Related Experiment Video
Updated: Jul 26, 2025
06:15
Author Spotlight: Anterior HR-OCT as a Non-Invasive Tool for Characterizing Ocular Surface Squamous Neoplasia
Published on: August 9, 2024
1.3K
Artificial intelligence for ocular oncology
Neslihan Dilruba Koseoglu1, Zélia Maria Corrêa2,3, T Y Alvin Liu1
1Wilmer Eye Institute, Johns Hopkins University, Baltimore, Maryland.
Current Opinion in Ophthalmology
|June 16, 2023
Summary
Deep learning (DL) and machine learning (ML) show promise in detecting and predicting outcomes for eye cancers. DL is leading prognostication for uveal melanoma (UM), despite challenges with rare disease data.
Area of Science:
- Ophthalmology
- Oncology
- Artificial Intelligence
Background:
- Intraocular and ocular surface malignancies pose significant diagnostic and prognostic challenges.
- Machine learning (ML) and deep learning (DL) offer novel computational approaches to address these challenges.
Approach:
- This review synthesizes recent advancements in applying DL and classical ML techniques to ocular oncology.
- Focus is placed on the utility of these AI methods in the detection and prognostication of eye cancers.
Key Points:
- Recent research highlights the application of DL and ML in predicting patient outcomes for uveal melanoma (UM).
- Deep learning demonstrates superior performance in prognostication for ocular oncological conditions, especially UM.
Conclusions:
- Deep learning is becoming the primary ML technique for prognostication in ocular oncology, particularly for uveal melanoma.
- The clinical utility of DL in rare conditions like UM may be constrained by data availability due to disease rarity.
