CC BY
21B-PS-11
spectral features of conventional Raman spectroscopy and autofluorescence analysis of human skin in patients with kidney failure were studied
L. Bratchenko1, I. Bratchenko1, D. Artemyev1, O. Myakinin1, V. Grishanov1, D. Pimenova2, P. Lebedev2, V. Zakharov1
1Samara University, Laser and Biotechnical Systems Department, Samara, Russian Federation 2Samara State Medical University, Department of Internal Medicine, Samara, Russian Federation
Kidney failure leads to the breakdown of water, electrolyte, nitrogen and other metabolic disorders in the human body. Such changes are closely related to the skin condition and affect its component composition. A pathogenesis study demonstrated the accumulation of advanced glycation end-products (AGEs) in the vascular wall, vital organs tissues and skin in patients with severe stages of chronic kidney disease. Improving the diagnostic characteristics and more detailed analysis of pathologically-associated changes in the skin component composition are possible by utilizing the Raman spectroscopy (RS) and autofluorescence (AF) analysis. Therefore, the aim of this work is in vivo optical skin analysis in hemodialysis patients by means of AF in visible region and RS in NIR region.
Studies were performed for two subject's groups. The first group included 85 hemodialysis patients with kidney failure. 79 healthy volunteers constituted a second (control) group. The excitation of the skin AF response was stimulated by the light emitting diode in the 300-420 nm range. Detection of AF spectra was performed at the range of 435-600 nm. Along with AF measurements a backscattered signal was registered in order to compensate the individual differences in the optical properties of the skin. To study skin Raman spectral features the stimulation of collected spectra was performed by the laser module (LML-785) with central wavelength 785 nm. The Raman probe (RPB785) allows for focusing of the exciting radiation, collecting and filtering of the scattered radiation. The collected signal was decomposed into a spectrum using a portable spectrometer (QE65Pro). The spectra were recorded in the spectral range 780-1050 nm.
Firstly, we analyzed data on AF, while the target group is characteristically different by the presence of kidney failure and by level of average age. In order to discrimination target and control group the autofluorescence data is subjected to logistic regression analysis. The analysis demonstrated the detection of subjects associated with increased content of AGE products in the skin with specificity of 0.93, sensitivity of 0.82 and accuracy of 0.88. The PLS-DA analysis of skin Raman spectra set was carried out to determine the target group. The specificity, sensitivity and accuracy of identifying subjects with an elevated AGE-level by analyzing the skin Raman characteristics were 0.98, 0.88 and 0.95, respectively.
As expected, utilizing the combination of RS and AF analysis may improve the accuracy of detecting the AGEs products in the skin. Accuracy, sensitivity and specificity of the analysis are sufficient for clinical setting, which makes it a potential basis for screening AGE-associated diseases (for example, cardiovascular diseases) and methods for monitoring the status of hemodialysis patients.
