Jan 11, 2021
Optical heart rate sensor simulation using Zemax OpticStudio
In recent years, handheld, portable, and wearable medical imaging devices have grown in popularity, from fast on-site measurements during surgeries, to comfortable continuous monitoring during everyday tasks. The most common wearable example is the optical heart rate sensor used in sport bracelets and smart watches. The applied photoplethysmography (PPG) technology is a low cost, non-invasive optical method used for taking physiological measurements at the surface of the skin. This article demonstrates how OpticStudio can be used to implement a layered human skin model and simulate optical heart rate monitoring via the API.
Skin model in OpticStudio
The Henyey-Greenstein distribution is recognized by the biomedical community as representative of bulk scattering in human tissue. In the non-sequential mode of OpticStudio the Henyey-Greenstein model is available as a DLL to characterize scattering in the bulk of the volume. Using the DLL, a layered skin model was created consisting of an epidermis layer, four different layers of the dermis having distinct blood content, namely the papillary dermis, upper blood net dermis, reticular dermis, and deep blood net dermis, as well as a subcutaneous fat layer. The blood vessels for each layer were represented by taking the blood content of the layers into account as a weighted average of the optical parameters of blood and the rest of the tissue. An LED source model with a wavelength of 575 nm was used as the light source, which falls in the range of commercially applied green LEDs used in reflectance PPG-based devices, as its penetration depth is best suited to analyze pulsatile blood flow in the dermis layers. The amount of back-scattered light from the tissue was measured on detector objects. Since in most of the applications, the time-dependent effects are not relevant, and are typically eliminated from the results, an average layered skin model in OpticStudio was implemented, as shown in the image below.
Heart rate sensor simulation using Python ZOS-API
In order to simulate heart rate monitoring, the ZOS-API was used to mimic pulsatile blood flow in the skin layers and examined the detected back-scattered light as a function of the changes. An empirical function was used to characterize the pulsation of the relative blood content in the skin due to heartbeat. Accordingly, in each step, the reflective index and the parameters of the Henyey-Greenstein scattering distribution from the API were updated, a ray trace was ran, and the results were analyzed. Since the absorption coefficient of the blood is much larger than the rest of the tissue, as the relative blood content increases in the layers, the measured power on the detectors decreases. The results of the simulation are shown on the image below.
According to the literature referenced for this modelling, as the blood flows into the vessels the relative blood content of the skin layers approximately doubles. Based on the simulations, this results in ~10…15% change in the detector signal, which is in good agreement with published observations. In commercial heart rate sensors typically a straightforward signal processing algorithm is applied, which first smooths the noisy signal, and then counts the number peaks per minute to calculate the heart rate.
Further modelling possibilities
This example was only one of many different modelling possibilities in OpticStudio. A few other options include source – detector position optimization for heart rate sensors, signal analysis for specific skin characteristics – parameter settings, multiple wavelength applications and database generation for healthy and diseased tissue. As these medical imaging devices continue to gain popularity, more opportunities arise to use this non-invasive technology to diagnose and monitor various health issues. The fast-growing field will continue to offer new and exciting technologies to help with the advancement of medical care.
To learn more about how OpticStudio helps the life sciences industry create life-changing optical designs read the knowledge based article, How to model the human skin and optical heart rate sensors in OpticStudio and visit the OpticStudio product page.
- T. Maeda, N. Arakawa, M.Takahashi, Y. Aizu. Monte Carlo Simulation of Spectral Reflectance Using a Multilayered Skin Tissue Model. Optical Review, 17(3):223–229 (2010)
- Y. P. Sinichkin, S. R. Utz, A. H. Mavliutov, H. A. Pilipenko. In Vivo Fluorescence Spectroscopy of The Human Skin: Experiments and Models. Journal of Biomedical Optics, 3(2):201–211 (1998)
- V. V. Tuchin. Light scattering study of tissues. Physics – Uspekhi, 40(5):495-515 (1997)
- H. Li, C. Zhang, X. Feng. Monte Carlo simulation of light scattering in tissue for the design of skin-like optical devices. Biomedical Optics Express, 10(2):868-878 (2019)
- M. N. Salihin Yusoff, M. S. Jaafar. Performance of CUDA GPU in Monte Carlo Simulation of Light-Skin Diffuse Reflectance Spectra. IEEE EMBS International Conference on Biomedical Engineering and Sciences (2012)
- V. Meglinski, S. J. Matcher. Quantitative assessment of skin layers absorption and skin reflectance spectra simulation in the visible and near-infrared spectral regions. Physiological Measurement 23(4):741-753 (2002)
- Doronin, I. Fine, I. V. Meglinski. Assessment of the Calibration Curve for Transmittance Pulse-Oximetry. Laser Methods in Chemistry, Biology, and Medicine (2011)
- Meglinski, Monte Carlo simulation of reflection spectra of random multilayer media strongly scattering and absorbing light. Quantum Electronics, 31(12):1101-1107 (2001)