JKOS Journal of the Korean Ophthalmological Society J Korean Ophthalmol Soc 0378-6471 2092-9374 Korean Ophthalmological Society 10.3341/jkos.2014.55.9.1320 jkos-55-1320 Original Article Influence of Image Compression on the Interpretation of Optical Coherence Tomography in Diabetic Macular Edema KangSunahMD ChangYoung SukMD KimJong WooMD KimChul GuMD LeeTae GonMD KimJae HuiMD Department of Ophthalmology, Kim's Eye Hospital, Konyang University College of Medicine, Seoul, Korea Address reprint requests to Jae Hui Kim, MD, Department of Ophthalmology, Konyang University Kim's Eye Hospital, #136 Yeongsin-ro, Yeongdeungpo-gu, Seoul 150-034, Korea, Tel: 82-2-2671-7665, Fax: 82-2-2671-6359, E-mail: kjh7997@daum.net 09 2014 02 09 2014 55 9 1320 1326 08032014 24042014 01082014 Copyright © 2014 The Korean Ophthalmological Society 2014 This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract Purpose

To evaluate the influence of image compression on optical coherence tomography (OCT) images in eyes with diabetic macular edema (DME).

 Methods

Twenty eyes of 30 patients diagnosed with DME were included in this retrospective observational case series. Horizontal OCT scans centered at the center of the fovea were conducted using spectral-domain OCT (Spectral OCT/SLO®). The images were exported to Tag Image File Format (TIFF) and then transformed to 10, 5, and 1 quality of Joint Photographic Experts Group (JPEG) format using Photoshop. OCT images were taken before and after intravitreal bevacizumab injection. The presence of intraretinal fluid, foveolar detachment, and photoreceptor inner segment/outer segment (IS/OS) disruption were evaluated in each image.

 Results

The mean (+/- standard deviation) size of TIFF images and 10, 5 and 1 quality JPEG images were 1712.0, 183.3 ± 6.8, 90.9 ± 4.3, 42.8 ± 1.4 kilobytes (KB), respectively, before the injection and 1712.0, 189.5 ± 9.1, 94.9 ± 5.6, 43.4 ± 1.8 KB, respectively, after the injection. The presence of intraretinal fluid, foveolar detachment, and photoreceptor IS/OS disruption identified in TIFF images was also identified in the compressed JPEG images.

 Conclusions

Quality of retinal OCT image did not influence the estimation of DME despite the JPEG image being compressed to approximately 1/40 of the original TIFF image size.

 Diabetic macular edema Image compression Optical coherence tomography References AlamSZawadzkiRJChoiSClinical application of rapid serial fourier-domain optical coherence tomography for macular imagingOphthalmology2006113142531 NassifNCenseBParkBHIn vivo human retinal imaging by ultrahigh-speed spectral domain optical coherence tomographyOpt Lett2004294802 LeeSHChungHKimHCSubfoveal choroidal thickness in fellow eyes of patients with central serous chorioretinopathyJ Korean Ophthalmol Soc2012539827 KimJHKangSWParkDYAsymmetric elongation of foveal tissue after macular hole surgery and its impact on metamorphopsiaOphthalmology2012119213340 BrownDMChenEMarianiAMajorJCJrSuper-dose anti-VEGF (SAVE) trial: 2.0 mg intravitreal ranibizumab for recalcitrant neovascular macular degeneration-primary end pointOphthalmology201312034954 CohenSYDuboisLNghiem-BuffetSSpectral domain optical coherence tomography analysis of macular changes in tilted disk syndromeRetina201333133845 YeungLLimaVCGarciaPCorrelation between spectral domain optical coherence tomography findings and fluorescein angiography patterns in diabetic macular edemaOphthalmology2009116115867 BolzMKriechbaumKSimaderCIn vivo retinal morphology after grid laser treatment in diabetic macular edemaOphthalmology201011753844 HaritoglouCKookDNeubauerAIntravitreal bevacizumab (Avastin) therapy for persistent diffuse diabetic macular edemaRetina2006269991005 PaccolaLCostaRAFolgosaMSIntravitreal triamcinolone versus bevacizumab for treatment of refractory diabetic macular oedema (IBEME study)Br J Ophthalmol2008927680 YanyaliAAytugBHorozogluFNohutcuAFBevacizumab (Avastin) for diabetic macular edema in previously vitrectomized eyesAm J Ophthalmol20071441246 JeongJHKimESLeeJKThe effects of intravitreal bevacizumab injection according to the type of diabetic macular edemaJ Korean Ophthalmol Soc2010517006 LeeSJKimETMoonYSIntravitreal bevacizumab alone versus combined with macular photocoagulation in diabetic macular edemaKorean J Ophthalmol201125299304 MaheshwaryASOsterSFYusonRMThe association between percent disruption of the photoreceptor inner segment-outer segment junction and visual acuity in diabetic macular edemaAm J Ophthalmol2010150637.e1 OtaniTYamaguchiYKishiSCorrelation between visual acuity and foveal microstructural changes in diabetic macular edemaRetina20103077480 ShinHJLeeSHChungHKimHCAssociation between photoreceptor integrity and visual outcome in diabetic macular edemaGraefes Arch Clin Exp Ophthalmol20122506170 SakamotoANishijimaKKitaMAssociation between foveal photoreceptor status and visual acuity after resolution of diabetic macular edema by pars plana vitrectomyGraefes Arch Clin Exp Ophthalmol2009247132530 LimJHKimIHBaeGHAnalysis of optical coherence tomographic patterns and clinical courses in diabetic macular edema after treatmentJ Korean Ophthalmol Soc2014552229 KimJHKangSWHaHSOverestimation of subfoveal choroidal thickness by measurement based on horizontally compressed optical coherence tomography imagesGraefes Arch Clin Exp Ophthalmol201325110916 Figures and Table

