CornOrb: A Multimodal Dataset of Orbscan Corneal Topography and Clinical Annotations for Keratoconus Detection

Mohammed El Amine Lazouni; Leila Ryma Lazouni; Zineb Aziza Elaouaber; Mohammed Ammar; Sofiane Zehar; Mohammed Youcef Bouayad Agha; Ahmed Lazouni; Amel Feroui

cs.CVMar 22, 2026

CornOrb: A Multimodal Dataset of Orbscan Corneal Topography and Clinical Annotations for Keratoconus Detection

Authors: Mohammed El Amine Lazouni, Leila Ryma Lazouni, Zineb Aziza Elaouaber, Mohammed Ammar, Sofiane Zehar, Mohammed Youcef Bouayad Agha, Ahmed Lazouni, Amel Feroui,

Organizations: Biomedical Engineering Laboratory, Abou Bakr Belkaid University, Tlemcen, Algeria · LIST Laboratory, M’Hamed Bougara Boumerdes University, Algeria · Lazouni Clinic, Tlemcen, Algeria · College of Excellence, University of Baghdad, Iraq ·

Abstract

In this paper, we present CornOrb, a publicly accessible multimodal dataset of Orbscan corneal topography images and clinical annotations collected from patients in Algeria. The dataset comprises 1,454 eyes from 744 patients, including 889 normal eyes and 565 keratoconus cases. For each eye, four corneal maps are provided (axial curvature, anterior elevation, posterior elevation, and pachymetry), together with structured tabular data including demographic information and key clinical parameters such as astigmatism, maximum keratometry (Kmax), central and thinnest pachymetry, and anterior/posterior asphericity. All data were retrospectively acquired, fully anonymized, and pre-processed into standardized PNG and CSV formats to ensure direct usability for artificial intelligence research. This dataset represents one of the first large-scale Orbscan-based resources from Africa, specifically built to enable robust AI-driven detection and analysis of keratoconus using multimodal data. The data are openly available at Zenodo.

Figures & tables

Panel of Figure 1 — Figure 1: Representative Orbscan 3 corneal maps included in the CornOrb dataset: (a) axial curvature, (b) anterior elevation, (c) posterior elevation, and (d) pachymetry. Each map is standardized to a resolution of $560\times 560$ pixels and saved in PNG format.

Panel of Figure 2 — Figure 2: Schematic representation of the CornOrb dataset structure. Data are organized hierarchically by patient (from 001 to 744) and eye (OD = right eye, OS = left eye). Each eye includes four Orbscan 3 corneal maps. A CSV file accompanies the images with demographic and clinical annotations. Note: D = diopters, the standard unit for refractive power and keratometry.

Variable name	Unit / Type	Description
patient_code	string	Unique pseudonymized identifier assigned to each patient.
sex	categorical (M/F)	Patient sex.
age	years (integer)	Age at examination.
eye	categorical (OD/OS)	Eye laterality: OD = right eye, OS = left eye.
astigmatism_D	diopters (float)	Magnitude of corneal astigmatism.
astigmatism_axis	degrees (0–180)	Axis of astigmatism.

Table 1: Data dictionary for the clinical_annotations.csv file.

Variable	Normal (n=889)	Keratoconus (n=565)	$p_{\text{BH}}$
Age (years)	$33.5\pm 7.4$	$27.9\pm 7.1$	$<0.001$
Central pachymetry ( $\mu$ m)	$530\pm 43$	$465\pm 54$	$<0.001$
Thinnest pachymetry ( $\mu$ m)	$521\pm 44$	$427\pm 59$	$<0.001$
K max (D)	$43.7\pm 2.9$	$53.1\pm 6.9$	$<0.001$
Astigmatism (D)	$-1.35\pm 0.91$	$-5.32\pm 3.32$	$<0.001$
Anterior asphericity	$-0.15\pm 0.45$	$-1.20\pm 0.81$	$<0.001$

Table 2: Summary statistics with one additional column reporting the BH-adjusted

p

-value for the Normal vs. Keratoconus comparison.

Panel of Figure 3 — Figure 3: Distributions of quantitative clinical measures by diagnosis (Normal vs. Keratoconus). Panels display age (years), K max (D), central pachymetry ( $\mu$ m), astigmatism (D), and anterior/posterior asphericity (dimensionless). Colors: Normal (blue), Keratoconus (red).

Figure 4: The Bausch & Lomb Orbscan 3 device used for all corneal topography acquisitions in the CornOrb dataset.

Medical Image Benchmarks

Medical image benchmarks compare AI methods on curated clinical images, tasks, and evaluation protocols to measure accuracy, robustness, and clinical usefulness.

