.72 | 0.61 | 0.82 | 0.81 | 0.74 | 0.76 | 0.77 | 0.79 | 0.77 | 0.38 | 0.81 | 0.83 | 0.71 | 0.83 | 0.80 | 0.78 | 0.67 | 0.81 | 0.41 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Seo, J.-W.; Choi, J.; Lee, K.; Kim, J.U.
Age-Related Changes in the Characteristics of the Elderly Females Using the Signal Features of an Earlobe Photoplethysmogram. *Sensors* **2021**, *21*, 7782.
https://doi.org/10.3390/s21237782

**AMA Style**

Seo J-W, Choi J, Lee K, Kim JU.
Age-Related Changes in the Characteristics of the Elderly Females Using the Signal Features of an Earlobe Photoplethysmogram. *Sensors*. 2021; 21(23):7782.
https://doi.org/10.3390/s21237782

**Chicago/Turabian Style**

Seo, Jeong-Woo, Jungmi Choi, Kunho Lee, and Jaeuk U. Kim.
2021. "Age-Related Changes in the Characteristics of the Elderly Females Using the Signal Features of an Earlobe Photoplethysmogram" *Sensors* 21, no. 23: 7782.
https://doi.org/10.3390/s21237782