Deep data analysis via physically constrained linear unmixing: universal framework, domain examples, and a community-wide platform
Tóm tắt
Many spectral responses in materials science, physics, and chemistry experiments can be characterized as resulting from the superposition of a number of more basic individual spectra. In this context, unmixing is defined as the problem of determining the individual spectra, given measurements of multiple spectra that are spatially resolved across samples, as well as the determination of the corresponding abundance maps indicating the local weighting of each individual spectrum. Matrix factorization is a popular linear unmixing technique that considers that the mixture model between the individual spectra and the spatial maps is linear. Here, we present a tutorial paper targeted at domain scientists to introduce linear unmixing techniques, to facilitate greater understanding of spectroscopic imaging data. We detail a matrix factorization framework that can incorporate different domain information through various parameters of the matrix factorization method. We demonstrate many domain-specific examples to explain the expressivity of the matrix factorization framework and show how the appropriate use of domain-specific constraints such as non-negativity and sum-to-one abundance result in physically meaningful spectral decompositions that are more readily interpretable. Our aim is not only to explain the off-the-shelf available tools, but to add additional constraints when ready-made algorithms are unavailable for the task. All examples use the scalable open source implementation from
https://github.com/ramkikannan/nmflibrary
that can run from small laptops to supercomputers, creating a user-wide platform for rapid dissemination and adoption across scientific disciplines.
Tài liệu tham khảo
Pennycook, S.J., Varela, M., Lupini, A.R., Oxley, M.P., Chisholm, M.F.: Atomic-resolution spectroscopic imaging: past, present and future. J. Electron Microsc. 58, 87–97 (2009)
Zhou, W., Kapetanakis, M.D., Prange, M.P., Pantelides, S.T., Pennycook, S.J., Idrobo, J.C.: Direct determination of the chemical bonding of individual impurities in graphene. Phys. Rev. Lett. 109, 206803 (2012)
Suenaga, K., Koshino, M.: Atom-by-atom spectroscopy at graphene edge. Nature 468, 1088–1090 (2010)
Varela, M., Gazquez, J., Pennycook, S.J.: STEM-EELS imaging of complex oxides and interfaces. MRS Bull. 37, 29–35 (2012)
Kumar, A., Ehara, Y., Wada, A., Funakubo, H., Griggio, F., Trolier-McKinstry, S., et al.: Dynamic piezoresponse force microscopy: spatially resolved probing of polarization dynamics in time and voltage domains. J. Appl. Phys. 112, 052021 (2012)
Guo, S., Jesse, S., Kalnaus, S., Balke, N., Daniel, C., Kalinin, S.V.: Direct mapping of ion diffusion times on LiCoO(2) surfaces with nanometer resolution. J. Electrochem. Soc. 158, A982–A990 (2011)
Kalinin, S., Balke, N., Jesse, S., Tselev, A., Kumar, A., Arruda, T.M., et al.: Li-ion dynamics and reactivity on the nanoscale. Mater. Today 14, 548–558 (2011)
Jesse, S., Balke, N., Eliseev, E., Tselev, A., Dudney, N.J., Morozovska, A.N., et al.: Direct mapping of ionic transport in a si anode on the nanoscale: time domain electrochemical strain spectroscopy study. ACS Nano 5, 9682–9695 (2011)
Kano, H., Segawa, H., Okuno, M., Leproux, P., Couderc, V.: Hyperspectral coherent Raman imaging—principle, theory, instrumentation, and applications to life sciences. J. Raman Spectrosc. 47, 116–123 (2016)
