Improving visualization and quantitative assessment of choriocapillaris with swept source OCTA through registration and averaging applicable to clinical systems

Abstract

Choriocapillaris (CC) visualization and quantification remains challenging. We propose an innovative three-step registration and averaging approach using repeated swept source optical coherence tomography angiography (SS-OCTA) scans to conduct automatic quantitative assessment on CC. Six subjects were enrolled, each imaged at several locations with SS-OCTA from macular to equatorial regions using 3 mm × 3 mm scanning pattern. Five repeated volumes were collected for each subject. The complex optical microangiography (OMAG) algorithm was applied to identify blood flow in CC slab. An automatic three-step registration of translation, affine and B-Spline was applied to en face OCTA images of CC, followed with averaging. A fuzzy clustering approach was used to segment vasculature and flow deficits from the averaged images. The improvement in visualization of CC was evaluated and the average intercapillary distance was estimated by calculating the averaged capillary lumen spacing. A series of quantitative indices of flow deficit density, number, size, complexity index and aspect ratio index (FDD, FDN, FDS, FDCI and FDARI) were designed and validated with the increase of repeated scan numbers for averaging. Quantitative assessment was applied and compared on CC in macular and equatorial regions. The intercapillary distance was observed to be around 24 µm at macula and increased toward equatorial regions. All five quantitative indices (FDD, FDN, FDS, FDCI and FDARI) showed significant changes with multiple averaging and tend to become stable with repeated number of 4. Our proposed registration and averaging algorithm significantly improved the visualization of CC with SS-OCTA. The designed five indices for CC provide more options in the quantitative assessment of CC and are of great potentials in assisting the understanding of disease pathology, early diagnosis and treatment monitoring.

Figure 1

Illustration of the registration and averaging method to improve the image quality of the choriocapillaris. A series of repeated projection images were successively performed translation, affine and B-spline registration with the reference to the first scan, to eliminate the deformation and motion between images.

Figure 2

Illustration of the CC FDs segmentation using fuzzy c-means on a chronic birdshot chorioretinopathy patient. (A) Averaged OCTA CC image; (B) elbow method to determine appropriate number of clusters; membership is designated to each cluster of signals; (C) fuzzy c-means membership map of original CC image, all pixels were assigned into 5 different memberships based on fuzzy logic; (D) binary CC FD map, pixels presented in black represent identified CC FDs.

Figure 3

Visual illustration of quantification maps derived from CC FD map. (A) Binary CC FD map, white pixels represent identified FDs; (B) CC FD perimeter map, white pixels represents identified perimeter of FDs; (C) CC FD length map, white pixels represents identified center length of FDs; (D) CC FD aspect ratio map, color bar represents calculated aspect ratio of FDs, unit-less; (E) CC FD complexity map, color represents calculated complexity index of FDs, unit-less; (F) CC FD size map, color bar represents calculated size of FDs, unit is µm².

Figure 4

Improved visualization of macular CC and equatorial CC can be achieved by registration and averaging algorithm. Top row (A) macular CC of single scan, 2 scans averaged, 3 scans averaged, 4 scans averaged, and 5 scans averaged, respectively. Bottom row (B) equatorial CC of single scan, 2 scans averaged, 3 scans averaged, 4 scans averaged and 5 scans averaged, respectively.

Figure 5

Performance evaluation of registration and averaging algorithm against the number of averaged scans. (A) Global entropy (0–1); (B) global standard deviation (y-axis ranging from 0–255); (C) local texture correlation (y-axis ranging from 0–1); (D) PSNR using five averages as reference.

Figure 6

Power spectrum analysis of the inter-capillary distance for macular CC, posterior pole CC and equatorial CC, respectively. (A) Three selected 650 µm *650 µm regions of macular CC; (B) 2D power spectrum of macular CC in A; (C) example radially averaged power spectrum plot of macular CC. (D) Three selected 650 µm ∗ 650 µm regions of posterior pole CC (~4 mm away from fovea); (E) 2D power spectrum of macular CC in D; (F) example radially averaged power spectrum plot of posterior pole CC. (G) Three selected 650 µm ∗ 650 µm regions of equatorial CC; (H) 2D power spectrum of equatorial CC in G; (I) example radially averaged power spectrum plot of equatorial CC. Scale bar represent 200 µm.

Figure 7

The inter-capillary distance increases as the increase of radial distance with respect to central fovea (depicted as degrees). X-axis indicates the estimated relative locations.

Figure 8

Visual illustration of CC FDs detection with averaging for the scans obtained from macular region. (Panel A) Macular CC OCTA images with single scan, three scans averaged, and five scans averaged, respectively. (Panel B) Macular CC vasculature segmented by fuzzy c-means algorithm with single scan, three scans averaged, and five scans averaged, respectively. (Panel C) Segmented macular CC FD with single scan, three scans averaged and five scans averaged, respectively.

Figure 9

Visual illustration of CC FDs detection with averaging for the scans obtained from equatorial region. (Panel A) Macular CC OCTA images with single scan, three scans averaged, and five scans averaged, respectively. (Panel B) Macular CC vasculature segmented by fuzzy c-means algorithm with single scan, three scans averaged, and five scans averaged, respectively. (Panel C) Segmented macular CC FD with single scan, three scans averaged, and five scans averaged, respectively.

Figure 10

Quantitative analysis of CC FDs with multiply averaged OCTA images. X-axis represents number of images averaged. Y-axis in (A) Flow deficits density; (B) Flow deficits number; (C) Flow deficits complexity index; and (D) Flow deficits size. ^*Denotes p < 0.05, ^**Denotes p < 0.01 and ^***Denotes p < 0.001.

Figure 11

Correlation and Bland-Altman plots of CC FDD with multiple averaged scans. (A) Comparing single scan CC FDD results with 5 scans averaged CC FDD results; (B) comparing 2 scans averaged CC FDD results with 5 scans averaged CC FDD results; (C) comparing 3 scans averaged CC FDD results with 5 scans averaged CC FDD results; (D) comparing 4 scans averaged CC FDD results with 5 scans averaged CC FDD results.