Estimation of Respiratory Frequency in Women and Men by Kubios HRV Software Using the Polar H10 or Movesense Medical ECG Sensor during an Exercise Ramp.
Bruce RogersMarcelle SchafarczykThomas GronwaldPublished in: Sensors (Basel, Switzerland) (2022)
Monitoring of the physiologic metric, respiratory frequency (RF), has been shown to be of value in health, disease, and exercise science. Both heart rate (HR) and variability (HRV), as represented by variation in RR interval timing, as well as analysis of ECG waveform variability, have shown potential in its measurement. Validation of RF accuracy using newer consumer hardware and software applications have been sparse. The intent of this report is to assess the precision of the RF derived using Kubios HRV Premium software version 3.5 with the Movesense Medical sensor single-channel ECG (MS ECG) and the Polar H10 (H10) HR monitor. Gas exchange data (GE), RR intervals (H10), and continuous ECG (MS ECG) were recorded from 21 participants performing an incremental cycling ramp to failure. Results showed high correlations between the reference GE and both the H10 (r = 0.85, SEE = 4.2) and MS ECG (r = 0.95, SEE = 2.6). Although median values were statistically different via Wilcoxon testing, adjusted median differences were clinically small for the H10 (RF about 1 breaths/min) and trivial for the MS ECG (RF about 0.1 breaths/min). ECG based measurement with the MS ECG showed reduced bias, limits of agreement (maximal bias, -2.0 breaths/min, maximal LoA, 6.1 to -10.0 breaths/min) compared to the H10 (maximal bias, -3.9 breaths/min, maximal LoA, 8.2 to -16.0 breaths/min). In conclusion, RF derived from the combination of the MS ECG sensor with Kubios HRV Premium software, tracked closely to the reference device through an exercise ramp, illustrates the potential for this system to be of practical usage during endurance exercise.
Keyphrases
- heart rate
- heart rate variability
- blood pressure
- mass spectrometry
- high intensity
- multiple sclerosis
- ms ms
- resistance training
- healthcare
- public health
- mental health
- data analysis
- pregnant women
- insulin resistance
- skeletal muscle
- deep learning
- human health
- metabolic syndrome
- middle aged
- social media
- pregnancy outcomes
- carbon dioxide