A machine learning framework to analyze hyperspectral stimulated Raman scattering microscopy images of expressed human meibum

• Published in: Journal of Raman spectroscopy Vol. 48; no. 6; pp. 803 - 812
• Main Authors: Alfonso‐García, Alba, Paugh, Jerry, Farid, Marjan, Garg, Sumit, Jester, James, Potma, Eric
• Format: Journal Article
• Language: English
• Published: England Wiley Subscription Services, Inc 01.06.2017
• Subjects: Aversion learning; Classification; Cluster analysis; Clustering; Correlation analysis; Datasets; human meibum; hyperspectral stimulated Raman scattering microscopy; Learning algorithms; Machine learning; Microscopy; multi‐image analysis; Quantitative analysis; Raman spectra; Scattering; Serial learning; Stacks; Stretching; Training; Vector quantization; multi-image analysis; human meibum; hyperspectral stimulated Raman scattering microscopy; machine learning
• ISSN: 0377-0486; 1097-4555
• DOI: 10.1002/jrs.5118

We develop and discuss a methodology for batch‐level analysis of hyperspectral stimulated Raman scattering (hsSRS) data sets of human meibum in the CH‐stretching vibrational range. The analysis consists of two steps. The first step uses a training set (n=19) to determine chemically meaningful reference spectra that jointly constitute a basis set for the sample. This procedure makes use of batch‐level vertex component analysis, followed by unsupervised k‐means clustering to express the data set in terms of spectra that represent lipid and protein mixtures in changing proportions. The second step uses a random forest classifier to rapidly classify hsSRS stacks in terms of the pre‐determined basis set. The overall procedure allows a rapid quantitative analysis of large hsSRS data sets, enabling a direct comparison among samples using a single set of reference spectra. We apply this procedure to assess 50 specimens of expressed human meibum, rich in both protein and lipid, and show that the batch‐level analysis reveals marked variation among samples that potentially correlate with meibum health quality. Copyright © 2017 John Wiley & Sons, Ltd. A machine learning methodology for batch‐level analysis of hyperspectral stimulated Raman of expressed human meibum is presented. A strategy based on clustering and classification of hyperspectral stimulated Raman scattering spectra in the CH‐stretching region provides a high throughput, rapid quantitative analysis of the protein composition of meibum. This approach enables direct comparison among samples using a single set of reference spectra and reveals marked variation among samples that potentially correlate with meibum health quality.