Thank you for submitting your work to the JustRAIGS challenge session!¶
Feedback on the review and acceptance of your paper is expected on April 27. Given the short timelines for updating your manuscript and submitting the camera-ready version (deadline on April 30), we’d like to draw your attention to some important general guidelines and requirements.¶
For the manuscripts:¶
Please ensure that the correct template is used (follow the link on https://biomedicalimaging.org/2024/authors-instructions).Make sure that you adhere to the page limit (4 pages, possibly one supplemental page – see https://biomedicalimaging.org/2024/authors-instructions).Include a statement on ethical compliance, according to the instructions on https://biomedicalimaging.org/2024/authors-instructions.¶
Additionally, you need to meet the following requirements:¶
In-person presentation (and not virtual) Full registration to the conference.¶
These guidelines and requirements are enforced because their papers are treated as regular papers to ISBI 2024 (part of the main proceedings). If they do not meet these guidelines and requirements, then their papers will not be accepted. If you require an invitation letter for your visa, please let us know so we can provide you with that as soon as possible after we inform you of the acceptance of your paper.¶
Justified Referral in AI Glaucoma Screening¶
Glaucoma, a leading cause of irreversible blindness, often goes unnoticed in its early stages but can lead to significant vision impairment and hazards like stumbling or traffic accidents in advanced stages. Detectable through changes in the optic nerve head on imaging modalities like color fundus photographs (CFPs) or optical coherence tomography (OCT), early detection and treatment are crucial to halt its progression. CFPs, in particular, are cost-effective for screening, offering vital insights into glaucomatous damage through indicators like neuroretinal rim thinning and optic disc hemorrhages, and provide a baseline record for ongoing monitoring.
To initiate the development of such AI algorithms for glaucoma screening and to evaluate their performance, we propose the Justified Referral in AI Glaucoma Screening (JustRAIGS) challenge, for which we have provided a unique large dataset with over 110k carefully annotated fundus photographs collected from about 60,000 screenees. We have generated a training subset with 101,442 gradable fundus images (from "referable glaucoma" eyes and "no referable glaucoma" eyes) and a test subset with 9,741 fundus images. Each fundus photograph thus has been labeled as either "referable glaucoma" or "no referable glaucoma". In addition, all fundus images of referable glaucoma eyes have been further annotated with up to ten additional labels associated with different glaucomatous features.
In this challenge, participants will be tasked with analyzing the fundus images and assigning each image to one of two classes: "referable glaucoma" or "no referable glaucoma". "Referable glaucoma" refers to eyes where the fundus image exhibits signs or features indicative of glaucoma that require further examination or referral to a specialist. In this case, visual field damage is expected. On the other hand, "no referable glaucoma" refers to cases where the fundus image does not show significant indications of glaucoma and does not require immediate referral. Very early disease, in which visual field damage is not yet expected, would also be classified as ‘"no referable glaucoma". In addition to the referable glaucoma classification, participants will be further instructed to perform multi-label classification for ten additional features related to glaucoma. These features are specific characteristics or abnormalities that may be present in the fundus images of glaucoma patients. The multi-label classification task involves assigning relevant labels to each fundus image based on the presence or absence of these specific features. These additional features provide more detailed information about the specific characteristics observed in the fundus images of "referable glaucoma" cases. By combining both the binary classification task (referable vs. no referable glaucoma) and the multi-label classification task (for the ten additional features), we aim to evaluate the participant's ability to accurately identify and classify fundus images associated with referable glaucoma. The results of this classification task can provide insights into the development of automated systems or algorithms for glaucoma detection, ultimately assisting in the early identification and treatment of glaucoma patients, thereby reducing avoidable visual impairment and blindness from glaucoma.
Task 1: Referral performance¶
Binary classification of referable glaucoma and no referable glaucoma¶
- No referable glaucoma (NRG)
- Referable glaucoma (RG)
Task 2: Justification performance¶
Multi-label classification of ten additional features¶
- Appearance neuroretinal rim superiorly (ANRS)
- Appearance neuroretinal rim inferiorly (ANRI)
- Retinal nerve fiber layer defect superiorly (RNFLDS)
- Retinal nerve fiber layer defect inferiorly (RNFLDI)
- Baring circumlinear vessel superiorly (BCLVS)
- Baring circumlinear vessel inferiorly (BCLVI)
- Nasalisation of vessel trunk (NVT)
- Disc hemorrhages (DH)
- Laminar dots (LD)
- Large cup (LC)