應用於鑑別糖尿病患者之氣體酸鹼值感測系統

A pH Gas Sensing System for the Identification of Diabetes Patients

本論文提出一個應用於鑑別糖尿病患者之酸鹼值（pH）氣體感測系統，當食物殘留於口腔內時，細菌會開始分解口腔內食物殘渣中的蛋白質，這個過程中會使口腔氣體之pH值提升。而糖尿病患者之唾液分泌量相較於正常人還要少，因此更利於細菌分解導致更多腐敗氣體的產生。使用此pH氣體感測系統量測口腔呼出氣體之pH值變化量，可以用來初步判斷受試者是否罹患糖尿病。
在系統設計方面，我們選用以ARM Cortex-M3為核心的Silicon Labs EFM32微處理晶片為系統之控制核心，並搭配超極生技公司所開發之免收集氣體酸鹼電極（Humming Probe UH2-Gas）。起初，受試者會往一吹氣袋內吹氣至飽和，再將氣體酸鹼感測試片放入吹氣袋之導管內，以固定速度擠壓吹氣袋，氣體酸鹼感測試片會將氣體樣本的pH反應轉換成電壓信號，透過微處理器將電壓信號轉換成數位信號，傳送給電腦端，利用LabVIEW開發之使用者介面，可以即時顯示pH值變化量之結果。
在系統驗證方面，我們收集了5名糖尿病患者及20名正常人的數據當作對比。每位受試者會在飯後1個小時後進行吹氣收集，分別為飯後不漱口7次及飯後漱口7次，共測驗14次。實驗結果顯示，糖尿病患者之測得平均pH值要高於無糖尿病的正常人。
本論文所提出的氣體感測系統可以應用在鑑別受試者是否罹患糖尿病之症狀，本系統之特點是無須透過傳統的抽血取樣方式就能初步得到受試者之檢驗結果，未來希望能應用於檢查正常人之身體狀況是否有出現異樣，可以幫助受試者得到初步的健康檢測資訊而即時改善。

In this paper, a pH gas sensing system is proposed for identifying diabetic patients. When food residues remain in the oral cavity, bacteria begin to break down the proteins in the food debris, which increases the pH value of oral gases. Diabetic patients tend to have lower saliva secretion than healthy individuals, which further facilitates bacterial decomposition and leads to the production of more putrid gases. By using this pH gas sensing system to measure changes in the pH value of exhaled oral gases, it is possible to make a preliminary determination of whether a subject has diabetes.
In terms of system design, we selected the Silicon Labs EFM32 microprocessor, based on the ARM Cortex-M3 core, as the system's control unit, and integrated it with the Humming Probe UH2-Gas, a non-collecting gas pH electrode developed by UltraE. Initially, the subject exhales into an airbag until it is saturated. The gas pH sensing strip is then placed inside the airbag’s conduit, and the airbag is compressed at a constant speed. The gas pH sensing strip converts the pH response of the gas sample into a voltage signal, which is then digitized by the microprocessor and transmitted to a computer. Using a user interface developed with LabVIEW, real-time pH variation results can be displayed.
For system validation, we collected data from five diabetic patients and twenty healthy individuals for comparison. Each subject performed breath collection one hour after a meal, including seven tests without rinsing their mouth and seven tests after rinsing, for a total of fourteen tests. Experimental results show that the average pH value measured in diabetic patients is higher than that of healthy individuals.
The gas sensing system proposed in this paper can be applied to the preliminary identification of diabetes symptoms in individuals. Its key advantage is that it provides an initial test result without requiring traditional blood sampling. In the future, this system could be used to monitor the health status of healthy individuals, helping them obtain early health insights and take timely corrective actions.