안저 영상에서 헤이지안 알고리즘을 이용한 혈관 검출 = Detection of Retinal Vessels of Fundus Photograph Using Hessian Algorithm

다국어 초록 (Multilingual Abstract)

Fundus images are highly useful in evaluating patients' retinal conditions in diagnosing eye diseases. In particular, vessel regions are essential in diagnosing diabetes and hypertension. In this paper, we used top-hat filter to compensate for non-uniform background. Image contrast was enhanced by using contrast limited adaptive histogram equalization (CLAHE) method. Hessian matrix was next applied to detect vessel regions. Results indicate that our method is 1.3% more accurate than matched filter method. Our proposed method is expected to contribute to diagnosing eye diseases.

국문 초록 (Abstract)

망막 질환의 진단에서 안저영상은 환자의 망막 상태에 대한 객관적인 평가와 기록에 중요하다. 특히 혈관의 분석은 당뇨병, 고혈압 등의 진단과 경과 관찰에 매우 중요하다. 혈관 영역을 검출하기 위해 톱-햇(Top-hat) 필터를 사용하여 균일하지 않은 배경 영상을 보상하고, 대비 제한의 적응적 히스토그램 보정 (contrast limited adaptive histogram equalization) 방법을 적용하여 대비를 향상시켰다. 영상에 전처리를 한 후 헤이지안 행렬(hessian matrix)을 적용하여 혈관 성분을 검출한 결과 제안된 방법이 기존의 정합 필터(matched filter) 방법보다 약 1.3% 더 정확하였다. 결론으로 제안한 알고리즘은 안저 영상에서 혈관 영역을 검출하는데 있어서 기존 방법에 비해서 향상되었다.

참고문헌 (Reference)

1 M. Niemeijer and B. van Ginneken,

2 이병일, "의료영상처리 및 분석 방법" 4 (4): 51-69, 2000

3 강전권, "안저영상 해석을 위한 특징영역의 분할에 관한 연구" 16 (16): 121-128, 1995

4 Zuiderveld, K., "“Contrast limited adaptive histogram equalization,” Graphics gems IV" Academic Press Professional, Inc.

5 T. N. Pappas, "new method for estimation of coronary artery dimensions in angiograms" 36 : 1501-1512, 1988

6 Lim JI, "comparison of digital nonmydriatic fundus imaging with standard 35-millimeter slides for diabetic retinopathy" 107 (107): 866-870, 2000

7 Y. Sato, "Tissue classification based on 3D local intensity structures for volume rendering" 6 (6): 160-180, 2000

8 Sato, Y., "Three-dimensional multi-scale line filter for segmentation and visualization of curvilinear structures in medical images" 2 (2): 143-168, 1998

9 B. M. Ege, "The relationship between age and colour content in fundus images" 80 (80): 485-489, 2002

10 Yen KG, "Telephotoscreening to detect retinopathy of prematurity : preliminary study of the optimum time to employ digital fundus camera imaging to detect ROP" 6 (6): 64-70, 2002

11 M. Sofka, "Retinal vessel centerline extraction using multiscale matched filters, confidence and edge measures" 25 (25): 1531-1546, 2006

12 A. Hoover, "Locating the optic nerve in a retinal image using the fuzzy convergence of the blood vessels" 22 (22): 951-958, 2003

13 Hoover, A., "Locating blood vessels in retinal images by piecewise threshold probing of a matched filter response" 19 (19): 203-210, 1998

14 M. Sofka, "Erratum to retinal vessel centerline extraction using multiscale matched filters, confidence and edge measures" 26 (26): 133-133, 2007

15 Gonzalez RC, "Digital image processing using MatLab" ITC 2004

16 Otsu, N., "A threshold selection method from gray_level histograms" 9 (9): 62-66, 1979

17 M. A. T. Figuereido, "A nonsmoothing approach to the estimation of vessel contours in angiograms" 14 : 162-172, 1995

18 Y. A. Tolias, "A fuzzy vessel tracking algorithm for retinal images based on fuzzy clustering" 17 (17): 263-273, 1998

19 Y. Sato, "3d multi-scale line filter for segmentation and visualization of curvilinear structures in medical images" Springer Verlag 213-222, 1997