Repeated double cross-validation applied to the PCA-LDA classification of SERS spectra: a case study with serum samples from hepatocellular carcinoma patients

Abstract

Intense label-free surface-enhanced Raman scattering (SERS) spectra of serum samples were rapidly obtained on Ag plasmonic paper substrates upon 785 nm excitation. Spectra from the hepatocellular carcinoma (HCC) patients showed consistent differences with respect to those of the control group. In particular, uric acid was found to be relatively more abundant in patients, while hypoxanthine, ergothioneine, and glutathione were found as relatively more abundant in the control group. A repeated double cross-validation (RDCV) strategy was applied to optimize and validate principal component analysis-linear discriminant analysis (PCA-LDA) models. An analysis of the RDCV results indicated that a PCA-LDA model using up to the first four principal components has a good classification performance (average accuracy was 81%). The analysis also allowed confidence intervals to be calculated for the figures of merit, and the principal components used by the LDA to be interpreted in terms of metabolites, confirming that bands of uric acid, hypoxanthine, ergothioneine, and glutathione were indeed used by the PCA-LDA algorithm to classify the spectra.

Label-free Surface Enhanced Raman Scattering (SERS) on Centrifugal Silver Plasmonic Paper (CSPP): A Novel Methodology for Unprocessed Biofluids Sampling and Analysis

“…A possible explanation can reside in the origin of the serum used in our study. The variation in relative intensity in HX and ERG bands has been linked to the spectral variability due to different clinical states of patients [ 33 , 34 , 35 ]. With respect to that, a standard commercial serum can easily present a lower HX and ERG concentration since no metabolite standardization or quantification is available for commercial serum.…”

The key role of ergothioneine in label‐free surface‐enhanced Raman scattering spectra of biofluids: a retrospective re‐assessment of the literature

“…Example of how characteristic ET bands (highlighted with grey areas) are observed in LF‐SERS spectra of a variety of human biofluids of healthy donors. SERS spectrum of (a) ET [19] is compared to SERS spectra of (b) erythrocytes lysate [19], (c) serum [29], (d) gingival crevicular fluid [28], (e) seminal plasma [50] and (f) cerebrospinal fluid [32]. All spectra were obtained with an excitation at 785 nm, and using Ag nanoparticles as substrates, with the exception of (f), for which Au nanoparticles were used.…”

“…For instance, preliminary studies have suggested that LF‐SERS spectroscopy on serum/plasma samples can be used to distinguish patients with several types of hepatocellular carcinoma (HCC) from patients with chronic hepatitis B (CHB), normal individuals, or patients with oesophageal carcinoma (EC) [8,29,33,45]. Recent serum data published by Gurian et al.…”

“…Recent serum data published by Gurian et al. reported how ET bands were relatively less intense in HCC patients when compared with healthy controls [29] . A relative decrease in the SERS intensity of ET bands for liver cancer patients with respect to controls can also be found in the works by Xiao et al.…”

Automorphism Groups of Alkane Graphs

“…The rdCV approach [35][36][37] estimates the optimum model complexity (given by the number of PLS components, AOPT) separately from estimating the prediction performance for new objects. [38] Furthermore, the variability of the performance criteria is characterized by repeated random splits into calibration and test sets. The essential parameters used for rdCV are as follows: the numbers of segments in outer and inner loop are 3 and 7, respectively; the number of repetitions is 50, resulting in 50 test-set predictions ŷ for each object, and 150 estimations of the optimum number of PLS components, with the most frequent value taken as the final AOPT.…”