Hostname: page-component-78c5997874-fbnjt Total loading time: 0 Render date: 2024-11-20T04:48:37.348Z Has data issue: false hasContentIssue false
  • English
  • Français

Semi-Supervised Nonnegative Matrix Factorization of Wide-Field Fluorescence Microscopic Images for Tissue Diagnosis

Published online by Cambridge University Press:  14 April 2020

Shania M. Soman ,
Charuvil Radhakrishna Pillai Rekha ,
Hema Santhakumar ,
Uttamchand Narendrakumar  and
Ramapurath S. Jayasree Open the ORCID record for Ramapurath S. Jayasree [Opens in a new window]
Shania M. Soman
Affiliation:
School of Electronics and Engineering, VIT University, Vellore, TamilNadu632014, India
Charuvil Radhakrishna Pillai Rekha
Affiliation:
Division of Biophotonics and Imaging, Sree Chitra Tirunal Institute of Medical Science and Technology, Trivandrum, Kerala695012, India
Hema Santhakumar
Affiliation:
Division of Biophotonics and Imaging, Sree Chitra Tirunal Institute of Medical Science and Technology, Trivandrum, Kerala695012, India
Uttamchand Narendrakumar
Affiliation:
School of Mechanical Engineering, VIT University, Vellore, TamilNadu632014, India
Ramapurath S. Jayasree*
Affiliation:
Division of Biophotonics and Imaging, Sree Chitra Tirunal Institute of Medical Science and Technology, Trivandrum, Kerala695012, India
*
*Author for correspondence: Ramapurath S. Jayasree, E-mail: [email protected]
Get access

Abstract

This study tests the use of a constrained nonnegative matrix factorization (NMF) algorithm to explore the comparatively new field of chemometric microscopy to support tissue diagnosis. The algorithm can extract the spectral signature and the absolute concentration map of endogenous fluorophores from wide-field microscopic images. The resultant data distinguished normal and fibrous calvarial tissues, based on the changes in their spectral signatures. The absolute concentration map of endogenous fluorophores, nicotinamide adenine dinucleotide (NADH), flavin adenine dinucleotide (FAD), and lipofuscin were derived from microscopic images and compared with the fluorescence from pure fluorophores. While the absolute concentration of NADH increased, the same of FAD and lipofuscin decreased from a normal to fibrous calvarial condition. An increase in the optical redox ratio, possibly due to the metabolic changes during the development of fibrosis, was observed. Differentiating tissue types using the absolute concentration map was found to be considerably more precise than that achievable with relative concentration. The quantification of fluorophores with reference to the absolute concentration map can eliminate uncertainties due to system responses or measurement details, thereby generating more biologically apposite data. Wide-field microscopy augmented with a constrained NMF algorithm could emerge as an advanced diagnostic tool, potentially heralding the emergence of chemometric microscopy.

Keywords

chemometric microscopyendogenous fluorescencenonnegative matrix factorizationoptical redox ratiospatial distributionspatial diversity
Type
Software and Instrumentation
Information
Microscopy and Microanalysis , Volume 26 , Issue 3 , June 2020 , pp. 419 - 428
Copyright
Copyright © Microscopy Society of America 2020