See all 337 papers

Fundus Image-based Glaucoma Screening via Retinal Knowledge-Oriented Dynamic Multi-Level Feature Integration
Apr 14, 2026Chi Liu, Yuzhuo Zhou, Sheng Shen +9Glaucoma Detection Medical Image Analysis
Faculty of Data Science, City University of Macau, Avenida Padre Tomás Pereira, Taipa, 999078, Macao, China · State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangdong Provincial Key Laboratory of Ophthalmology and Vision Science, Guangdong Provincial Clinical Research Center for Ocular Diseases, Guangzhou, 510000, Guangdong, China · Hainan International College, Minzu University of China, Li’an International Education Innovation Pilot Zone, Lingshui, 572423, Hainan, China +4
Detecting Glaucoma Across Multi-ethnic Myopic and Non-Myopic Populations Using an Uncertainty-Aware Vision Transformer: A Multicentre Model Development and Validation Study
Sep 24, 2026Raghavan Lavanya, Yangqin Feng, Ten Cheer Quek +36Fundus Imaging Glaucoma Detection
Singapore Eye Research Institute, Singapore National Eye Centre, Singapore · Ophthalmology & Visual Sciences Academic Clinical Program (Eye ACP), Duke-NUS Medical School, Singapore · Institute of Advanced Intelligence and Computing, Agency for Science, Technology and Research, Singapore +14
ARCOS: Zero-shot Boundary Localization for Corneal Layer Segmentation Across Optical Coherence Tomography Devices
Sep 3, 2026Nuno Vivas Brás, Benjamin Memmi, Maëlle Bouhassane +4Domain Generalization Optical Coherence Tomography
Laboratoire d’Optique et Biosciences, CNRS, INSERM, ´Ecole polytechnique, Institut Polytechnique de Paris, Palaiseau, France · GRC 32, Transplantation et Th´erapies Innovantes de la Corn´ee, Sorbonne Universit´e, Centre Hospitalier National d’Ophtalmologie des Quinze-Vingts, Paris, France

Apr 14, 2026J/K move · Enter open · S save

Fundus Image-based Glaucoma Screening via Retinal Knowledge-Oriented Dynamic Multi-Level Feature Integration

Chi Liu, Yuzhuo Zhou, Sheng Shen, Zongyuan Ge+8

Faculty of Data Science, City University of Macau, Avenida Padre Tomás Pereira, Taipa, 999078, Macao, China · State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangdong Provincial Key Laboratory of Ophthalmology and Vision Science, Guangdong Provincial Clinical Research Center for Ocular Diseases, Guangzhou, 510000, Guangdong, China · Hainan International College, Minzu University of China, Li’an International Education Innovation Pilot Zone, Lingshui, 572423, Hainan, China+4

Glaucoma Detection Medical Image Analysis Medical Diagnosis Retinal Disease Classification Fundus Imaging

Open paper AskPDF arXiv

While deep learning has advanced automated glaucoma screening via color fundus photography, existing purely data-driven models often overfit to confounding imaging artifacts and struggle to capture unpredictable pathological cues located beyond predefined anatomical boundaries. To address these limitations, we propose a retinal knowledge-oriented framework that synergizes dynamic multi-scale feature learning with domain-specific anatomical priors. The proposed architecture adopts a tri-branch structure to jointly model the global retinal context, the structural characteristics of the optic cup and disc, and dynamically cropped pathological regions. Specifically, we devise a Dynamic Window Mechanism (DWM) that adaptively discovers diagnostically informative patches via image-level supervision. Furthermore, we introduce a Knowledge-Enhanced Convolutional Block Attention Module (KE-CBAM) that explicitly incorporates retinal priors from a pre-trained foundation model RETFound to guide spatial attention, preventing the network from assigning spurious weights to irrelevant background noise. Extensive evaluations on the large-scale AIROGS dataset demonstrate that our method achieves a state-of-the-art AUC of

98.5\%

and an accuracy of

94.6\%

. More importantly, the integration of anatomical priors effectively mitigates the inherent class imbalance, significantly improving the detection of referable glaucoma cases. Additional validations on the SMDG-19 benchmark further confirm its superior cross-domain generalization, indicating that our contribution provides a robust, interpretable, and scalable solution for real-world clinical glaucoma diagnosis. Our code is available at https://github.com/magiczhuo/Glaucoma-Detection

Abstract

Medical Image Benchmarks

Explore similar work

Fundus Image-based Glaucoma Screening via Retinal Knowledge-Oriented Dynamic Multi-Level Feature Integration

Detecting Glaucoma Across Multi-ethnic Myopic and Non-Myopic Populations Using an Uncertainty-Aware Vision Transformer: A Multicentre Model Development and Validation Study

ARCOS: Zero-shot Boundary Localization for Corneal Layer Segmentation Across Optical Coherence Tomography Devices