Multivariate analysis of Brillouin imaging data by supervised and unsupervised learning

Brillouin imaging relies on the reliable extraction of subtle spectral information from hyperspectral datasets. To date, the mainstream practice has been using line fitting of spectral features to retrieve the average peak shift and linewidth parameters. Good results, however, depend heavily on sufficient SNR and may not be applicable in complex samples that consist of spectral mixtures. In this work, we thus propose the use of various multivariate algorithms that can be used to perform supervised or unsupervised analysis of the hyperspectral data, with which we explore advanced image analysis applications, namely unmixing, classification and segmentation in a phantom and live cells. The resulting images are shown to provide more contrast and detail, and obtained on a timescale $10^2$ faster than fitting. The estimated spectral parameters are consistent with those calculated from pure fitting.

[1]  T. W. Anderson An Introduction to Multivariate Statistical Analysis, 2nd Edition. , 1985 .

[2]  Robert Prevedel,et al.  Imaging mechanical properties of sub-micron ECM in live zebrafish using Brillouin microscopy , 2018, bioRxiv.

[3]  Paola Borri,et al.  Quantitative Chemical Imaging and Unsupervised Analysis Using Hyperspectral Coherent Anti-Stokes Raman Scattering Microscopy , 2013, Analytical chemistry.

[4]  S. Rice,et al.  Probing the internal micromechanical properties of Pseudomonas aeruginosa biofilms by Brillouin imaging , 2017, npj Biofilms and Microbiomes.

[5]  Guang-Zhong Yang,et al.  Spatio-spectral classification of hyperspectral images for brain cancer detection during surgical operations , 2018, PloS one.

[6]  J. C. Dunn,et al.  A Fuzzy Relative of the ISODATA Process and Its Use in Detecting Compact Well-Separated Clusters , 1973 .

[7]  Vladislav V. Yakovlev,et al.  Seeing cells in a new light: a renaissance of Brillouin spectroscopy , 2016 .

[8]  Luc Moens,et al.  The Biodata toolbox for MATLAB , 2009 .

[9]  S. Gupta,et al.  Assessment of corneal substrate biomechanics and its effect on epithelial stem cell maintenance and differentiation , 2019, Nature Communications.

[10]  C. Paterson,et al.  Fiber-integrated Brillouin microspectroscopy: Towards Brillouin endoscopy , 2017 .

[11]  Antonio J. Plaza,et al.  Hyperspectral Unmixing Overview: Geometrical, Statistical, and Sparse Regression-Based Approaches , 2012, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[12]  D. Fioretto,et al.  Brillouin-Raman mapping of natural fibers with spectral moment analysis , 2019, Biomedical optics express.

[13]  Carl Paterson,et al.  Water content, not stiffness, dominates Brillouin spectroscopy measurements in hydrated materials , 2017, Nature Methods.

[14]  A. Mahadevan-Jansen,et al.  Automated Method for Subtraction of Fluorescence from Biological Raman Spectra , 2003, Applied spectroscopy.

[15]  Noël Bonnet,et al.  New applications of multivariate statistical analysis in spectroscopy and microscopy , 1992 .

[16]  Giuliano Scarcelli,et al.  Direct Three-Dimensional Measurement of Refractive Index via Dual Photon-Phonon Scattering. , 2019, Physical review letters.

[17]  Arvid Lundervold,et al.  An overview of deep learning in medical imaging focusing on MRI , 2018, Zeitschrift fur medizinische Physik.

[18]  Anastasios Koulaouzidis,et al.  Use of Hyperspectral/Multispectral Imaging in Gastroenterology. Shedding Some–Different–Light into the Dark , 2019, Journal of clinical medicine.

[19]  M. Kendall,et al.  Kendall's advanced theory of statistics , 1995 .

[20]  Juyang Weng,et al.  Using Discriminant Eigenfeatures for Image Retrieval , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  Carl Paterson,et al.  Assessing corneal biomechanics with Brillouin spectro-microscopy. , 2016, Faraday discussions.

[22]  M. Barker,et al.  Partial least squares for discrimination , 2003 .

[23]  Max Diem,et al.  Spectral unmixing and clustering algorithms for assessment of single cells by Raman microscopic imaging , 2011 .

[24]  Károly Héberger,et al.  Classification of olive oils using high throughput flow 1H NMR fingerprinting with principal component analysis, linear discriminant analysis and probabilistic neural networks , 2005 .

[25]  Alice Berthelot,et al.  High-Frequency Mechanical Properties of Tumors Measured by Brillouin Light Scattering. , 2019, Physical review letters.

[26]  Karl Pearson F.R.S. LIII. On lines and planes of closest fit to systems of points in space , 1901 .

[27]  David Rubin,et al.  Introduction to Continuum Mechanics , 2009 .

