A-eye: Automated 3D Segmentation of Healthy Human Eyeand Orbit Structures and Axial Length Extraction
Barranco, J.; Luyken, A. K.; Stachs, O.; Langner, S.; Franceschiello, B.; Bach Cuadra, M.
Show abstract
This study addresses the need for accurate 3D segmentation of the human eye and orbit from MRI to improve ophthalmic diagnostics. Past efforts focused on small sample sizes and varied imaging methods. Here, two techniques (atlas-based registration and supervised deep learning) are tested for automated segmentation on a large T1-weighted MRI dataset. Results show accurate segmentations of the lens, globe, optic nerve, rectus muscles, and fat. Additionally, the study automates the estimation of axial length, a key biomarker.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- OpenMAP-T1: A Rapid Deep Learning Approach to Parcellate 280 Anatomical Regions to Cover the Whole Brain 96%
- Automatic Landmark-guided Bijective Brain Image Registration by Composing Region-based Locally Diffeomorphic Warpings 95%
- Deep Bayesian networks for uncertainty estimation and adversarial resistance of white matter hyperintensity segmentation 95%
Similar papers in this journal
Similar papers in this journal
- pyKNEEr: An image analysis workflow for open and reproducible research on femoral knee cartilage 96%
- Freewater EstimatoR using iNtErpolated iniTialization (FERNET): Toward Accurate Estimation of Free Water in Peritumoral Region Using Single-Shell Diffusion MRI Data 94%
- Differentiation of white matter histopathology using b-tensor encoding and machine learning 94%
Similar papers in this journal
- GLAPAL-H: Global, Local, And Parts Aware Learner for Hydrocephalus Infection Diagnosis in Low-Field MRI 94%
- An Automated and Robust Tool for Musculoskeletal and Finite Element Modeling of the Knee Joint 93%
- Maximum Classifier Discrepancy Generative Adversarial Network for Jointly Harmonizing Scanner Effects and Improving Reproducibility of Downstream Tasks 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.