Kybernetika 47 no. 3, 455-481, 2011

Fusion based analysis of ophthalmologic image data

Jiří Jan, Radim Kolář, Libor Kubečka, Jan Odstrčilík and Jiří Gazárek

Abstract:

The paper presents an overview of image analysis activities of the Brno DAR group in the medical application area of retinal imaging. Particularly, illumination correction and SNR enhancement by registered averaging as preprocessing steps are briefly described; further mono- and multimodal registration methods developed for specific types of ophthalmological images, and methods for segmentation of optical disc, retinal vessel tree and autofluorescence areas are presented. Finally, the designed methods for neural fibre layer detection and evaluation on retinal images, utilising different combined texture analysis approaches and several types of classifiers, are shown. The results in all the areas are shortly commented on at the respective sections. In order to emphasise methodological aspects, the methods and results are ordered according to consequential phases of processing rather then divided according to individual medical applications.

Keywords:

image analysis, image fusion, 2D and 3D image registration, ophthalmology, retina imaging, subtractive angiography, computed tomography, illumination correction, image averaging, spatial transforms

Classification:

93E12, 62A10

paper.pdf

References:

  1. R. Bock et al.: Glaucoma risk index: automated glaucoma detection from color fundus images. Medical Image Analysis 14 (2000), 3, 471-481.   CrossRef
  2. B. H. Brinkmann, A. Manduca and R. A. Robb: Optimised homomorphic unsharp masking for MR greyscale inhomogeneity correction. IEEE Trans. Med. Imag. 17 2, 161-171.   CrossRef
  3. D. L. Budenz: Reproducibility of retinal nerve fiber thickness measurements using the stratus OCT in normal and glaucomatous eyes. Invest. Ophthalmology and Visual Science 46 (2005), 2440-2443.   CrossRef
  4. C Bellmann: Topography of fundus autofluorescence with a new confocal scanning laser ophthalmoscope. Ophthalmology 94 (1997), 385-91.   CrossRef
  5. T. Chanwimaluang and G. Fan: An efficient blood vessel detection algorithm for retinal images using local entropy tresholding. Proc. Int. Symp. Circuits \   CrossRef
  6. R. Chrástek, H. Niemann, L. Kubečka, J. Jan, V. Derhartunian and G. Michelson: Optic nerve head segmentation in multimodal retinal images. In: Proc. SPIE 2005, Bellingham 2005, pp. 1604-1615.   CrossRef
  7. R. Chrástek et al.: Segmentation of the optic nerve head for glaucoma diagnosis. In: Proc. SPIE 2005, Bellingham 2005, pp. 1604-1615.   CrossRef
  8. T. A. Ciulla, C. D. Regillo and A. H. Harris: Retina and Optic Nerve Imaging. Lippincott Williams and Wilkins, Philadelphia 2003.   CrossRef
  9. M. J. Cree, D. Cornforth and H. F. Jelinek: Vessel segmentation and tracking using a two-dimensional model. IVC New Zealand (2005), 345-350.   CrossRef
  10. B. M. Dawant, P. Zijdenbos and R. A. Margolin: Correction of intensity variations in MR images for computer-aided tissue classification. IEEE Trans. Med. Imag. 12 (1993), 4, 770-781.   CrossRef
  11. F. C. Delori et al.: In vivo fluorescence of the ocular fundus exhibits retinal pigment epithelium lipofuscin characteristics. Invest. Ophthalmol. Vis. Sci. 12 (1995), 718-29.   CrossRef
  12. P. M. Dodson: Diabetic Retinopathy. Oxford University Press 2008.   CrossRef