Wabuyele, M.B., Yan, F., Griffin, G.D., Vo-Dinh, T.: Hyperspectral surface-enhanced Raman imaging of labeled silver nanoparticles in single cells. Rev. Sci. Instrum. 76, 063710 (2005)
Fu, D., Holtom, G., Freudiger, C., Zhang, X., Xie, X.S.: Hyperspectral imaging with stimulated raman scattering by chirped femtosecond lasers. J. Phys. Chem. B 117, 4634–4640 (2013)
Bouillard, J.-S.G., Dickson, W., Wurtz, G.A., Zayats, A.V.: Near-field hyperspectral optical imaging. ChemPhysChem 15, 619–629 (2014)
Jung, S., Foston, M., Kalluri, U.C., Tuskan, G.A., Ragauskas, A.J.: 3D chemical image using TOF-SIMS revealing the biopolymer component spatial and lateral distributions in biomass. Angew. Chem. Int. Ed. 51, 12005–12008 (2012)
Ievlev, A.V., Maksymovych, P., Trassin, M., Seidel, J., Ramesh, R., Kalinin, S.V., et al.: Chemical state evolution in ferroelectric films during tip-induced polarization and electroresistive switching. ACS Appl. Mater. Interfaces. 8, 29588–29593 (2016)
McDonnell, L.A., Heeren, R.M.A.: Imaging mass spectrometry. Mass Spectrom. Rev. 26, 606–643 (2007)
Zimmermann, T.: Spectral imaging and linear unmixing in light microscopy. In: Rietdorf, T., Denert, E. (eds.) Microscopy Techniques: −/−, pp. 245–265. Springer, Berlin (2005)
Peckner, R., Myers, S.A., Egertson, J.D., Johnson, R.S., Carr, S.A., MacCoss, M.J., et al.: Specter: linear deconvolution as a new paradigm for targeted analysis of data-independent acquisition mass spectrometry proteomics. bioRxiv (2017). https://doi.org/10.1101/152744
Kalinin, S.V., Jesse, S., Rodriguez, B.J., Shin, J., Baddorf, A.P., Lee, H.N., et al.: Spatial resolution, information limit, and contrast transfer in piezoresponse force microscopy. Nanotechnology 17, 3400 (2006)
Collins, L., Okatan, M.B., Li, Q., Kravenchenko, I.I., Lavrik, N.V., Kalinin, S.V., et al.: Quantitative 3D-KPFM imaging with simultaneous electrostatic force and force gradient detection. Nanotechnology 26, 175707 (2015)
Collins, L., Belianinov, A., Somnath, S., Balke, N., Kalinin, S.V., Jesse, S.: Full data acquisition in Kelvin probe force microscopy: mapping dynamic electric phenomena in real space. Sci. Rep. 6, 30557 (2016)
Cohen, G., Halpern, E., Nanayakkara, S.U., Luther, J.M., Held, C., Bennewitz, R., et al.: Reconstruction of surface potential from Kelvin probe force microscopy images. Nanotechnology 24, 295702 (2013)
Kirkland, E.J.: Linear image approximations. In: Kirkland, E.J. (ed.) Advanced Computing in Electron Microscopy, pp. 29–60. Springer, Boston (2010)
Björck, Å: Numerical Methods for Least Squares Problems. SIAM (1996)
Kannan, R.: Scalable and Distributed Constrained Low Rank Approximations. Georgia Institute of Technology, Atlanta (2016)
Kim, J., He, Y., Park, H.: Algorithms for nonnegative matrix and tensor factorizations: a unified view based on block coordinate descent framework. J. Glob. Optim. 58, 285–319 (2014)
Kannan, R., Ishteva, M., Drake, B., Park, H.: Bounded matrix low rank approximation. In: Non-negative Matrix Factorization Techniques, pp. 89–118. Springer, Berlin (2016)
Kannan, R., Ishteva, M., Park, H.: Bounded matrix factorization for recommender system. Knowl. Inf. Syst. 39, 491–511 (2014)
Keshava, N., Mustard, J.F.: Spectral unmixing. IEEE Signal Process. Mag. 19, 44–57 (2002)
Dobigeon, N., Moussaoui, S., Coulon, M., Tourneret, J.Y., Hero, A.O.: Joint Bayesian endmember extraction and linear unmixing for hyperspectral imagery. IEEE Trans. Signal Process. 57, 4355–4368 (2009)
