Glaucoma is a chronic progressive disease that causes loss of retinal ganglion cells, which impairs the visual field. In optic coherence tomography (OCT) image, the retinal ganglion cell (RGC) axons in the optic nerve head (ONH) can be quantified as the minimal thickness from the ONH Pigmental epithelium Central Limit (OPCL) to the Inner limit of the Retina Closest Point (IRCP). Alternatively, the minimal cross-sectional surface area can be measured. In peripapillary atrophy, the morphometry of the retinal pigmental epithelium is affected. Purpose: To design and test a new computational algorithm for estimation of Pigment epithelium to Inner limit of the Retina Minimal Area (PIMA) and evaluate a new method to estimate the Pigment epithelium to Inner limit of the Retina Minimal Distance (PIMD). OPCL can be detected and annotated by a deep learning algorithm in individuals with peripapillary atrophy. Methods: A deep learning algorithm has been trained to automatically detect OPCL, IRCP and calculate PIMD. A new computational algorithm was developed to estimate PIMA in OCT images of young adults. The mean between the first and second version of estimating PIMD was evaluated. The difference of distance between the ONH center-OPCL and ONH center-atrophic edge was estimated in eyes with peripapillary atrophy. Results: A 95% confidence interval for PIMA-2π was estimated to 1.97 ± 0.19 mm2 (df = 15). A confidence interval for the difference between PIMDv1-2π and PIMDv2-2π was 0 ± 1 μm (df = 15). A 95 % confidence interval for the mean difference between ONH-OPCL and ONH-atrophic edge was estimated to 692 ± 192 µm (df = 5). Conclusions: The computational algorithm for estimation of PIMA was developed and applied. An initial analysis indicated the capacity of the deep learning algorithm to detect OPCL in subjects with PPA. Keywords: deep learning, optic nerve head, ONH, retinal pigmental epithelium, RPE, PIMD, PIMD-2π, minimal distance, PIMA, PIMA-2π, minimal area, peripapillary atrophy, PPA, optic coherence tomography, OCT, glaucoma, quantification, retinal ganglion cell axons
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:uu-502042 |
| Date | January 2023 |
| Creators | Kisonaite, Konstancija |
| Publisher | Uppsala universitet, Oftalmiatrik, Uppsala : Acta Universitatis Uppsaliensis, 2023 |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Licentiate thesis, comprehensive summary, info:eu-repo/semantics/masterThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0026 seconds