Running pattern of choroidal vessel in en face OCT images determined by machine learning–based quantitative method
To evaluate the new method to quantitate the running pattern of the vessels in Haller’s layer in en face optical coherence tomographic (OCT) images using the new algorithm.
A retrospective and cross-sectional study. The en face image of top 25% slab of Haller’s layer was analyzed. The vascular area in these images was calculated after binarization. Then, the vessels were thinned, and the total length of the vessels and the mean vessel diameter were calculated. Based on the angle of vessel running, “natural oblique vessel” was defined. The ratio of the natural oblique vessel to the whole vessels was defined as the “symmetry index”. To examine the reproducibility of the software, the images obtained on two different examination dates of the same subject (25 eyes of 25 healthy subjects) were analyzed. Also, to compare the symmetry index and subjective evaluations, 180 eyes and 180 healthy subjects were analyzed. The subjective evaluations classified the images into 3 groups, the Symmetrical, Semi-symmetrical, and Asymmetrical types. Symmetry index was compared in each group.
The inter-measurement correlation coefficient (ICC) of the vessel area, vessel length, and vessel diameter were 0.955, 0.934, and 0.954, respectively. The ICC of the symmetry index was 0.926. The symmetry index of the Symmetrical type was 60.4 ± 7.2%, that of the Semi-symmetry type was 56.2 ± 4.6%, and that of the Asymmetry type was 52.6 ± 5.2%.
The present algorithm can analyze vessels in Haller’s layer of the en face images of choroid in an objective manner with good repeatability.
KeywordsChoroid En face imaging Haller’s layer Quantitative method
The authors thank Professor Emeritus Duco Hamasaki of the Bascom Palmer Eye Institute of the University of Miami for providing critical discussions and suggestions to our study and revision of the final manuscript. We also thank Mrs. Kazumi Sakaguchi and Mr. Saiki Fujii of Softcube Inc. for technical assistance.
This study was financially supported by JSPS KAKENHI grant number 15H04996.
Compliance with ethical standards
The procedures used in this study were approved by the Institutional Review Board of Kagoshima University Hospital, and the procedures conformed to the tenets of the 1989 Declaration of Helsinki.
Informed consent was obtained from all individual participants included in the study.
Conflict of interest
The authors declare that they have no conflict of interest.
- 1.Lutty GA, Cao J, McLeod DS (1997) Relationship of polymorphonuclear leukocytes to capillary dropout in the human diabetic choroid. Am J Pathol 151:707Google Scholar
- 12.Esmaeelpour M, Považay B, Hermann B, Hofer B, Kajic V, Kapoor K, Sheen NJ, North RV, Drexler W (2010) Three-dimensional 1060-nm OCT: choroidal thickness maps in normal subjects and improved posterior segment visualization in cataract patients. Invest Ophthalmol Vis Sci 51:5260–5266CrossRefGoogle Scholar
- 14.Adhi M, Liu JJ, Qavi AH, Grulkowski I, Lu CD, Mohler KJ, Ferrara D, Kraus MF, Baumal CR, Witkin AJ (2014) Choroidal analysis in healthy eyes using swept-source optical coherence tomography compared to spectral domain optical coherence tomography. Am J Ophthalmol 157:1272–1281. e1271CrossRefGoogle Scholar
- 20.Stroustrup B (2013) The C++ programming language, 4th edn. Pearson Education India, LondonGoogle Scholar
- 21.Zuiderveld K (1994) Contrast limited adaptive histogram equalization. Graphics gems: 474–485Google Scholar
- 22.Buades A, Coll B, Morel J-M (2005) A non-local algorithm for image denoising. In Proc IEEE CVPR 60–65Google Scholar