Estimating Continuous Blood Pressure from a Smartphone PPG Using 1D U–Net

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26220%2F26%3A0191348" target="_blank" >RIV/00216305:26220/26:0191348 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Estimating Continuous Blood Pressure from a Smartphone PPG Using 1D U–Net

Popis výsledku v původním jazyce

The development of new methods for non-invasive blood pressure (BP) monitoring is nowadays moving towards accessible and inexpensive cuffless devices (e.g. smartphone and smartwatch). This paper presents an experimental cuff-less measurement of BP using a smartphone. A photoplethysmographic (PPG) signal measured by a smartphone camera is used to estimate continous BP. Reference continuous BP was captured using Finapres NOVA device based on Penaz principle. This paper verifies the effect of input data personalization on the accuracy of BP estimation using 1D U-Net. Personalization consists in adding the information about age, gender, weight and height of the subject to the PPG signal. The study involves 56 measured subjects. The output of U-Net is continuous BP waveform, which accuracy is compared with reference using Mean Absolute Error (MAE). Personalized approach shows better results MAE = 8.28±6.29 mmHg than non-personalized approach with raw PPGs inputs where MAE = 10.06±7.12 mmHg. In the subsequent estimation of systolic BP (SBP) for the personalized U-Net input, the MAE was 8.45±5.21 mmHg, and in the estimation of diastolic BP (DBP), the MAE was 6.13±4.61 mmHg. Proposed method of non-invasive cuffless smartphone-based is promising and will be further developed to enable it is general and widespread use.

Název v anglickém jazyce

Estimating Continuous Blood Pressure from a Smartphone PPG Using 1D U–Net

Popis výsledku anglicky

The development of new methods for non-invasive blood pressure (BP) monitoring is nowadays moving towards accessible and inexpensive cuffless devices (e.g. smartphone and smartwatch). This paper presents an experimental cuff-less measurement of BP using a smartphone. A photoplethysmographic (PPG) signal measured by a smartphone camera is used to estimate continous BP. Reference continuous BP was captured using Finapres NOVA device based on Penaz principle. This paper verifies the effect of input data personalization on the accuracy of BP estimation using 1D U-Net. Personalization consists in adding the information about age, gender, weight and height of the subject to the PPG signal. The study involves 56 measured subjects. The output of U-Net is continuous BP waveform, which accuracy is compared with reference using Mean Absolute Error (MAE). Personalized approach shows better results MAE = 8.28±6.29 mmHg than non-personalized approach with raw PPGs inputs where MAE = 10.06±7.12 mmHg. In the subsequent estimation of systolic BP (SBP) for the personalized U-Net input, the MAE was 8.45±5.21 mmHg, and in the estimation of diastolic BP (DBP), the MAE was 6.13±4.61 mmHg. Proposed method of non-invasive cuffless smartphone-based is promising and will be further developed to enable it is general and widespread use.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20601 - Medical engineering

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Computing in Cardiology 2024

ISBN

—

ISSN

—

e-ISSN

2325-887X

Počet stran výsledku

4

Strana od-do

—

Název nakladatele

—

Místo vydání

—

Místo konání akce

Karlsruhe

Datum konání akce

8. 9. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—