No abstract

No abstract

PurposeTo evaluate a machine learning AI‐tool for automatic segmentation of the waist of the nerve fiber layer at the optic nerve head (ONH).MethodsOCT‐volumes of the ONH were captured from eyes with early to manifest glaucoma (OCT‐2000) and eyes without glaucoma (OCT‐Triton). The angularly resolved segmentation consisted of first detecting the inner limit of the retina in the OCT volume with AI‐model I. Subsequently, the OCT volume was resampled into 500 radial images distributed at equidistant angles in the frontal plane. In each radial image, the Optic nerve head Pigment epithelium Central Limit (OPCL) was identified manually with a specially developed tool and automatically with AI‐model II. Model II was trained for OPCL location on a training set of radial images. Ground truth was provided as manual annotations. Finally, the Pigment‐epithelium‐Inner‐limit of the retina‐Minimal Distance (PIMD) was estimated at each angle as the shortest distance between OPCL and the closest point on the inner limit of the retina. PIMD estimated both automatically and based on manual annotation of OPCL were compared in a test data set. All radial images in the test data set were manually annotated by three different annotators.ResultsFor both groups of eyes examined, the automatic and manual estimates of the angular distribution of PIMD overlapped. The eyes without glaucoma expressed the expected peak thickness of the waist of the nerve fiber layer inferiorly and superiorly, while the glaucomatous eyes lost most of this angular dependence. A 95% confidence interval for the mean difference of automatically estimated PIMD between eyes without glaucoma and glaucomatous eyes was 220 ± 85 μm. An analysis of interobserver variability in PIMD‐estimates demonstrated minor differences.ConclusionsAI‐based estimation of the waist of the nerve fiber layer at the ONH angularly resolved is reliable and precise enough to detect a difference between early to manifest glaucoma and non‐glaucoma.

PurposeGlaucoma leads to pathological loss of axons in the retinal nerve fibre layer at the optic nerve head (ONH). This study aimed to develop a strategy for the estimation of the cross‐sectional area of the axons in the ONH. Furthermore, improving the estimation of the thickness of the nerve fibre layer, as compared to a method previously published by us.MethodsIn the 3D‐OCT image of the ONH, the central limit of the pigment epithelium and the inner limit of the retina, respectively, were identified with deep learning algorithms. The minimal distance was estimated at equidistant angles around the circumference of the ONH. The cross‐sectional area was estimated by the computational algorithm. The computational algorithm was applied on 16 non‐glaucomatous subjects.ResultsThe mean cross‐sectional area of the waist of the nerve fibre layer in the ONH was 1.97 ± 0.19 mm2. The mean difference in minimal thickness of the waist of the nerve fibre layer between our previous and the current strategies was estimated as CIμ (0.95) 0 ± 1 μm (d.f. = 15).ConclusionsThe developed algorithm demonstrated an undulating cross‐sectional area of the nerve fibre layer at the ONH. Compared to studies using radial scans, our algorithm resulted in slightly higher values for cross‐sectional area, taking the undulations of the nerve fibre layer at the ONH into account. The new algorithm for estimation of the thickness of the waist of the nerve fibre layer in the ONH yielded estimates of the same order as our previous algorithm.