Uncovering of intraspecies macular heterogeneity in cynomolgus monkeys using hybrid machine learning optical coherence tomography image segmentation

Abstract

The fovea is a depression in the center of the macula and is the site of the highest visual acuity. Optical coherence tomography (OCT) has contributed considerably in elucidating the pathologic changes in the fovea and is now being considered as an accompanying imaging method in drug development, such as antivascular endothelial growth factor and its safety profiling. Because animal numbers are limited in preclinical studies and automatized image evaluation tools have not yet been routinely employed, essential reference data describing the morphologic variations in macular thickness in laboratory cynomolgus monkeys are sparse to nonexistent. A hybrid machine learning algorithm was applied for automated OCT image processing and measurements of central retina thickness and surface area values. Morphological variations and the effects of sex and geographical origin were determined. Based on our findings, the fovea parameters are specific to the geographic origin. Despite morphological similarities among cynomolgus monkeys, considerable variations in the foveolar contour, even within the same species but from different geographic origins, were found. The results of the reference database show that not only the entire retinal thickness, but also the macular subfields, should be considered when designing preclinical studies and in the interpretation of foveal data.

Cynomolgus monkey’s choroid reference database derived from hybrid deep learning optical coherence tomography segmentation

“…Given the fact that OCT imaging is used rather extensively in animal models and that, in an increasing number of cases, morphological OCT assessment is highly comparable to histopathology, the use of OCT as a constantly evolving imaging technique has been included in the framework of drug safety profiling 14 , 15 . In this context, it is noteworthy that significant differences in retinal thickness were found between Mauritius and Asian Macaques, despite being the same species but with different origins 16 . Although several studies have evaluated retinal and choroidal blood supply in macaques, few measurements have been conducted on a large number of individuals while taking into account both their origin and sex, thus providing appropriate reference data for research 17 , 18 .…”

“…These monkeys were between 30 and 50 months of age and had weights between 2.5 and 5.5 kg. OCT measurements were performed under anesthesia, as previously reported, with the pupil dilated using the Spectralis HRA + OCT Heidelberg device (Heidelberg Engineering, Heidelberg, Germany) 16 . The scanning protocol was the same for all animals and included a horizontal line scan pattern (centered over the fovea) with a size of 20° × 20°, consisting of 25 B-scans spaced 221 μm apart (scan length 5.3 mm, 512 × 496 pixels, scan depth 1.9 mm).…”

“…The obtained images were analyzed via two automatic processes (Fig. 1 ): (1) Using a previously developed and validated deep learning (DL) procedure, the OCT images were segmented into their corresponding compartments 16 , allowing the choroid to be segmented just above the choriocapillaris down to the choroid-sclera junction.

Figure 1 Illustration of automatic deep learning choroidal segmentation.

…”

“…In summary the DL procedure used a modified U-Net architecture 21 , a type of convolutional neural network (CNN). Training and validation of the CNN was done using a representative subset of the OCT cynomolgus monkey data set 16 . This subset—the ground truth (GT)—contains 1100 B-scans obtained from 44 eyes from 44 individuals (each eye contributed 25 B-scans).…”

“…Because the internal limiting membrane (ILM) line extracted as the border between the segmentation of the vitreous and retinal compartments was rather noisy, the extracted ILM was smoothed using a moving average with a two-dimensional sampling window to determine the deepest point within the fovea. Thus, it was possible to automatically identify and define the deepest point of the fovea from the smoothed ILM, which was denoted as the nulla 16 . The nulla was therefore defined as the deepest position within a series of OCT B-scans of a particular macular OCT volume scan.…”