Deconvolution Methods for Low Exposure Fluorescence Microscopy

Abstract

Deconvolution microscopy is the only viable form of fluorescence microscopy when we study sensitive cells such as yeasts with imaging experiments of longer durations. It makes the full utilisation of emission photons in order to make the actual or practical/obtainable resolution to be as close as to the theoretical resolution limit. The postprocessing step, known as the deconvolution, which refers to computational inversion of microscope's image blurring model, plays a crucial role in maximisation of the obtainable resolution. In particular, the obtainable resolution is determined by the mathematical technique used for stabilising the computational inversion process against the noise, which is known as regularisation. The regularisation is an expression of some prior knowledge about the structure of a typical image, and its suitability to the actual structure greatly determines the maximum resolution obtainable by a deconvolution method. Advancements in the design of regularisation methods that are specifically tuned for fluorescence imaging will make new type of experiments – with imaging durations longer than ever – possible.

Key Concepts

  • The problem of phototoxicity limits the imaging duration in 4D microscopy.
  • In 4D deconvolution microscopy, actual resolution is significantly lower than the theoretical resolution.
  • In 4D deconvolution microscopy, extending imaging duration causes loss of resolution, and setting a minimum requirement on actual resolution limits imaging duration.
  • Extent of the loss of resolution in 4D deconvolution microscopy is determined by the efficiency of the deconvolution method.
  • The efficiency of deconvolution method is determined by the suitability of the regularisation functional used in deconvolution.

Keywords: 4D microscopy; deconvolution; live microscopy; fluorescence microscopy; phototoxicity; regularisation

Figure 1. Imaging model of a widefield microscope. (a) Schematic of the imaging set‐up; (b) xz section of widefield amplitude OTF Ha(X, Y, Z); (c) xz section of intensity OTF H(X, Y, Z); (d) xz section of intensity PSF h(x, y, z).
Figure 2. Demonstration of the role deconvolution. (a) xz section of a widefield image measured from Drosophila spindle; (b) xz section of the same image corresponding to (a); (c) xy section of the image deconvolved from measured image corresponding to (a); (d) xz section of the same image corresponding to (c).
Figure 3. Demonstration of the effect of noise on resolution.
Figure 4. Comparing the performance of our recently published method ER‐Decon with SOTV (Lefkimmiatis et al., ), MLTL (Vonesch and Unser, ) and cMLE‐Huygens (commercial software). (a) xy and xz sections of the raw image measured from yeast vacuole with very low exposure; (b) z‐ and y‐projections of the raw image; the severity of noise level is evident from the fact that even the projected images do not show the location of the vacuole clearly; (c) xy and xz sections of the deconvolution result obtained from ER‐Decon; (d.1, e.1, f.1) xy and xz of the results obtained from SOTV, MLTL and cMLE‐Huygens with carefully tuned smoothing parameters; (d.2, e.2, f.2) results obtained from SOTV, MLTL and cMLE‐Huygens with smoothing parameter chosen to be slightly less than the optimal value; (d.3, e.3, f.3) results obtained from SOTV, MLTL and cMLE‐Huygens with even further reduced smoothing parameter.
close

References

Agard DA, Hiraoka Y and Sedat JW (1989) Three‐dimensional microscopy: image processing for high resolution subcellular imaging. 33rd Annual Technical Symposium, pp. 24–30. International Society for Optics and Photonics.

Arigovindan M, Fung JC, Elnatan D, et al. (2013) High‐resolution restoration of 3d structures from widefield images with extreme low signal‐to‐noise‐ratio. Proceedings of the National Academy of Sciences 110 (43): 17344–17349.

Beaudouin J, Gerlich D, Daigle N, Eils R and Ellenberg J (2002) Nuclear envelope breakdown proceeds by microtubule‐induced tearing of the lamina. Cell 108 (1): 83–96.

Conchello J‐A and Hansen EW (1990) Enhanced 3‐d reconstruction from confocal scanning microscope images. 1: deterministic and maximum likelihood reconstructions. Applied Optics 29 (26): 3795–3804.

Conchello J‐A, Kim JJ and Hansen EW (1994) Enhanced three‐dimensional reconstruction from confocal scanning microscope images. ii. depth discrimination versus signal‐to‐noise ratio in partially confocal images. Applied Optics 33 (17): 3740–3750.

Conchello J‐A (1995) Fluorescence photobleaching correction for expectation‐maximization algorithm. IS&T/SPIE's Symposium on Electronic Imaging: Science & Technology, pp 138–146. International Society for Optics and Photonics.

Conchello J‐A (1998) Superresolution and convergence properties of the expectation‐maximization algorithm for maximum‐likelihood deconvolution of incoherent images. Journal of the Optical Society of America. A, Optics, Image Science, and Vision 15 (10): 2609–2619.

Dey N, Blanc‐Feraud L, Zimmer C, et al. (2006) Richardson–Lucy algorithm with total variation regularization for 3d confocal microscope deconvolution. Microscopy Research and Technique 69 (4): 260–266.

