|
|
Automatic Analysis Method of Macular Edema Classification for Diabetic Retinopathy Images#br# |
College of Physics and Information Engineering, Fuzhou University, Fuzhou 350108, China |
|
|
Abstract Diabetic macular edema (DME) can appear in any stage of the diabetic retinopathy (DR) and is the leading cause of vision impairment in people with diabetes. Therefore, the automatic analysis of DME plays a key role in the DR screening program. According to the DME international clinical classification standard, detecting and judging the presence of hard exudates (HEs) close or relate to macula fovea (MF) is a standard method to assess DME in the fundus images. A synthetic improvement method based on existing mathematical morphology technique is selected for HEs detection. A novel method of macular fovea center location is proposed based on the directional local contrast filter and the local vessel density. Then the optic disk region can be easily located and removed for reducing the impact to the HEs detection. Only the large vessel segmentations were used for the vessel density calculation. From the testing results on the 169 images from HEIMED public dataset, the sensitivity and specificity of HEs detection on image level are 100% and 922%, respectively. The accuracy of macular fovea center location reaches 982%. And the accuracy for each DME classification is more than 88%. The proposed method would have important clinical application potentials.
|
|
|
|
|
[1]Deepak KS, Sivaswamy J. Automatic assessment of macular edema from color retinal images [J]. IEEE Transactions on Medical Imaging, 2012, 31(3): 766-776.
[2]Abràmoff MD, Garvin MK, Sonka M. Retinal imaging and image analysis[J]. IEEE Reviews in Biomedical Engineering, 2010, 3: 169-208.
[3]惠延年, 王琳. 糖尿病视网膜病变和黄斑水肿的国际临床分类法[J]. 国际眼科杂志, 2004, 4(1): 56-59.
[4]Tariq A, Akram M U, Shaukat A, et al. Automated detection and grading of diabetic maculopathy in digital retinal images[J]. Journal of digital imaging, 2013, 26(4): 803-812.[5]Kayal D, Banerjee S. A new dynamic thresholding based technique for detection of hard exudates in digital retinal fundus image[C] //2014 International Conference on Signal Processing and Integrated Networks (SPIN). Noida: IEEE, 2014:141-144.
[6]Sopharak A, Uyyanonvara B, Barman S, et al. Automatic detection of diabetic retinopathy exudates from nondilated retinal images using mathematical morphology methods[J]. Computerized Medical Imaging and Graphics, 2008, 32(8): 720-727.
[7]Sinthanayothin C, Boyce JF, Williamson TH, et al. Automated detection of diabetic retinopathy on digital fundus images [J]. Diabetic Medicine, 2002, 19(2): 105-112.
[8]Sánchez CI, García M, Mayo A, et al. Retinal image analysis based on mixture models to detect hard exudates [J]. Medical Image Analysis, 2009, 13(4): 650-658.
[9]Ranamuka NG, Meegama RGN. Detection of hard exudates from diabetic retinopathy images using fuzzy logic [J]. IET Image Processing, 2012,7(2):121-130.
[10]García M, Snchez CI, López M I, et al. Neural network based detection of hard exudates in retinal images [J]. Computer Methods and Programs in Biomedicine, 2009, 93(1): 9-19.
[11]Harangi B, Antal B, Hajdu A. Automatic exudate detection with improved Navebayes classifier[C] //25th International Symposium on ComputerBased Medical Systems (CBMS). Rome: IEEE, 2012: 1-4.
[12]高玮玮, 沈建新, 王玉亮, 等. 改进的快速 FCM及SVM实现糖网白色病灶的自动检测[J]. 中国生物医学工程学报, 2013, 32(3): 305-312.
[13]Li H, Chutatape O. Automated feature extraction in color retinal images by a model based approach [J]. IEEE Transactions on Biomedical Engineering, 2004, 51(2): 246-254.[14]Niemeijer M, Abràmoff MD, Van Ginneken B. Segmentation of the optic disc, macula and vascular arch in fundus photographs [J]. IEEE Transactions on Medical Imaging, 2007, 26(1): 116-127.
[15]Niemeijer M, Abràmoff MD, Van Ginneken B. Fast detection of the optic disc and fovea in color fundus photographs [J]. Medical Image Analysis, 2009, 13(6): 859-870.
[16]Tobin KW, Chaum E, Govindasamy VP, et al. Detection of anatomic structures in human retinal imagery [J]. IEEE Transactions on Medical Imaging, 2007, 26(12): 1729-1739.
[17]Hamilton Eye Institute Macular Edema Dataset [EB/OL]. http://vibot.ubourgogne.fr/luca/heimed.php, 2012-06/2014-06-18.
[18]Giancardo L, Meriaudeau F, Karnowski TP, et al. Exudatebased diabetic macular edema detection in fundus images using publicly available datasets [J]. Medical Image Analysis, 2012, 16(1): 216-226.
[19]Zhang Ming, Liu JC. Directional local contrast based blood vessel detection in retinal images [C] //IEEE International Conference on Image Processing. San Antonio: IEEE, 2007, 4: IV-317-IV-320.
[20]Stark J A. Adaptive image contrast enhancement using generalizations of histogram equalization [J]. IEEE Transactions on Image Processing, 2000, 9(5): 889-896.
[21]Patel VM, Easley GR, Healy DM. A new multiresolution generalized directional filter bank design and application in image enhancement[C] // Proceedings of 15th IEEE International Conference on Image Processing. San Diego: IEEE, 2008: 2816-2819.
[22]Sanchez CI,Hornero R,Lopez MI,et al.A novel automated image processing algorithm for detection of hard exudates based on retinal images analysis[J].Med Eng Phys,2007,30(3):350-357.
[23]高玮玮, 沈建新, 王玉亮. 眼底图像中硬性渗出自动检测方法的对比[J]. 南京航空航天大学学报, 2013, 45(1): 55-61. |
|
|
|