Plastic Optical Fiber With Spherical Lens Embedded in Mattress for Sleep Monitoring

The quality of sleep is closely related to human health; however, traditional polysomnography (PSG) is expensive and time-consuming, making it impractical for routine home sleep diagnosis. This study introduces a smart mattress based on a spherical lens enhanced plastic optical fiber (SL-POF) sensor. UV-fused silica ball lenses are embedded into PMMA POFs, which are then encapsulated with Ecoflex silicone rubber. The smart mattress employs a dual-channel design and collected data from eight subjects in various sleep postures. The signals are transformed into the frequency domain for processing using the Welch power spectral density estimation method. The results indicate that the smart mattress has a mean absolute error (MAE) of less than 1.92 bpm for heart rate (HR) and 0.95 bpm for respiratory rate (RR). Additionally, consistency analysis was conducted using Bland-Altman analysis and linear regression methods. The results showed that the 95% limits of agreement (LOA) for HR were −3.047 to 2.915 bpm, with an ${R}\,^{{2}}$ value of 0.95729, and for RR they were −1.741 to 1.380 bpm, with an ${R}\,^{{2}}$ value of 0.9738. Furthermore, Random Forest algorithm was applied to the dataset, successfully classifying between five different sleep behaviors with a classification accuracy of 97.67%. This smart mattress is capable of long-term, stable monitoring of physiological parameters and sleep behavior monitoring, providing a low-cost solution for home sleep monitoring.