Application of Polynomial Regression Method in Non-invasive Measurement of Blood Sugar, Cholesterol, and Non-invasive Uric Acid Based on IoT

Early disease avoidance depends much on health monitoring. However, the general examination methods still used today are invasive, namely, using a syringe to take blood samples. Many do not undergo routine examinations because this method is uncomfortable and expensive. In this study, the MAX30105 optical sensor is used as a non-invasive measuring device that can read the reflection of infrared light from the fingertip. After that, the second-order polynomial regression method is used to process the sensor data and determine the blood sugar, cholesterol, and uric acid levels. Using calibration data, this tool will change the reflected light signal into numbers for these three substances. The quantitative experimental method was conducted on 15 participants, The quantitative experimental method was carried out on 15 participants, the test results showed that blood sugar levels reached 91.50%, cholesterol levels reached 86.07%, and uric acid levels reached 89.33%. Real-time data transmission is carried out through the Adafruit IO platform, which was chosen for its accessibility and ease of integration. At the same time, a mobile application was developed using MIT App Inventor for user-friendly health data visualization. A preliminary Quality of Service (QoS) assessment showed an average data latency of 500–700 ms and a 97% transmission success rate via Wi-Fi. These results indicate that this device is reasonably practical and comfortable. However, several factors, such as skin thickness, finger position, and skin cleanliness, can affect the accuracy of the measurement results. Therefore, this tool cannot yet replace regular medical standards.