Pearson, K.: LIII. On lines and planes of closest fit to systems of points in space. In: Philosophical Magazine Series 6, vol. 2, pp. 559–572. (1901)
Jolliffe, I.: Principal component analysis. In: Wiley StatsRef: Statistics Reference Online. Wiley, London (2014)
Medina, J.M., Pereira, L.M., Correia, H.T., Nascimento, S.M.C.: Hyperspectral optical imaging of human iris in vivo: characteristics of reflectance spectra. J. Biomed. Opt. 16, 076001 (2011)
Bonnet, N.: Artificial intelligence and pattern recognition techniques in microscope image processing and analysis. In: Hawkes, P.W. (ed.) Advances in Imaging and Electron Physics, vol. 114, pp. 1–77. Elsevier Academic Press Inc, San Diego (2000)
Bonnet, N.: Multivariate statistical methods for the analysis of microscope image series: applications in materials science. J. Microsc. Oxf. 190, 2–18 (1998)
Serin, V., Andrieu, S., Serra, R., Bonell, F., Tiusan, C., Calmels, L., et al.: TEM and EELS measurements of interface roughness in epitaxial Fe/MgO/Fe magnetic tunnel junctions. Phys. Rev. B 79, 144413 (2009)
Bosman, M., Watanabe, M., Alexander, D.T.L., Keast, V.J.: Mapping chemical and bonding information using multivariate analysis of electron energy-loss spectrum images. Ultramicroscopy 106, 1024–1032 (2006)
Biesinger, M.C., Paepegaey, P.-Y., McIntyre, N.S., Harbottle, R.R., Petersen, N.O.: Principal component analysis of TOF-SIMS images of organic monolayers. Anal. Chem. 74, 5711–5716 (2002)
Race, A.M., Steven, R.T., Palmer, A.D., Styles, I.B., Bunch, J.: Memory efficient principal component analysis for the dimensionality reduction of large mass spectrometry imaging data sets. Anal. Chem. 85, 3071–3078 (2013)
Kalinin, S.V., Rodriguez, B.J., Budai, J.D., Jesse, S., Morozovska, A.N., Bokov, A.A., et al.: Direct evidence of mesoscopic dynamic heterogeneities at the surfaces of ergodic ferroelectric relaxors. Phys. Rev. B 81, 064107 (2010)
Jesse, S., Kalinin, S.V.: Principal component and spatial correlation analysis of spectroscopic-imaging data in scanning probe microscopy. Nanotechnology 20, 085714 (2009)
Kalinin, S.V., Rodriguez, B.J., Jesse, S., Morozovska, A.N., Bokov, A.A., Ye, Z.G.: Spatial distribution of relaxation behavior on the surface of a ferroelectric relaxor in the ergodic phase. Appl. Phys. Lett. 95, 142902 (2009)
Ovchinnikov, O.S., Jesse, S., Bintacchit, P., Trolier-McKinstry, S., Kalinin, S.V.: Disorder identification in hysteresis data: recognition analysis of the random-bond-random-field ising model. Phys. Rev. Lett. 103, 157203 (2009)
Koren, Y., Bell, R., Volinsky, C.: Matrix factorization techniques for recommender systems. Computer 42(8), 30–37 (2009). https://doi.org/10.1109/MC.2009.263
Shiga, M., Muto, S., Tatsumi, K., Tsuda, K.: Matrix factorization for automatic chemical mapping from electron microscopic spectral imaging datasets. Trans. Mater. Res. Soc. Jpn 41, 333–336 (2016)
Shiga, M., Tatsumi, K., Muto, S., Tsuda, K., Yamamoto, Y., Mori, T., et al.: Sparse modeling of EELS and EDX spectral imaging data by nonnegative matrix factorization. Ultramicroscopy 170, 43–59 (2016)
Kuang, D., Park, H.: Fast rank-2 nonnegative matrix factorization for hierarchical document clustering. In: Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 739–747. (2013)
Xu, W., Liu, X., Gong, Y.: Document clustering based on non-negative matrix factorization. In: Proceedings of the 26th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 267–273. (2003)
Candes, E., Recht, B.: Exact matrix completion via convex optimization. Commun. ACM 55, 111–119 (2012)
