Non-invasive and rapid diagnosis of type 2 diabetes mellitus based on the analysis of hair by front-face fluorescence spectroscopy

Abstract

Rapid and early detection of pathologies for screening purposes is not always easy with traditional methods. New low-cost optical techniques that are suitable for preventive detection and, more specifically, for developing countries must be developed. The objective of this work is to propose UV-VIS fluorescence of hair as a promising technique for a simple and rapid preliminary diagnosis of type II diabetes. In this work, we analyzed 130 hair samples taken from volunteers of different ages at the Habib Thameur Hospital in Tunis, Tunisia. These samples were analyzed clinically beforehand, and 50 were classified as healthy acting as a control, 24 were classified as low-level diabetics with a glycated hemoglobin A1C (HbA1c) &lt;<!-- 7 % , and the other 56 were classified as high-level diabetics having an HbA1c &gt; 7 % . The fluorescence of the patients’ hair led to several interesting results. Indeed, a clear discrimination was obtained not only between the healthy and patients with a disease, but also a discrimination between diabetics with low levels and high levels or diabetics with and without complication. Also, a strong correlation between fluorescence spectra and glycated HbA1c for the diabetic population was clearly established. A linear discriminant analysis shows that it is possible to predict the status of test patients after having trained a fraction of the population. All these results show the ability of the front-face fluorescence (FFF) technique to detect disease and predict advanced states simply by capturing the fluorescence of the hair illuminated by a single LED. This work shows for the first time, to our knowledge, the capability of the FFF technique on hair samples for the diagnosis of diabetes.

Green Synthesis and Characterization of Novel Silver Nanoparticles Using Achillea maritima subsp. maritima Aqueous Extract: Antioxidant and Antidiabetic Potential and Effect on Virulence Mechanisms of Bacterial and Fungal Pathogens

“…For the experiment, the setup employed, based on previous studies [ 32 , 33 , 34 , 35 ], bifurcated optical fiber and a UV light electroluminescent diode (LED) at 365 nm with an 8 nm half-width to excite the samples. UV-Vis front-face fluorescence spectroscopy (FFFS) provides fluorescence emission excitation and frontal collection simultaneously.…”

“…Using SpectraSuite software version 2.0, spectra were acquired between 400 and 800 nm. The same methodology and identical conditions were used for all measurements [ 32 , 33 , 34 , 35 ]. We made five recordings for the sample and left a 3 s exposure delay between each recording to allow for stability of the fluorescence signal before each capture.…”

“…The measured spectra were processed by averaging the five recorded spectra that were acquired for the silver nanoparticles using the software program Igor Pro 6.01, and then, the measured spectra were subjected to deconvolution analysis using Gaussian functions to identify the peak positions. Python was utilized for the spectral deconvolution and treatment [ 32 , 33 , 34 , 35 ].…”