advertisement

Editors Selection IGR 10-3

Clinical Examination Methods: ONH imaging: A 3D perspective

Comment by Linda Zangwill & Jalil Jalili on:

This study introduces a novel algorithm for evaluating the Optic Nerve Head (ONH) parameters using Light Field (LF) fundus camera data, comparing the results with those obtained from Optical Coherence Tomography (OCT). This camera has several advantages over current fundus cameras including fewer motion artifacts, and the ability to generate 3D images and geometric ONH parameters.

Using a small sample of 17 healthy subjects a custom algorithm that uses LF camera's images to measure ONH parameters was compared to OCT measurements. The proposed algorithm is comprehensive and sophisticated, particularly in its approach to preprocessing, surface reconstruction, and ONH parameter determination based on the rising and falling edges in the 3D point cloud analysis. The results suggest that the median values of the ONH parameters derived from LF 3D imaging were largely in agreement with the OCT data. However, additional analysis such as scatterplots that might offer more precise evaluation of the algorithm's performance by showing the direction and magnitude of differences found at the image and also patient level. In addition, the data acquired could be used to measure the repeatability and reproducibility of the LDF ONH measurements as several scans were acquired on individuals over a short period of time.

As this first study included only healthy participants, the effectiveness of the algorithm in broader clinical scenarios, including patients with glaucoma and myopia who exhibit a range of ONH variations will need to be evaluated before determining the role in disease detection. As the authors suggest, there are several critical limitations of the LF camera including the 30-minute image acquisition time, lack of metric calibration, and wavefront errors, reflections and scattered light. The results suggest that if strategies proposed to overcome these limitations are successful, the LF fundus camera, combined with the processing algorithms, could be an addition tool for evaluating the ONH.

• ← Previous
• Next →