  13. M. A. T. Figueiredo and A. K. Jain: Unsupervised learning of finite mixture models. IEEE Trans. Pattern Analysis and Machine Intelligence 24 ({2002}), 3, 381-396.   CrossRef
  14. M. J. Greaney et al.: Comparison of optic nerve imaging methods to distinguish normal eyes from those with glaucoma. Invest Ophthalmol Vis. Sci. 43 (2002), 1, 140-145.   CrossRef
  15. E. Grisan et al.: A new tracking system for robust extraction of retinal vessel structure. In: Proc. 26th IEEE EMBC 2004, San Francisco 2004, pp. 1620-1623.   CrossRef
  16. J. Gazárek, R. Kolář, J. Jan and J. Odstrčil\'{i}k: Blood vessel tree recontruction in retinal OCT data. In: Proc. EURASIP Conf. BIOSIGNAL 2010, Brno 2010, CD issue, 4 pp.   CrossRef
  17. R. Guillemaud and M. Brady: Estimating the bias field of MR images. IEEE Trans. Med. Imag. 16 (1997), 3, 238-251.   CrossRef
  18. Y. Hayashi et al.: Detection of retinal nerve fiber layer defects in retinal fundus images using Gabor filtering. In: Proc. of SPIE 6514 (2006).   CrossRef
  19. St. Hoh: Evaluating the optic nerve head and retinal nerve fibre layer: The role of Heidelberg retina tomography, scanning laser polarimetry and optical coherence tomography. Annals Academy of Medicine 16 (2007), 195-202.   CrossRef
  20. J. Jan, J. Odstrčil\'{i}k, J. Gazárek and R. Kolář.: Retinal image analysis aimed at early detection of neural-layer deterioration. Submitted.   CrossRef
  21. J. Jan, R. Chrástek and L. Kubečka: Automated optic disc segmentation in multimodal images of retina. In: Proc. DOG/SOE Congress 2005, Berlin 2005, CD issue.   CrossRef
  22. J. Jan and R. Kolář et al.: Analysis of fused ophthalmologic image data. In: Proc. 6th EURASIP conf. Speech \   CrossRef
  23. J. Jan: Retinal image analysis - Brno group). In: SAOT Retina Image Processing Workshop 2009, Erlangen Univ.   CrossRef
  24. J. Jan: Retinal image analysis aimed at blood vessel structure segmentation and neural layer detection. In: Proc. BEC 2008, Tallin 2008, pp. 31-38   CrossRef
  25. J. Jan: Medical Image Processing, Reconstruction and Restoration - Concepts and Methods. CRC Press, Taylor and Francis Group 2006.   CrossRef
  26. J. Jan, J. Odstrčil\'{i}k, J. Gazárek and R. Kolář: Retinal image analysis aimed at support of early neural-layer deterioration diagnosis. In: Proc. ITAB 2009, Larnaca, 4 pp., CD issue.   CrossRef
  27. P. Janknecht and J. Funk: Optic nerve head analyser and Heidelberg retina tomograph: accuracy and reproducibility of topographic measurements in a model eye and in volunteers. British Journal of Ophthalmology 78 (1994), 760-768.   CrossRef
  28. J. Jorge, G. Leandro and M. Roberto et al.: Vessels segmentation in retina: Preliminary assessment of the mathematical morphology and of the wavelet transform techniques. In: XIV Brazilian SIBGRAPI'01 2001, pp. 84-91.   CrossRef
  29. R. Kolář, V. Šikula and M. Base: Retinal image registration using phase correlation. In: Proc. 20th EURASIP Conf. BIOSIGNAL 2010, Brno 2010, CD issue, 4 pp.   CrossRef
  30. R. Kolář, J. Jan, R. Chrástek, R. Laemmer and Ch. Y. Mardin: Autofluorescence areas detection in HRA images. In: Proc. EMBEC'05, Prague 2005, CD issue.   CrossRef
  31. R. Kolář, L. Kubečka, J. Jan and R. Chrastek: Disparity estimation in uncalibrated stereo retina images. In: Proc. EMBEC'05, Prague 2005, CD issue.   CrossRef
  32. R. Kolář and J. Jan: Detection of glaucomatous eye via color fundus images using fractal dimensions. In: Proc. 6th EURASIP Conf. Speech \   CrossRef