Zhou, T., Tao, D.: Godec: randomized low-rank & sparse matrix decomposition in noisy case. In: International Conference on Machine Learning. (2011)
Kannan, R., Ballard, G., Park, H.: MPI-FAUN: an MPI-based framework for alternating-updating nonnegative matrix factorization. IEEE Trans. Knowl. Data Eng. 30(3), 544–558 (2018)
Ding, C., He, X., Simon, H.D.: On the equivalence of nonnegative matrix factorization and spectral clustering. In: Proceedings of the 2005 SIAM International Conference on Data Mining, pp. 606–610. (2005)
Choo, J., Lee, C., Clarkson, E., Liu, Z., Lee, H., Chau, D.H.P., et al.: VisIRR: interactive visual information retrieval and recommendation for large-scale document data. Georgia Institute of Technology, Atlanta (2013)
Choo, J., Lee, C., Kim, H., Lee, H., Liu, Z., Kannan, R., et al.: VisIRR: visual analytics for information retrieval and recommendation with large-scale document data. In: Visual Analytics Science and Technology (VAST), 2014 IEEE Conference on, pp. 243–244. (2014)
Kim, J., Park, H.: Sparse nonnegative matrix factorization for clustering. Georgia Institute of Technology, Atlanta (2008)
Bishop, C.M.: Pattern recognition and machine learning. Springer, Berlin (2006)
Wit, E., Heuvel, E.V.D., Romeijn, J.-W.: ‘All models are wrong…’: an introduction to model uncertainty. Stat. Neerlandica 66, 217–236 (2012)
Bischl, B., Richter, J., Bossek, J., Horn, D., Thomas, J., Lang, M.: mlrMBO: a modular framework for model-based optimization of expensive black-box functions. arXiv preprint arXiv:1703.03373 (2017)
Bergstra, J., Yamins, D., Cox, D.D.: Hyperopt: a python library for optimizing the hyperparameters of machine learning algorithms. (2013)
Singh, A., Gordon, G.: A unified view of matrix factorization models. In: Machine Learning and Knowledge Discovery in Databases, pp. 358–373. (2008)
Collins, M., Dasgupta, S., Schapire, R.E.: A generalization of principal component analysis to the exponential family
Lee, D.D., Seung, H.S.: Learning the parts of objects by non-negative matrix factorization. Nature 401, 788–791 (1999)
Cai, D., He, X., Han, J., Huang, T.S.: Graph regularized nonnegative matrix factorization for data representation. IEEE Trans. Pattern Anal. Mach. Intell. 33, 1548–1560 (2011)
Golub, G.H., Van Loan, C.F.: Matrix Computations. JHU Press, Baltimore (2012)
Collins, M., Dasgupta, S., Schapire, R.E.: A generalization of principal components analysis to the exponential family. In: Advances in Neural Information Processing Systems, pp. 617–624. (2001)
Lee, D.D., Sebastian, S.H.: Learning the parts of objects by non-negative matrix factorization. Nature 401, 788–791 (1999)
Singh, A.P., Gordon, G.J.: A unified view of matrix factorization models. In: Joint European Conference on Machine Learning and Knowledge Discovery in Databases, pp. 358–373, Berlin (2008)
Pacholski, M.L., Winograd, N.: Imaging with mass spectrometry. Chem. Rev. 99, 2977 (1999)
Ievlev, A.V., Belianinov, A., Jesse, S., Allison, D.P., Doktycz, M.J., Retterer, S.T., et al.: Automated interpretation and extraction of topographic information from time of flight secondary ion mass spectrometry data. Sci. Rep. 7, 17099 (2017)
Seidel, J., Trassin, M., Zhang, Y., Maksymovych, P., Uhlig, T., Milde, P., et al.: Electronic properties of isosymmetric phase boundaries in highly strained Ca-Doped BiFeO3. Adv. Mater. 26, 4376–4380 (2014)
Ievlev, A.V., Maksymovych, P., Trassin, M., Seidel, J., Ramesh, R., Kalinin, S.V., et al.: Chemical state evolution in ferroelectric films during tip-induced polarization and electroresistive switching. ACS Appl. Mater. Interfaces. 8, 29588–29593 (2016)
