Evaluation of Raman spectra of human brain tumor tissue using the learning vector quantization neural network

Abstract

The Raman spectra of tissue of 20 brain tumor patients was recorded using a confocal microlaser Raman spectroscope with 785 nm excitation in vitro. A total of 133 spectra were investigated. Spectra peaks from normal white matter tissue and tumor tissue were analyzed. Algorithms, such as principal component analysis, linear discriminant analysis, and the support vector machine, are commonly used to analyze spectral data. However, in this study, we employed the learning vector quantization (LVQ) neural network, which is typically used for pattern recognition. By applying the proposed method, a normal diagnosis accuracy of 85.7% and a glioma diagnosis accuracy of 89.5% were achieved. The LVQ neural network is a recent approach to excavating Raman spectra information. Moreover, it is fast and convenient, does not require the spectra peak counterpart, and achieves a relatively high accuracy. It can be used in brain tumor prognostics and in helping to optimize the cutting margins of gliomas.

Raman spectroscopy as a promising noninvasive tool in brain cancer detection

“…Liu et al employed learning vector quantization neural network to distinguish white matter and tumors. 37 which yielded a better classi¯cation e±ciency. In a recent and interesting study, band-wise feature extraction, sound synthesis from Raman spectra and feedback mechanism was employed to achieve classi¯cation of spectral data from¯xed and parafn embedded tissue sections.…”

In-vivo Raman spectroscopy: from basics to applications

“…In addition to the significant shot noise level, a convolution of the broad polynomial background with the filter function can easily result in additional peaks and can even emulate to some degree Raman spectra, making the analysis very challenging and the results often misleading. Jermyn and coauthors also demonstrated the potential of boosted trees 109 and artificial neuronal networks (ANN) 210 to distinguish tissue with and without the presence of light artifacts, concluding that ANN achieves an accuracy of 90%, sensitivity of 91%, and specificity of 89% when measuring with light artifacts. In contrast, boosted trees have a lower accuracy.…”

Rapid and Low-Cost Detection of Thyroid Dysfunction Using Raman Spectroscopy and an Improved Support Vector Machine

“…The 74 spectral data of PCA feature extraction of Section 3.2 were loaded, and a random number column vector from 1-74 was created to disorganize the spectral data [27]. We chose the first 40 vectors as the training set; the remaining 34 were used as the prediction set to validate the diagnostic accuracy of the SVM model optimized by the AFSA and AFUD algorithms.…”