  33. R. Kolář, J. Jan and L. Kubečka: Registration and fusion of the autofluorescent and infrared retinal images. Internat. J. Biomedical Imaging (2008), 513478, pp. 1-11.   CrossRef
  34. R. Kolář, J. Jan and L. Kubečka: Computer support for early glaucoma diagnosis based on the fused retinal images. Scripta Medica (2006), 79, 269-276.   CrossRef
  35. R. Kolář, J. Jan and R. Jiřík: Semiautomatic detection and evaluation of autofluorescent areas. In: Proc. IEEE-EMBC 2007, Lyon 2007, pp. 3327-3330.   CrossRef
  36. R. Kolář, R. Laemmer, J. Jan and C. Mardin: The segmentation of zones with increased autofluorescence in the junctional zone of parapapillary atrophy. Physiological Measurement (2009), 30, 505-516.   CrossRef
  37. L. Kubečka and J. Jan: Retinal image fusion and registration. In: Proc. EMBEC'05, Prague 2005, CD issue.   CrossRef
  38. L. Kubečka, M. Skokan and J. Jan: Optimization methods for registration of multimodal images of retina. In: Proc. IEEE-EMBC, Cancun 2003, pp. 599-601.   CrossRef
  39. L. Kubečka and J. Jan: Registration of bimodal retinal images - improving modifications. In: Proc. 26th IEEE EMBC, San Francisco 2004, pp. 1695-1698.   CrossRef
  40. L. Kubečka, J. Jan and R. Kolář: Retrospective illumination correction of retinal images. J. Biomedical Imaging (2010), 5, 201-210.   CrossRef
  41. L. Kubečka, J. Jan, R. Kolář and R. Jiřík: Improving quality of autofluorescence images using non-rigid image registration. In: Proc. EUSIPCO 2006, Florence 2006, CD issue, pp. 357-361.   CrossRef
  42. L. Kubečka, J. Jan, R. Kolář and R. Jiřík: Elastic registration for auto-fluorescence image averaging. In: Proc. IEEE-EMBC 2006, New York 2006, CD issue, pp. 1948-1951.   CrossRef
  43. R. Laemmer et al.: Measurement of autofluorescence in the parapapillary atrophic zone in patients with ocular hypertension. Graefes Arch. Clin. Exp. Ophthalmol. (2007), 245, 51-58.   CrossRef
  44. M. Lalondey, L. Gagnony and M. C. Boucherz: Non-recursive paired tracking for vessel extraction from retinal images. In: Proc. Vision Interface 2000, Montreal 2000, pp. 61-68.   CrossRef
  45. S. Y. Lee et al.: Automated quantification of retinal nerve fiber layer atrophy in fundus photograph. In: Proc. IEEE EMBC San Francisco 2004, 1, pp. 1241-1243.   CrossRef
  46. S. Z. Li et al.: Markov Radnom Field Modeling in Image Analysis. Springer 2009.   CrossRef
  47. R. Linde et al.: Reproducibility of parapapillary autofluorescence measurement in glaucoma diagnostics. In: Proc. DOG 2005, p. 482.   CrossRef
  48. B. Likar, J. Derganc and F. Pernus: Segmentation-based retrospective correction of intensity non-uniformity in multispectral MR images. In: Proc. Conf. Medical Imaging: Image Processing, San Diego (M. Sonka, J. M. Fitzpatrick, eds.), Proc. SPIE 4684 (2002), pp. 1531-1540.   CrossRef
  49. B. Likar, J. B. Maintz, M. Viergever and F. Pernus: Retrospective shading correction based on entropy minimization. J. Microscopy 197 (2000), 3, 285-295.   CrossRef
  50. N. Lois, A. S. Halfyard, A. C. Bird and F. W. Fitzke: Quantitative evaluation of fundus autofluorescence imaged in vivo in eyes with retinal disease. Br. J. Ophthalmol. 84 (2000), 741-5.   CrossRef
  51. M. Lundström and O. J. Eklundh: Computer densitometry of retinal nerve fibre atrophy - a pilot study. Acta Ophthalmologica 58 (1980), 4, 639-644.   CrossRef
