In the trial participated 9 subjects including males and females, age average is 43. Subjects were asked not to use coffee or tobacco and 15 minutes relax before measurement starts. During experiment various parameters measured to estimate heart risk. Record duration limited to approx. 15 sec. HRW prototype attached to each subject i.e. ECG, PPG sensors and standard blood pressure monitor. ECG electrodes placed on the subject's chest and PPG sensor on the left hand index finger. Standard blood pressure monitor which is a gold standard for blood pressure estimation used to compare results over state-of-the-art BP estimation technique.
First, signals acquired with HRW prototype and immediately after blood pressure (BP) estimated by a standard BP monitor. Blood pressure accuracy as well as other parameters are greatly dependent on PPG sensors' position. This is mostly due made hardware errors. Therefore it is necessary, first to check PPG signal's quality and verify that quality is acceptable for measurement. For each participant experiment repeated for at least 2 times.
Systolic and diastolic blood pressure and resting heart rate graphs summarized with the following pictures. All experiments performed with the alpha software version and the latest HRW hardware version which is sent to supporters all around the world. Software alpha version release are not yet available, it will be posted once it became more stable. Some records can be found here .
No subjects were excluded from experiment and all records considered correct. In a trial two males had elevated heart risk score, this is due to increased blood pressure levels. Algorithm estimated SBP and RHR parameters with rather high accuracy than DBP. Arrhythmia haven't been detected none of the participants. Blood pressure and some other parameters which require PPG sensor can only be estimated in a still state while the heart rate and some other parameters are available during physical activity. In order to increase overall performance it is important to collect more signals from different people and store them in a database. This would allow much more to improve overall algorithm quality.
SBP (Mean diff ± SD) 3.7143 ± 5.6188 mmHg
DBP (Mean diff ± SD) 10.7143 ± 5.1547 mmHg
HRW prototype: RHR (Mean ± SD) 73.0909 ± 2.8793 bpm
Standard monitor: RHR (Mean ± SD) 73 ± 3.8729 bpm