Comparison of image quality between an image of Tag Image File Format (TIFF) format (A, A-1, A-2) and images of 1 quality of Joint Photographic Experts Group (JPEG) compression format (B, B-1, B-2) in an eye with diabetic macular edema before treatment. A-1, B-1: ×600% magnified view of areas enclosed by white square (A, B), A-2, B-1: ×3,200% magnified view of areas enclosed by white square (A-1, B-1). Magnified views show the distorted image. The image was composed of small squares (dotted square, A-2) in the TIFF image, whereas the basic unit composing the JPEG image was relative large square (dotted square, B-1, B-2) which composed of 8×8 small squares. However, the presence of subfoveal serous detachment and small intraretinal cysts were clearly identified even in compressed image.

Comparison of image quality between an image of Tag Image File Format (TIFF) format (A, A-1, A-2) and images of 1 quality of Joint Photographic Experts Group (JPEG) compression format (B, B-1, B-2) in an eye with diabetic macular edema after treatment. A-1, B-1: ×600% magnified view of areas enclosed by white square (A, B), A-2, B-2: ×3,200% magnified view of areas indicated by white arrows (A-1, B-1). Magnified view shows the distorted pixels in compressed JPEG image (B-1, B-2). The presence of very small intraretinal cysts white arrows was clearly identified even in compressed image (B-1).

Size of image, prevalence of intraretinal fluid, foveolar detachment, and foveal photoreceptor inner segment/outer segment (IS/OS) disruption, according to the file form at and degree of compression

Findings File format and degree of image compression
TIFF (uncompressed) JPEG (10 quality) JPEG (5 quality) JPEG (1 quality)
Before treatment (n = 30)
Size of image (KB) 1712 183.3 ± 6.8 90.9 ± 4.3 42.8 ± 1.4
Intraretinal fluid (+/-) (n) 30/0 30/0 30/0 30/0
Foveolar detachment (+/-) (n) 10/20 10/20 10/20 10/20
IS/OS disruption (+/-) (n) 3/27
 3/27
 3/27
 3/27
After intravitreal bevacizumab injections (n = 30)
Size of image (KB) 1712 189.5 ± 9.1 94.9 ± 5.6 43.4 ± 1.8
Intraretinal fluid (+/-) (n) 23/7 23/7 23/7 23/7
Foveolar detachment (+/-) (n) 5/25 5/25 5/25 5/25
IS/OS disruption (+/-) (n) 3/27 3/27 3/27 3/27

Values are presented as mean ± SD unless otherwise indicated.

TIFF = tag image file format; JPEG = joint photographic experts group; KB = kilobytes.