BLOOD PRESSURE ESTIMATION VIA THE OPTIBP MOBILE APPLICATION - PERFORMANCES ACROSS DIFFERENT BODY MASS INDEX. (June 2022)

• Record Type: Journal Article
• Authors: Caillat, Mary; Degott, Jean; Wuerzner, Arlène; Proença, Martin; Bonnier, Guillaume; Knebel, Jean-François; Stoll, Chloé; Christen, Urvan; Durgnat, Virginie; Burnier, Michel; Wuerzner, Grégoire; Schoettker, Patrick
• Is Part Of: Journal of hypertension. Volume 40(2022)Supplement 1
• Page Start: e91
• Publication Date: 2022-06
• Subjects: Hypertension -- Periodicals; Hypertension -- Periodicals; 616.132005
• DOI: 10.1097/01.hjh.0000836108.49528.36
• Languages: English
• ISSNs: 1473-5598

Abstract

Abstract : Objective: The association between overweight-obesity and hypertension is well known with both pathologies considered as public health priorities. The current reference techniques for Blood Pressure (BP) measurement represent a challenge in the obese population since the cuff may not be fully adapted to patient's anatomy. OptiBP mobile application is a cuffless approach that estimates BP with the patient's fingertip on the camera of a smartphone that could improve access to the measure while decreasing the challenges related to the cuff-based technique. Design and method: This study assessed the accuracy of OptiBP across different Body Mass Index (BMI) categories via a dedicated analysis of an existing dataset from a previously published clinical study (NCT038752488). The study protocol followed the ISO 81060–2:2018 standard for "Non-invasive sphygmomanometers - Clinical investigation of intermittent automated measurement type'' adapted for a cuffless device. The performance of OptiBP was assessed with ISO 81060–2 Criterion 1 - the mean error value between the reference BP value and the estimated BP - and Criterion 2 - standard deviation of the patients' mean error on 414 valid recordings obtained from 95 patients. Patients were divided in three BMI categories according to WHO classification (Table 1). Results: Data processing shows an acceptance of OptiBP recordings of 82% suggesting a high quality dataset. Results show that OptiBP estimations ulfil Criterion 1 with a mean error lower than ± 5 mmHg and SD < = 8mmHg for systolic BP and diastolic BP in each subgroup (Table 2) and Criterion 2 with all calculated values under the passing threshold (Table 3). The accuracy of these estimations are strengthened by absolute error proportions in accordance with the recommendations of the European Society of Hypertension and clinically relevant margin of error (error-grid analysis; at least 98% of "no risk'' range). Conclusions: Altogether these results illustrate the accuracy of OptiBP estimation regardless of the patient's BMI. This is a further step toward reliable, comfortable and inclusive medical care for all population diversity. Further research is in progress to improve the algorithm and validate it across other population diversities and pathologies.