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Alfano, RR, Das, BB, Cleary, J, Romulo, P & Celmer, JE (1991). Light sheds light on cancer — Distinguishing malignant tumors from benign tissues and tumors. Bull N Y Acad Med 67(2), 143–150.Google ScholarPubMed
Alfano, RR, Tang, GC, Pradhan, A, Lam, W, Daniel, SJC & Qpher, E (1987). Fluorescence spectra from cancerous and normal human breast and lung tissue. IEEE J Quantum Electron 23(10), 1806–1811.CrossRefGoogle Scholar
Andrezej, C & Pando, G (2003). Blind source separation algorithms with matrix constraints. IEICE Trans Fundam 1(1), 1–9.Google Scholar
Aubin, JE (1979). Autofluorescence. J Histochem Cytochem 27, 36–43.CrossRefGoogle ScholarPubMed
Benson, RC, Meyer, RA, Zaruba, ME & McKhann, GM (1979). Cellular autofluorescence — Is it due to Flavins? J Histochem Cytochem 27(1), 44–48.CrossRefGoogle ScholarPubMed
Chance, B, Williamson, JR, Jamieson, D & Schoener, B (1965). Properties and kinetics of reduced pyridine nucleotide fluorecence of the isolated and in vivo rat heart. Biochem Zeit 341, 377.Google Scholar
Chang, C & Lin, C (2011). LIBSVM: A library for support vector machines. ACM Trans Intell Syst Technol 2(3), 27.CrossRefGoogle Scholar
Christian, FW, Wei, M, Brian, SG, Sijia, L, Gary, HR, Chengwei, H, Jason, TC, Jing, KX & Sunney, X (2008). Label-free biomedical imaging with high sensitivity by stimulated Raman Scattering Microscopy. Science 322, 1857–1862.Google Scholar
Cichocki, A, Zdunek, R & Amari, S (2006). New Algorithms for Non-Negative Matrix Factorization in Applications to Blind Source Separation Riken. In IEEE International Conference on Acoustics Speech and Signal Processing Proceedings, 14–19 May, Toulouse, FranceGoogle Scholar
Fatakdawala, H, Poti, S, Zhou, F, Sun, Y, Bec, J, Liu, J, Yankelevich, DR, Tinling, SP, Gandour-edwards, RF, Farwell, DG & Marcu, L (2013). Multimodal in vivo imaging of oral cancer using fluorescence lifetime, photoacoustic and ultrasound techniques. Biomed Optics Express 4(9), 1724–1741.CrossRefGoogle ScholarPubMed
Fujimoto, JG, Pitris, C, Boppart, SA & Brezinski, ME (2000). Optical coherence tomography: An emerging technology for biomedical imaging and optical biopsy. Neoplasia 2(1–2), 9–25.CrossRefGoogle ScholarPubMed
Ge, H, Yan, Z, Dou, J, Wang, Z & Wang, ZQ (2018). A semisupervised framework for automatic image annotation based on graph embedding and multiview nonnegative matrix factorization. Math Probl Eng 2018, 1–11. doi: 10.1155/2018/5987906.Google Scholar
Graf, BW & Boppart, SA (2012). Multimodal in vivo skin imaging with integrated optical coherence and multiphoton microscopy. IEEE J Sel Top Quantum Electron 18(4), 1280–1286.CrossRefGoogle ScholarPubMed
Haris, PS, Anita, B, Jayasree, RS & Gupta, AK (2009). Autofluorescence spectroscopy for the in vivo evaluation of oral submucous fibrosis. Photomed Laser Surg 27(5), 757–761.CrossRefGoogle ScholarPubMed
Hoyer, P (2002). Non-negative sparse coding. In Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing, pp. 557–565.CrossRefGoogle Scholar
Hoyer, PO (2004). Non-negative matrix factorization with sparseness constraints. J Mach Learn Res 5, 1457–1469.Google Scholar
Hucker, WJ, Ripplinger, CM, Fleming, CP & Rollins, AM (2009). Bimodal biophotonic imaging of the structure-function relationship in cardiac tissue. J Biomed Optics 13(5), 1–15.Google Scholar
Jiji, S, Smitha, KA, Gupta, AK, Pillai, VPM & Jayasree, RS (2013). Segmentation and volumetric analysis of the caudate nucleus in Alzheimer's disease. Eur J Radiol 82(9), 1525–1530.CrossRefGoogle ScholarPubMed
