Related Experiment Video
Updated: Aug 22, 2025
07:03
Medical-grade Sterilizable Target for Fluid-immersed Fetoscope Optical Distortion Calibration
Published on: February 23, 2017
7.8K
A Generic Pixel Pitch Calibration Method for Fundus Camera via Automated ROI Extraction
Tengfei Long1, Yi Xu2, Haidong Zou2,3
1Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, China.
Sensors (Basel, Switzerland)
|November 11, 2022
Summary
This study introduces a new method for calibrating pixel pitch in fundus images, crucial for accurate disease diagnosis. The approach uses automated region of interest and optic disc detection, eliminating the need for specialized equipment and improving measurement accuracy.
Area of Science:
- Ophthalmology
- Medical Imaging
- Biomedical Engineering
Background:
- Accurate pixel pitch calibration is vital for quantitative analysis of fundus images in diagnosing diseases like diabetes and arteriosclerosis.
- Conventional calibration methods require specific camera parameters or specialized images (e.g., chessboard), which are often inaccessible and lack generalizability.
- Existing methods limit the accurate measurement of fundus structures and quantitative diagnosis.
Purpose of the Study:
- To develop and validate a novel, accessible method for pixel pitch calibration in fundus images.
- To enable accurate quantitative measurements of fundus structures without requiring specific camera details or chessboard images.
- To improve the generalizability and accuracy of fundus image analysis for clinical applications.
Main Methods:
- Automated detection of the region of interest (ROI) and optic disc within fundus images.
- Quantitative analysis of the diameter ratio between the ROI and the optic disc (ROI-disc ratio).
- Statistical estimation of pixel pitch using prior knowledge of average optic disc diameter and the calculated ROI-disc ratio from a large dataset of fundus images.
Main Results:
- The average ROI-disc ratio was found to be approximately constant (6.404 ± 0.619 pixels) across 40,600 fundus images from various cameras with a 45° field-of-view (FOV).
- The proposed method accurately estimated pixel pitches for different fundus cameras (Canon CR2, Topcon NW400, Zeiss Visucam 200, Newvision RetiCam 3100), with biases less than 5% compared to ISO 10940:2009 standards.
- Pixel pitches were determined as 6.825 ± 0.666 μm, 6.625 ± 0.647 μm, 5.793 ± 0.565 μm, and 5.884 ± 0.574 μm, respectively.
Conclusions:
- The developed method provides an accurate and accessible approach for pixel pitch calibration in fundus imaging.
- This technique eliminates the need for camera-specific parameters or chessboard images, enhancing usability and generalizability.
- The findings facilitate accurate quantitative measurements of fundus structures, supporting improved diagnosis and treatment of eye diseases.
