Sunday 02 February 2025
A team of researchers has made a significant breakthrough in developing a non-invasive method to measure glycated hemoglobin (HbA1c), a key indicator of diabetes management. By using only two wavelengths of light, they have achieved results comparable to those obtained using three wavelengths, making the system more compact and cost-effective.
The new method involves shining light through the fingertip, which is rich in capillaries containing oxygenated and deoxygenated hemoglobin. The absorption of light by these molecules varies depending on their concentration and type, allowing researchers to estimate HbA1c levels.
To develop this technique, the team employed a machine learning algorithm, XGBoost, to calibrate the system using a dataset of 150 participants. They found that two wavelengths, 465 nm (blue) and 525 nm (green), provided the best results for estimating HbA1c. The accuracy of their method was comparable to that of previous studies using three wavelengths.
The researchers also explored the use of different wavelength pairs and compared their performance using an evaluation metric called the Bland-Altman analysis. This statistical test assesses the agreement between two methods by plotting the difference against the average value of both methods. The results showed that the green-red pair performed better than other combinations, with most data points falling within a narrow range.
In addition to estimating HbA1c levels, the team also developed a method for non-invasive estimation of oxygen saturation (SpO2). This is an important parameter in medical diagnosis, as it can indicate respiratory distress or other conditions. The researchers found that their system performed well in this regard, with a correlation coefficient of 0.986.
The potential applications of this technology are significant. A non-invasive HbA1c monitor could revolutionize diabetes management, allowing patients to track their glucose levels easily and accurately at home. This could lead to better disease control, reduced complications, and improved quality of life for millions of people worldwide.
While further research is needed to refine the system and validate its performance in larger populations, this breakthrough marks an important step towards developing a practical and accessible non-invasive HbA1c monitor. With its potential to transform diabetes care, this technology has the potential to make a significant impact on global health.
Cite this article: “Non-Invasive Method for Measuring Glycated Hemoglobin Levels”, The Science Archive, 2025.
Glycated Hemoglobin, Hba1C, Diabetes Management, Non-Invasive Measurement, Machine Learning, Xgboost, Wavelength, Light Absorption, Oxygen Saturation, Spo2, Bland-Altman Analysis