Kalinin, S.V., Pennycook, S.J.: Microscopy: hasten high resolution. Nature 515, 487 (2014)
He, Q., Woo, J., Belianinov, A., Guliants, V.V., Borisevich, A.Y.: Better catalysts through microscopy: mesoscale M1/M2 intergrowth in Molybdenum–Vanadium based complex oxide catalysts for propane ammoxidation. ACS Nano 9, 3470–3478 (2015)
Vasudevan, R.K., Ziatdinov, M., Jesse, S., Kalinin, S.V.: Phases and interfaces from real space atomically resolved data: physics-based deep data image analysis. Nano Lett. 16, 5574–5581 (2016)
Ziatdinov, M., Fujii, S., Kiguchi, M., Enoki, T., Jesse, S., Kalinin, S.V.: Data mining graphene: correlative analysis of structure and electronic degrees of freedom in graphenic monolayers with defects. Nanotechnology 27, 495703 (2016)
He, Q., Woo, J., Belianinov, A., Guliants, V.V., Borisevich, A.Y.: Better catalysts through microscopy: mesoscale M1/M2 Intergrowth in Molybdenum–Vanadium based complex oxide catalysts for propane ammoxidation. ACS Nano 9, 3470–3478 (2015)
Ziatdinov, M., Maksov, A., Li, L., Sefat, A.S., Maksymovych, P., Kalinin, S.V.: Deep data mining in a real space: separation of intertwined electronic responses in a lightly doped BaFe2As2. Nanotechnology 27, 475706 (2016)
Sefat, A.S., Marty, K., Christianson, A.D., Saparov, B., McGuire, M.A., Lumsden, M.D., et al.: Effect of molybdenum 4d hole substitution in BaFe2As2. Phys. Rev. B 85, 024503 (2012)
Li, L., Cao, H., McGuire, M.A., Kim, J.S., Stewart, G.R., Sefat, A.S.: Role of magnetism in superconductivity of BaFe2As2: study of 5d Au-doped crystals. Phys. Rev. B 92, 094504 (2015)
Fäth, M., Freisem, S., Menovsky, A.A., Tomioka, Y., Aarts, J., Mydosh, J.A.: Spatially inhomogeneous metal-insulator transition in doped manganites. Science 285(5433), 1540–1542 (1999)
Holt, M., Harder, R., Winarski, R., Rose, V.: Nanoscale hard X-ray microscopy methods for materials studies. Ann. Rev. Mater. Res. 43, 183–211 (2013)
Tanner, B.K.: X-ray Diffraction Topography, vol. 10. Pergamon (1976)
Larson, B.C., Yang, W., Ice, G.E., Budai, J.D., Tischler, J.Z.: Three-dimensional X-ray structural microscopy with submicrometre resolution. Nature 415, 887–890 (2002)
Ice, G.E., Budai, J.D., Pang, J.W.L.: The race to X-ray microbeam and nanobeam science. Science 334, 1234 (2011)
Hofmann, F., Abbey, B., Liu, W., Xu, R., Usher, B.F., Balaur, E., et al.: X-ray micro-beam characterization of lattice rotations and distortions due to an individual dislocation. Nat. Commun. 4, 2774 (2013)
Hruszkewycz, S.O., Highland, M.J., Holt, M.V., Kim, D., Folkman, C.M., Thompson, C., et al.: Imaging local polarization in ferroelectric thin films by coherent X-ray Bragg projection ptychography. Phys. Rev. Lett. 110, 177601 (2013)
Laanait, N., Zhang, Z., Schlepütz, C.M.: Imaging nanoscale lattice variations by machine learning of X-ray diffraction microscopy data. Nanotechnology 27, 1–10 (2016)
Laanait, N., Zhang, Z., Schlepütz, C.M., Vila-Comamala, J., Highland, M.J., Fenter, P.: Full-field X-ray reflection microscopy of epitaxial thin-films. J. Synchrotron Radiat. 21, 1252–1261 (2014)
Oh, S.H., Park, C.G.: Misfit strain relaxation by dislocations in SrRuO3/SrTiO3 (001) heteroepitaxy. J. Appl. Phys. 95, 4691–4704 (2004)
Koster, G., Klein, L., Siemons, W., Rijnders, G., Dodge, J.S., Eom, C.B., et al.: Structure, physical properties, and applications of SrRuO3 thin films. Rev. Mod. Phys. 84, 253–298 (2012)