Breathing Rate Detection in Preterm Infants Using a Fiber Optic Pressure-Sensitive Mattress: a feasibility study

Continual and accurate monitoring of respiration is essential in neonatal care. Conventional techniques, including respiratory belts, adhesive sensors, or impedance pneumography, can often be obtrusive and uncomfortable, and may suffer from artifacts related to displacement, signal disturbances, or skin irritation.This study aims to evaluate the feasibility of detecting the breathing rate (BR), during periods in which the infants are still, using a fiber optic pressure-sensitive mattress (FM) by comparing the BR derived from data collected from the patient monitor, which stores in the data warehouse two types of data: the chest impedance waveform (CI-WAV) and the processed parameter data (CI-PAR). The BRs retrieved from these two served as the benchmark for the evaluation.In the analysis, the Mean Absolute Error (MAE) for the FM rate compared to the CI-PAR rate was calculated to have a mean (standard deviation) of 11.83bpm (4.20bpm). Similarly, when contrasting the FM rate to the measured CI-WAV rate, the mean (standard deviation) was 18.06bpm (9.16bpm). The discrepancy can be attributed to the intrinsic difference in the waveforms from which the BRs were extracted. However, a correspondence between the FM-based BR and the PAR-based or WAV-based BR could be found when plotting the BR over time.In conclusion, the FM was able to detect slight pressure variations, from which respiration data could be extracted. Its non-obtrusive design and high sensitivity make it an interesting candidate for monitoring respiration in preterm infants.