[28]  X. Ceamanos,et al.  Processing Hyperspectral Images , 2016 .

[29]  Hyperspectral analysis applied to micro-Brillouin maps of amyloid-beta plaques in Alzheimer's disease brains , 2018, The Analyst.

[30]  J. Roodenburg,et al.  In vivo detection of dysplastic tissue by Raman spectroscopy. , 2000, Analytical chemistry.

[31]  José M. Bioucas-Dias,et al.  Minimum Volume Simplex Analysis: A Fast Algorithm to Unmix Hyperspectral Data , 2008, IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium.

[32]  R. Capasso,et al.  Anandamide-derived Prostamide F2α Negatively Regulates Adipogenesis , 2013, The Journal of Biological Chemistry.

[33]  José M. Bioucas-Dias,et al.  Vertex Component Analysis: A~Fast Algorithm to Extract Endmembers Spectra from Hyperspectral Data , 2003, IbPRIA.

[34]  Avinash C. Kak,et al.  PCA versus LDA , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[35]  YuChen Xiang,et al.  SNR enhancement in brillouin microspectroscopy using spectrum reconstruction , 2020, Biomedical optics express.

[36]  R. Dasari,et al.  Prospects for in vivo Raman spectroscopy , 2000 .

[37]  J. H. Ward Hierarchical Grouping to Optimize an Objective Function , 1963 .

[38]  P. Török,et al.  Precision and informational limits in inelastic optical spectroscopy , 2019, Scientific Reports.

[39]  Adam J Engler,et al.  Preparation of Hydrogel Substrates with Tunable Mechanical Properties , 2010, Current protocols in cell biology.

[40]  Antonio J. Plaza,et al.  Minimum Volume Simplex Analysis: A Fast Algorithm for Linear Hyperspectral Unmixing , 2015, IEEE Transactions on Geoscience and Remote Sensing.

[41]  Giuliano Scarcelli,et al.  Evaluating biomechanical properties of murine embryos using Brillouin microscopy and optical coherence tomography. , 2017, Journal of biomedical optics.

[42]  R. Prevedel,et al.  Recent progress and current opinions in Brillouin microscopy for life science applications , 2020, Biophysical Reviews.

[43]  A. de Juan,et al.  Multivariate unmixing approaches on Raman images of plant cell walls: new insights or overinterpretation of results? , 2018, Plant Methods.

[44]  Sarah E. Bohndiek,et al.  Deep learning applied to hyperspectral endoscopy for online spectral classification , 2020, Scientific Reports.

[45]  Jürgen Popp,et al.  Discriminating isogenic cancer cells and identifying altered unsaturated fatty acid content as associated with metastasis status, using k-means clustering and partial least squares-discriminant analysis of Raman maps. , 2010, Analytical chemistry.

[46]  N. Koukourakis,et al.  Impulsive stimulated Brillouin microscopy for non-contact, fast mechanical investigations of hydrogels. , 2019, Optics express.

[47]  William J. Polacheck,et al.  Noncontact three-dimensional mapping of intracellular hydro-mechanical properties by Brillouin microscopy , 2015, Nature Methods.

[48]  G. Enikolopov,et al.  Role of pericytes in skeletal muscle regeneration and fat accumulation. , 2013, Stem cells and development.

[49]  K. Elsayad Spectral Phasor Analysis for Brillouin Microspectroscopy , 2019, Front. Phys..

[50]  Christoph Krafft,et al.  Classification of malignant gliomas by infrared spectroscopic imaging and linear discriminant analysis , 2007, Analytical and bioanalytical chemistry.

[51]  R. Fisher THE STATISTICAL UTILIZATION OF MULTIPLE MEASUREMENTS , 1938 .

[52]  Hans-Peter Kriegel,et al.  A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise , 1996, KDD.

[53]  W F Velicer,et al.  Component Analysis versus Common Factor Analysis: Some issues in Selecting an Appropriate Procedure. , 1990, Multivariate behavioral research.

[54]  K B Li,et al.  Distributed parallel processing for multidimensional maximum entropy reconstruction. , 1998, Journal of magnetic resonance.

[55]  Hugh Barr,et al.  Raman spectroscopy, a potential tool for the objective identification and classification of neoplasia in Barrett's oesophagus , 2003, The Journal of pathology.

[56]  G. Scarcelli,et al.  Single etalon design for two-stage cross-axis VIPA spectroscopy. , 2019, Biomedical optics express.

[57]  Silvia Caponi,et al.  Non-contact mechanical and chemical analysis of single living cells by microspectroscopic techniques , 2017, Light: Science & Applications.

[58]  Alfred O. Hero,et al.  Nonlinear Unmixing of Hyperspectral Images: Models and Algorithms , 2013, IEEE Signal Processing Magazine.