An attempt of the quantitative analysis of cyclical constituents in structure of scanning microfluorescence signals from human hairs

13 Jul 2020, 20:27
2m
Poster SR for medicine and biology application Poster Session

Speakers

Mr Dmitry Sorokoletov Maria Chernetskaja (Busker INP SB RAS)

Description

The distribution of some chemical elements along human hairs is proved to be an valuable indicator in a range of various cases. It may serve in order to use in medicine, toxicology, criminalistics, forensic, archaeologics, and so on [1]. However at the present any systematical studies focusing to analysing these distribution with submillimetrical and tens micron spacial resolution do not exist. We took an attempt to describe qualitatively as well as to test an possibility of the analyse in detail such distributions on the example of the limited sample (consisting of fragments of 5 human hairs).
Our work was partitioned by two parts: panoramical and quantitative. The first part allowed to identify three main types of superposing legible pseudoperiodical cyclical components of elemental distribution for all studied hairs. The diapason of all crutial parameters of needed scanning experiments that is optimal for applied equipment for ability to process the results (cycles and correlation] between elemental signals) by statictical methods with a good quality was also found. The existence of an amount of strong limitations of these parameters are fully explained by the specifics of using synchrotron radiation sources of second generation. We can assume that this obstacle results in the existence of some principal limits for ability to process a detail analysis of human hairs (those are also characteristic for any case of applying a similar equipment).
In the second part of this work we selected a concrete distribution of fluorescent signal from the sulfur with sufficiently acceptable relation «signal to noise». Two special mathematical methods of analysing time series were applied successfully for a detail analysis of the cyclical part of the signal. The algorithm «Hilbert-Huang transformation (HHT)» [2] found 4 pseudoperiodical constituents, the algorithm «Singular Spectrum Analysis (SSA)» [3] found 5 preudoperiodiс constituents. The sum of all found components is equal initial signal despite of a normally distributed stochastic noise with very good quality. The shapes of the cyclical components differ from approximating cosines by 10-30 % in average (until 40-50 % for singular points) for different constituents. The small amount of determinated pseudoperiodical components may be explained by human methabolizm intraday cycles as well as the characteristic features [1] of the hair structure.
The scanning experiments were carried out at Synchrotron and Terahertz Research (INP SB RAS, Novosibirsk) and the Kurchatov Complex of Synchrotron-Neutron Research (KISI, Moscow) using a confocal X-ray microscope. The work was carried out in the framework of the RFBR projects no. 19-05-50046. The work was partly done at the shared research center SSTRC on the basis of the VEPP-4-VEPP-2000 complex at BINP SB RAS, using equipment supported by project RFMEFI62119X0022.
[1] Hair in Toxicology. An Important Bio-monitor.
Edited by Desmond John Tobin. The Royal Society of Chemistry 2005
[2] https://www.mathworks.com/help/signal/ref/hht.html
[3] https://www.gistatgroup.com/cat

Primary authors

Co-authors

Dr Iakov Rakshun Dr Andrey Gogin (Kurchatov Institute) Fedor Darin (Budker Institute of Nuclear Physics SB RAS) Dr Roman Senin (Kurchatov Institute)

Presentation Materials