利用近紅外光譜做非侵入式之生理訊號量測

Abstract

本論文之目的是以近紅外線光譜做非侵入式的生理訊號量測，本論文主要是利用市上現有之各波長的LED光源元件來取代體積龐大的光源量測系統，藉以縮小系統的體積及驗證系統架構之可行性。由於含氧血紅素與去氧血紅素對於近紅外光譜的吸收波長不同，因此可將近紅外光入射至生物組織內，利用光的回彈反射特性，透過適當的電路設計以減少雜訊、放大訊號來取得反射光的強度，並計算出光的衰減比例；接著經由正確公式推演即可以求得血氧濃度值。本研究中，入射光在皮膚組織中的分佈情況也是發展重點之一。此研究討論利用蒙地卡羅方法模擬LED光源入射到皮膚組織中的分佈情況、光子路徑與穿透深度，藉以利用此方法模擬皮膚組織中的血氧濃度，並驗證硬體所量測到的結果。

The purpose of this thesis is to develop a non-invasive physiological signal detection by using the Near-Infrared Spectroscopy. In this thesis, we use the different of the wavelength of the LED to replace the optical system, for the reducing of the volume of the measurement system. By developing the related sensing circuits and computational algorithms, we are able to measure human oxyhemoglobin (HbO2), deoxyhemoglobin (Hb) and oxygen saturation (SO2) using the porposed system. One of the direction of the research is for the penetration of the photons. We use the Monte Carlo simulation to simulate the light of the LED which is in the skin organization and discuss the condition of the distribution, the path of the photons and depth. By using the above method, the oxygen saturation of skin tissue also can be compared with the result of the hardware implementation.