Kim, P, Puoris, M, Cote, D, Lin, CP & Yun, SH (2009). In vivo confocal and multiphoton microendoscopy. J Biomed Opt 13(1), 1–7.Google Scholar
Kinan, A, Lisa, RG, Muldoon, TJ, Kyle, QP & Narasimhan, R (2016). Optical redox ratio identifies metastatic potential-dependent changes in breast cancer cell metabolism. Biomed Optics Express 7(11), 4364–4374.Google Scholar
Konig, K, Ehlers, A, Riemann, I, Schenkl, S & Bu, R (2007). Clinical twophoton microendoscopy. Microsc Res Technol 402, 398–402.CrossRefGoogle Scholar
Kumar, N, Uppala, P, Duddu, K, Sreedhar, H, Varma, V, Guzman, G, Walsh, M & Sethi, A (2019). Hyperspectral tissue image segmentation using semi-supervised NMF and hierarchical clustering. IEEE Trans Med Imaging 38(5), 1304–1313.CrossRefGoogle ScholarPubMed
Lee, DD, Hill, M & Seung, HS (2001) Algorithms for Non-Negative Matrix Factorization.Google Scholar
Lee, DD & Seung, HS (2000). Learning the parts of objects by non-negative matrix factorization. Nature 401, 788–791.CrossRefGoogle Scholar
Liberti, MV & Locasale, JW (2017). The Warburg effect: How does it benefit cancer cells? Trends Biochem Sci 41(3), 211–218.CrossRefGoogle Scholar
Nazeer, SS, Asish, R, Venugopal, C, Anita, B, Gupta, AK & Jayasree, RS (2014 a). Noninvasive assessment of the risk of tobacco abuse in oral mucosa using fluorescence spectroscopy: A clinical approach tobacco abuse in oral mucosa using fluorescence. J Biomed Opt 19, 057013.CrossRefGoogle ScholarPubMed
Nazeer, SS, Nazeer, SS, Saraswathy, A, Gupta, K & Jayasree, RS (2013). Fluorescence spectroscopy as a highly potential single-entity tool to identify chromophores and fluorophores: Study on neoplastic human brain lesions entity tool to identify chromophores and fluorophores. J Biomed Opt 18(6), 067002.10.1117/1.JBO.18.6.067002CrossRefGoogle Scholar
Nazeer, SS, Sandhyamani, S & Jayasree, RS (2015). Optical diagnosis of the progression and reversal of CCl4-induced liver injury in rodent model using minimally invasive autofluorescence spectroscopy. Analyst 140, 3773–3780.CrossRefGoogle ScholarPubMed
Nazeer, SS, Saraswathy, A, Gupta, AK & Jayasree, RS (2014 b). Fluorescence spectroscopy to discriminate neoplastic human brain lesions: A study using the spectral intensity ratio and multivariate linear discriminant analysis. Laser Phys 24(2), 25602.CrossRefGoogle Scholar
Ostrander, JH, Mcmahon, CM, Lem, S, Millon, SR, Brown, JQ, Seewaldt, VL & Ramanujam, N (2010). Optical redox ratio differentiates breast cancer cell lines based on estrogen receptor status. Cancer Res 70(11), 4759–4767.CrossRefGoogle ScholarPubMed
Paatero, P & Tappert, U (1994). Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values. Environmetrics 5, 111–126.CrossRefGoogle Scholar
Pavlova, I, Sokolov, K, Drezek, R, Malpica, A, Follen, M & Richards-kortum, R (2003). Microanatomical and biochemical origins of normal and precancerous cervical autofluorescence using laser-scanning fluorescence. Photochem Photobiol 77(5), 550–555.2.0.CO;2>CrossRefGoogle ScholarPubMed
Podoleanu, AG, Dobre, GM, Cucu, RG, Rosen, R, Garcia, P, Nieto, J, Will, D, Gentile, R, Muldoon, T, Walsh, J, Yannuzzi, LA, Fisher, Y, Orlock, D, Weitz, R, Rogers, JA, Dunne, S & Boxer, A (2004). Combined multiplanar optical coherence tomography and confocal scanning ophthalmoscopy. J Biomed Opt 9(1), 86–93.CrossRefGoogle ScholarPubMed
Pu, Y, Wang, W, Yang, Y & Alfano, RR (2012). Stokes shift spectroscopy highlights differences of cancerous and normal human tissues. Opt Lett 37(16), 33360–3362.CrossRefGoogle ScholarPubMed
Rakotomamonjy, A (2004). Optimizing Area under ROC Curve with SVMs. In Advances in Neural Information Processing Systems.Google Scholar