  52. F. Maes: Segmentation and Registration of Multimodal Medical Images. PhD. Thesis, Kath. Univ. Leuven 1998.   CrossRef
  53. J.-F. Mangin: Entropy minimization for automatic correction of intensity nonuniformity. In: IEEE Works. MMBIA, Hilton Head Island 2000, 162-169.   CrossRef
  54. G. Michelson and M. J. Groh: Screening models for glaucoma. Curr Opin Ophthalmol. 12 (2001), 2, 105-11.   CrossRef
  55. Ch. Muramatsu, Y. Hayashi and A. Sawada et al.: Detection of retinal nerve fiber layer defects on retinal fundus images for early diagnosis of glaucoma. J. Biomedical Optics 15 (2010), 1, 1-7.   CrossRef
  56. R. Nayak et al.: Automated diagnosis of glaucoma using digital fundus images. J. Med. Syst. 33 (2009), 337-346.   CrossRef
  57. H. Niemann et al.: Towards automated diagnostic evaluation of retina images. J. Pattern Recognition and Image Analysis 16 (2006), 4, 671-676.   CrossRef
  58. M. Niemeijer, J. Staal and B. Ginneken et al.: Comparative study of retinal vessel segmentation methods on a new publicly available database. In: Proc. SPIE Med. Imag., San Diego 5370 (2004), p. 648.   CrossRef
  59. J. Staal et al.: Ridge-based vessel segmentation in color images of the retina. IEEE Trans. on Medical Imaging 23 (2004), 4, 501-509.   CrossRef
  60. J. Odstrčilík, J. Jan, J. Gazárek and R. Kolář: Improvement of vessel segmentation by matched filtering in colour retinal images. In: Proc. World Congress on Med. Physics Biomed. Engrg., Munich 2009, p. 4.   CrossRef
  61. J. Odstrčilík, R. Kolář, V. Harabis, J. Gazárek and J. Jan: Retinal nerve fiber layer analysis via Markov random fields texture modelling. In: Proc. EUSIPCO 2010, Eurasip, Aalborg 2010.   CrossRef
  62. P. Perona and J. Malik: Scale-space and edge detection using anisotropic diffusion. IEEE Trans. Pattern Analysis Machine Intelligence 12 (1990), 629-639.   CrossRef
  63. R. Porter and N. Canagarajah: Robust rotation-invariant texture classification: wavelet, Gabor filter and GMRF based schemes. IEEE Proc. Vis.-Image Signal Processing 144 (1997), 3, 180-188.   CrossRef
  64. P. Scheunders: An orthogonal wavelet representation of multivalued images. IEEE Trans. Image Processing 12 (2003), 6, 718-725.   CrossRef
  65. M. Skokan, A. Skoupý and J. Jan: Registration of multimodal images of retina. In: Proc. 24th Conf. IEEE EMBC, Houston 2002, pp. 1094-1096.   CrossRef
  66. M. Styner, CH. Brechbuehler, G. Szekely and G. Gerig: Parametric estimate of intensity inhomogeneities applied to MRI. IEEE Trans. Med. Imag. 19 (2000), 3, 153-165.   CrossRef
  67. J. Tvrdík: Generalized controlled random search and competing heuristic. In: Proc. 10th Int. Conf. on Soft Computing MENDEL 2004, pp. 228-33.   CrossRef
  68. J. Tvrdík: Controlled random search algorithm with alternating heuristics. AUTOMA (2002), 1, 54-57.   CrossRef
  69. A. Viestenz, A. Langenbucher and C. Y. Mardin: Parapapillary autofluorescence as indicator for glaucoma. Klin. Monatsbl. Augenheilkd. 223 (2006), 315-20.   CrossRef
  70. K. A. Vermeer et al.: A model based method for retinal blood vessel detection. Computers in Biology and Medicine 34 (2004), 209-219.   CrossRef
  71. J. Zhu, B. Liu and S. C. Schwartz: General illumination correction and its application to face normalization. In: Proc. IEEE ICASSP'03 3 (2003), pp. III-133-6.   CrossRef
  72. B. Zitová and J. Flusser: Image registration methods: a survey image. Vis. Comput. 21 (2003), 977-1000.   CrossRef