Application of optical spectroscopy and artificial neural networks for monitoring the elimination of anticancer nanoagents from the body
Application of optical spectroscopy and artificial neural networks
for monitoring the elimination of anticancer nanoagents from the body
A.D. Kudryashov, O.E. Sarmanov, K.A. Laptinsky, S.A. Burikov, M. Yu. Khmeleva, A.A. Fedyanina, P.V. Golubtsov, T.A. Dolenko
Moscow State University, Department of Physics, Moscow, Russia
In the last decade, the use of carbon nanoparticles as drug carriers has become more and more widespread. USA Food and Drug Administration (FDA) has already approved nearly 20 nano-based drugs. The use of any drug is associated with the need to control its excretion from the body for a certain period, including control of the excretion of the drug carrier, which in this case are carbon nanoparticles.
This study presents an approach to solving the problem of controlling the excretion of nanoagents from the body through urine with the help of carbon dots (CD) using optical spectroscopy and artificial neural networks (ANN). The use of ANN depends on the significant variability of the optical properties of urine. Thus, the shape of the urine fluorescence spectrum depends on many factors that are difficult to control - the donor's age, gender, nutrition and health status, time of sampling, etc. That is why standard methods of processing spectroscopic data are not applicable to this problem.
624 suspensions of CD nanocomplexes were prepared with an anti-cancer drug adsorbed on them - doxorubicin (Dox) - in urine. CD and Dox concentrations varied from 0 to 1.2 mg/L with 0.05 mg/L increment, and from 0 to 1 mg/L with 0.042 mg/L increment, respectively. The problem was solved using three sets of optical spectroscopy data: fluorescence spectra of aqueous suspensions of CD and Dox in urine upon excitation by radiation with wavelengths of 405 nm, 532 nm, as well as optical absorption spectra in the range from 190 to 800 nm. Fully connected neural networks - multilayer perceptrons (MLP) - were applied to the obtained spectral data, which ensured the determination of the concentrations of CD and Dox in urine. To increase the accuracy of monitoring the excretion of CD and Dox with urine, autoencoders were additionally used.
The report contains a comparative analysis of the results obtained using the indicated spectroscopic methods, and also the factors influencing the results of using artificial neural networks are discussed in details. Eventually it was found that the smallest error in determining the desired parameters from the absorption spectra is provided by an MSP with one hidden layer and 64 neurons in it. By the way the average absolute error in determining the concentrations of UT and Dox was 0.042 mg / L (3.6% of the maximum value) and 0.024 mg / L (2.4% of the maximum value), respectively.
This study has been supported by Russian Foundation for Basic Research No. 19-01-00738 (K.A. Laptinskiy). The contribution of O.E. Sarmanova (programming and ANN training) was supported by the Foundation for the Advancement of Theoretical Physics and Mathematics “BASIS” (Project No. 19-2-6-6-1).
File with abstract
Moscow State University, Department of Physics