Fung JC, Marshall WF, Dernburg A, Agard DA and Sedat JW (1998) Homologous chromosome pairing in drosophila melanogaster proceeds through multiple independent initiations. The Journal of Cell Biology 141 (1): 5–20.

Hammond AT and Glick BS (2000) Raising the speed limits for 4d fluorescence microscopy. Traffic 1 (12): 935–940.

Hom EF, Marchis F, Lee TK, et al. (2007) Aida: an adaptive image deconvolution algorithm with application to multi‐frame and three‐dimensional data. Journal of the Optical Society of America. A, Optics, Image Science, and Vision 24 (6): 1580–1600.

Kumari S and Mayor S (2008) Arf1 is directly involved in dynamin‐independent endocytosis. Nature Cell Biology 10 (1): 30–41.

Lane R (1996) Methods for maximum‐likelihood deconvolution. Journal of the Optical Society of America. A, Optics, Image Science, and Vision 13 (10): 1992–1998.

Lefkimmiatis S, Bourquard A and Unser M (2012) Hessian‐based regularization for 3‐d microscopy image restoration. 2012 9th IEEE International Symposium on Biomedical Imaging (ISBI), pp. 1731–1734. IEEE.

Masters BR (2010) The development of fluorescence microscopy. Encyclopedia of Life Sciences. Chichester: John Wiley & Sons.

Moir RD, Yoon M, Khuon S and Goldman RD (2000) Nuclear lamins a and b1 different pathways of assembly during nuclear envelope formation in living cells. The Journal of Cell Biology 151 (6): 1155–1168.

Oliveira JP, Bioucas‐Dias JM and Figueiredo MA (2009) Adaptive total variation image deblurring: a majorization–minimization approach. Signal Processing 89 (9): 1683–1693.

Pelham RJ and Chang F (2002) Actin dynamics in the contractile ring during cytokinesis in fission yeast. Nature 419 (6902): 82–86.

Preza C, Miller MI, Thomas LJ Jr McNally JG, et al. (1992) Regularized linear method for reconstruction of three‐dimensional microscopic objects from optical sections. Journal of the Optical Society of America. A, Optics, Image Science, and Vision 9 (2): 219–228.

Preza C, Miller MI and Conchello J‐A (1993) Image reconstruction for 3d light microscopy with a regularized linear method incorporating a smoothness prior. IS&T/SPIE's Symposium on Electronic Imaging: Science and Technology, pp. 129–139. International Society for Optics and Photonics.

Preza C and Conchello J‐A (2004) Depth‐variant maximum‐likelihood restoration for three‐dimensional fluorescence microscopy. Journal of the Optical Society of America. A, Optics, Image Science, and Vision 21 (9): 1593–1601.

Rieder CL and Khodjakov A (2003) Mitosis through the microscope: advances in seeing inside live dividing cells. Science 300 (5616): 91–96.

Swedlow JR (2001) Deconvolution Fluorescence Light Microscopy. John Wiley and Sons, Ltd.

van Kempen GM, van der Voort HT, Bauman JG and Strasters KC (1996) Comparing maximum likelihood estimation and constrained Tikhonov‐Miller restoration. IEEE Engineering in Medicine and Biology Magazine 15 (1): 76–83.

Verveer PJ and Jovin TM (1998) Image restoration based on good's roughness penalty with application to fluorescence microscopy. Journal of the Optical Society of America. A, Optics, Image Science, and Vision 15 (5): 1077–1083.

Vonesch C and Unser M (2008) A fast thresholded landweber algorithm for wavelet‐regularized multidimensional deconvolution. IEEE Transactions on Image Processing 17 (4): 539–549.

Vonesch C and Unser M (2009) A fast multilevel algorithm for wavelet‐regularized image restoration. IEEE Transactions on Image Processing 18 (3): 509–523.

Further Reading

Andrews PD, Harper IS and Swedlow JR (2002) To 5d and beyond: quantitative fluorescence microscopy in the postgenomic era. Traffic 3 (1): 29–36.

McNally JG, Preza C, Conchello J‐A and Thomas LJ (1994) Artifacts in computational optical‐sectioning microscopy. Journal of the Optical Society of America. A, Optics, Image Science, and Vision 11 (3): 1056–1067.

Preza C and Myneni V (2010) Quantitative depth‐variant imaging for fluorescence microscopy using the cosmos software package. BiOS, p. 757003. International Society for Optics and Photonics.

Sarder P and Nehorai A (2006) Deconvolution methods for 3‐d fluorescence microscopy images. IEEE Signal Processing Magazine 23 (3): 32–45.

Wallace W, Schaefer LH and Swedlow JR (2001) A workingperson's guide to deconvolution in light microscopy. Biotechniques 31 (5): 1076–1097.

Swedlow JR (2001) Deconvolution Fluorescence Light Microscopy. John Wiley and Sons, Ltd.

Contact Editor close
Submit a note to the editor about this article by filling in the form below.

* Required Field

How to Cite close
Arigovindan, Muthuvel(Oct 2015) Deconvolution Methods for Low Exposure Fluorescence Microscopy. In: eLS. John Wiley & Sons Ltd, Chichester. http://www.els.net [doi: 10.1002/9780470015902.a0020895]