Pathology Insensitive Quality Assessment for Optical Coherence Tomography

P.C. Barnum1, M. Chen2, H. Ishikawa3, G. Wollstein3, J. Schuman3. 

1. Robotics Institute, Carnegie Mellon University, Pittsburgh, PA; 
2. Intel Research, Pittsburgh, PA; 
3. Eye and Ear Institute, Ophthalmology and Visual Science Research Center, 
Dept. of Ophthalmology, University of Pittsburgh School of Medicine, Pittsburgh, PA.

Abstract:


Purpose: The quality of optical coherence tomography (OCT) images has a significant impact 
on measurements and diagnosis. The device provides an overall score for signal strength, 
but when only part of the image is poor, global quality metrics might not suffice. 
The purpose of this study was to develop an automated method for determining the quality
of individual A-scans, in a way that is insensitive to pathology.


Methods: Three OCT experts independently labeled the quality of each A-scan in 270 images.
There were 90 images each of the macula, optic nerve head, and circumpapillary retinal 
nerve fiber layer. To test robustness to pathology, one-third of the eyes had no glaucoma,
one-third had early glaucoma, and one-third had advanced glaucoma. Using the experts’
labels, we trained a hierarchy of support vector machines (SVM) at multiple scales and 
used histogram-based metrics. Specifically for the SVM, we projected different sized 
groups of A-scans into a high dimensional space, then compute the hyperplane that gives
the largest margin. To improve the classification performance, the scans were normalized
to be insensitive to various effects, such as pathology and eye movement during acquisition.


Results: There was strong agreement between the experts and our new algorithm. The following
chart is the percent agreement between each of the three experts, their mode, and our new
algorithm

         Wollstein Ishikawa Schuman Mode Algorithm 
Wollstein - 93 94 97 93 
Ishikawa 93 - 92 95 95 
Schuman 94 92 - 97 92 
Mode 97 95 97 - 95 
Algorithm 93 95 92 95 - 

For our algorithm, the area under the receiver operating characteristic curve (AUC) was 0.94
 for the macula, 0.96 for the optic nerve head, and 0.95 for the retinal nerve fiber layer.


Conclusions: We show that our automated algorithm performs at the level of inter-operator
variability in discriminating local OCT image quality. This localized assessment can be
used to improve the performance of segmentation and image fusion algorithms.


Author Disclosure Information: P.C. Barnum, None; M. Chen, None; H. Ishikawa, None; 
G. Wollstein, Carl Zeiss Meditec, Inc., F; 
J. Schuman, Alcon, F; Allergan, F; Carl Zeiss Meditec, Inc., F; Merck, F; Optoview, F; 
Heidelberg Engineering, F; Carl Zeiss Meditec, Inc., P; Alcon, R; Allergan, R; 
Carl Zeiss Meditec, Inc, R; Clarity, R; Merck, R; Heidelberg Engineering, R.