Related Experiment Video
Updated: Jul 31, 2025
Longitudinal Morphological and Physiological Monitoring of Three-dimensional Tumor Spheroids Using Optical Coherence Tomography
Published on: February 9, 2019
Accurate C/D ratio estimation with elliptical fitting for OCT image based on joint segmentation and detection
Chenan Xu1, Zhongyue Chen2, Xiao Zhang2
1State Key Laboratory of Radiation Medicine and Protection, Collaborative Innovation Center of Radiological Medicine of Jiangsu Higher Education Institutions, and School for Radiological and Interdisciplinary Sciences (RAD-X), Soochow University, Suzhou, 215006, China.
A new deep learning method accurately measures the optic cup-to-disc ratio (C/D ratio) using optical coherence tomography (OCT) scans. This automated approach shows strong correlation with manual annotations and clinical data, offering valuable clinical application potential.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Accurate cup-to-disc ratio (C/D ratio) estimation is crucial in ophthalmic examinations.
- Current automated C/D ratio measurement methods lack efficiency.
- Optical coherence tomography (OCT) is a key imaging modality in ophthalmology.
Purpose of the Study:
- To develop and validate a novel, efficient, and accurate automated method for measuring the C/D ratio from OCT images.
- To assess the performance of the proposed method against manual annotations and existing commercial OCT machines.
Main Methods:
- Utilized an end-to-end deep convolution network for segmenting inner limiting membrane (ILM) and Bruch's membrane opening (BMO) terminations.
- Implemented an ellipse fitting technique for post-processing optic disc edge detection.
- Evaluated the method on 41 normal subjects using OCT scans from BV1000, Topcon 3D OCT-1, and Nidek ARK-1 machines.
Main Results:
- The proposed method demonstrated a strong correlation (coefficient=0.84) with manual C/D ratio annotations.
- BV1000's C/D ratio measurements (96.34% < 0.6) were closest to clinical statistics among tested OCT machines.
- The method showed robust performance in optic cup and disc detection and C/D ratio measurement.
Conclusions:
- The developed deep learning method provides an efficient and accurate approach for automated C/D ratio measurement from OCT scans.
- The method's results closely align with clinical reality, suggesting significant clinical application value in ophthalmic screening.
- This automated technique has the potential to improve the efficiency of glaucoma and other optic nerve head disease diagnosis.
