Related Experiment Video
Updated: Jan 27, 2026

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning
Published on: April 28, 2022
Embedded deep learning in ophthalmology: making ophthalmic imaging smarter.
Petteri Teikari1, Raymond P Najjar1, Leopold Schmetterer1
1Visual Neurosciences Group, Singapore Eye Research Institute, Singapore.
Deep learning in ophthalmology is advancing, with a focus on embedding AI into imaging devices for automated, high-quality image acquisition. This integration promises more accurate diagnostics and efficient data management in clinical practice.
Area of Science:
- Ophthalmology
- Artificial Intelligence
- Medical Imaging
Background:
- Deep learning (DL) shows promise in ophthalmology for diagnosis and prognosis.
- Limited research exists on embedding DL systems within ophthalmic imaging devices for automated acquisition.
- Current DL applications primarily focus on post-acquisition analysis rather than integrated image capture.
Purpose of the Study:
- To review current and future directions of 'active acquisition'-embedded deep learning in ophthalmology.
- To explore how embedded DL can enhance image quality with minimal human intervention.
- To discuss the potential impact of embedded DL on clinical diagnostics and data management.
Main Methods:
- Review of existing literature and future trends in embedded deep learning for ophthalmic imaging.
- Discussion of hardware advancements enabling low-cost, high-performance embedded systems.
- Exploration of a three-layer computation framework (edge, fog, cloud) for clinical systems.
Main Results:
- 'Active acquisition' via embedded DL can significantly improve image quality and reduce operator intervention.
- Embedded DL systems, particularly at the edge layer, can act as automatic data curation tools.
- Enhanced data quality from edge-layer processing benefits both electronic health records and cloud-based data mining.
Conclusions:
- Embedding deep learning into ophthalmic imaging devices ('active acquisition') is a key future direction.
- This approach enhances image quality, leading to more robust DL-based clinical diagnostics.
- Improved data curation at the edge layer optimizes data for electronic health records and cloud analytics.
Related Concept Videos
Avoidance Learning and Learned Helplessness
Avoidance learning occurs when an organism learns that a specific behavior can prevent an unpleasant outcome. For example, a student who receives a bad grade may start studying harder to avoid future poor grades. This behavior persists even when the negative outcome is no longer present. Avoidance learning is powerful because it maintains behavior in the absence of the...
Learning Disabilities
Dyslexia
Dyslexia is a...
Associative Learning
Classical conditioning, also known...
Purposive Learning
Observational Learning
Introduction to Learning
In contrast to learned behaviors, unlearned behaviors such as crying, sexual...

