Related Experiment Video
Updated: Jun 15, 2025

Longitudinal Morphological and Physiological Monitoring of Three-dimensional Tumor Spheroids Using Optical Coherence Tomography
Published on: February 9, 2019
Robust AMD Stage Grading with Exclusively OCTA Modality Leveraging 3D Volume
Haochen Zhang1, Anna Heinke2, Carlo Miguel B Galang2
1Electrical and Computer Engineering Department, UC San Diego.
Deep learning classifiers using Optical Coherence Tomography Angiography (OCTA) can accurately grade Age-related Macular Degeneration (AMD). Analyzing 3D OCTA volumes directly improves accuracy, outperforming human experts in AMD staging.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Age-related Macular Degeneration (AMD) causes central vision loss.
- Optical Coherence Tomography Angiography (OCTA) visualizes pathogenic vessels in AMD.
- Current diagnostic methods can be limited.
Purpose of the Study:
- To evaluate OCTA's effectiveness for AMD stage grading using deep learning.
- To develop a robust deep learning model for AMD classification from OCTA data.
- To compare AI performance against human experts in AMD staging.
Main Methods:
- Developed a 2D classifier using OCTA projections.
- Identified segmentation errors impacting classification accuracy.
- Proposed a 3D OCTA volume analysis using a 2D CNN with projection supervision.
Main Results:
- Achieved over 80% accuracy in a four-stage AMD grading task.
- The 3D volume analyzer demonstrated robustness with segmentation errors.
- AI performance significantly surpassed the 60% accuracy of human experts.
Conclusions:
- OCTA contains sufficient information for accurate AMD stage grading.
- Direct 3D OCTA volume analysis enhances classification robustness and accuracy.
- Deep learning models show promise for objective and precise AMD assessment.
Related Concept Videos
Unsoundness of Aggregate due to Volume Change
Control Volume and System Representations
The control volume approach considers a stationary region in space through which fluid flows. This region is bounded by a control surface. For instance, in the case of water...
Depth Perception and Spatial Vision
Differential Leveling
Perceiving Loudness, Pitch, and Location
Place theory, or place coding, suggests that different pitches are heard because various sound waves activate specific locations along the cochlea's basilar membrane. The brain determines the pitch of a sound by...
Introduction and Methods of Leveling