Saraswathy, A, Jayasree, RS, Baiju, KV, Gupta, AK & Pillai, VPM (2009). Optimum wavelength for the differentiation of brain tumor tissue using autofluorescence spectroscopy. Photomed Laser Surg 27(3), 425–433.CrossRefGoogle ScholarPubMed
Shaiju, SN, Ariya, S, Asish, R, Haris, SP, Anita, B, Arun Kumar, G & Jayasree, RS (2011). Habits with killer instincts: In vivo analysis on the severity of oral mucosal alterations using autofluorescence spectroscopy. J Biomed Opt 16, 087006.CrossRefGoogle Scholar
Smitha, KA, Gupta, AK & Jayasree, RS (2013). Total magnitude of diffusion tensor imaging as an effective tool for the differentiation of glioma. Eur J Radiol 82(5), 857–861.CrossRefGoogle ScholarPubMed
Smitha, KA, Gupta, AK & Jayasree, RS (2015 a). Fractal analysis: Fractal dimension and lacunarity from MR images for differentiating the grades of glioma. Phys Med Biol 60, 6937–6947.CrossRefGoogle ScholarPubMed
Smitha, KA, Gupta, AK & Jayasree, RS (2015 b). Relative percentage signal intensity recovery of perfusion metrics — An efficient tool for differentiating grades of glioma. Br J Radiol 88, 20140784.10.1259/bjr.20140784CrossRefGoogle ScholarPubMed
Uppal, A & Gupta, PK (2003). Measurement of NADH concentration in normal and malignant human tissues from breast and oral cavity. Biotechnol Appl Biochem 37(1), 45–50.CrossRefGoogle ScholarPubMed
Vander Heiden, MG, Cantley, LC, Thompson, CB, Mammalian, P, Exhibit, C & Metabolism, A (2009). Understanding the Warburg effect: Cell proliferation. Science 324, 1029–1034.CrossRefGoogle ScholarPubMed
Venugopal, C, Nazeer, SS, Balan, A & Jayasree, RS (2013). Autofluorescence spectroscopy augmented by multivariate analysis as a potential noninvasive tool for early diagnosis of oral cavity disorders. Photomed Laser Surg 31(12), 605–612.CrossRefGoogle ScholarPubMed
Wang, D, Gao, X & Wang, X (2016). Semi-supervised nonnegative matrix factorization via constraint propagation. IEEE Trans Cybern 46(1), 233–244.CrossRefGoogle ScholarPubMed
Warburg, O (1956). Injuring of respiration the origin of cancer cells. Science 123(3191), 309–314.CrossRefGoogle Scholar
Xu, C, Vinegoni, C, Ralston, TS, Luo, W, Tan, W & Boppart, SA (2006). Spectroscopic spectral-domain optical coherence microscopy. Opt Lett 31(8), 1079–1081.CrossRefGoogle ScholarPubMed
Xu, J, Xiang, L, Wang, G, Ganesan, S, Feldman, M, Shih, NN, Gilmore, H & Madabhushi, A (2015). Sparse non-negative matrix factorization (SNMF) based color unmixing for breast histopathological image analysis. Comput Med Imaging Grap 46, 20–29. doi:10.1016/j.compmedimag.2015.04.002.CrossRefGoogle ScholarPubMed
Xu, M & Wang, LV (2006). Photoacoustic imaging in biomedicine. Rev Sci Instrum 77, 041101.CrossRefGoogle Scholar
Yang, P, Tang, CG, Wang, WB, Savage, HE, Schantz, SP & Alfano, RR (2011). Native fluorescence spectroscopic evaluation of chemotherapeutic effects on malignant cells using nonnegative matrix factorization analysis. Technol Cancer Res Treat 10(2).Google Scholar
Yang, P, Wubao, W, Tang, G & Robert, AR (2013). Changes of collagen and nicotinamide adenine dinucleotide in human cancerous and normal prostate tissues studied using native fluorescence. J Biomed Opt 15(4), 1–5.Google Scholar
Yu, Q & Heikal, AA (2009). Two-photon autofluorescence dynamics imaging reveals sensitivity of intracellular NADH concentration and conformation to cell physiology at the single-cell level. J Photochem Photobiol B 95(1), 46–57.CrossRefGoogle ScholarPubMed
Zudaire, I & Ortiz-de-solorzano, C (2013). Efficient blind spectral unmixing of fluorescently labeled samples using multi-layer non-negative matrix factorization. PLoS ONE 8(11), e78504.Google Scholar