TY - JOUR
T1 - Deep learning for quality assessment of optical coherence tomography angiography images
AU - Dhodapkar, Rahul M.
AU - Li, Emily
AU - Nwanyanwu, Kristen
AU - Adelman, Ron
AU - Krishnaswamy, Smita
AU - Wang, Jay C.
N1 - Publisher Copyright:
© 2022, The Author(s).
PY - 2022/12
Y1 - 2022/12
N2 - Optical coherence tomography angiography (OCTA) is an emerging non-invasive technique for imaging the retinal vasculature. While there are many promising clinical applications for OCTA, determination of image quality remains a challenge. We developed a deep learning-based system using a ResNet152 neural network classifier, pretrained using ImageNet, to classify images of the superficial capillary plexus in 347 scans from 134 patients. Images were also manually graded by two independent graders as a ground truth for the supervised learning models. Because requirements for image quality may vary depending on the clinical or research setting, two models were trained—one to identify high-quality images and one to identify low-quality images. Our neural network models demonstrated outstanding area under the curve (AUC) metrics for both low quality image identification (AUC = 0.99, 95%CI 0.98–1.00, κ = 0.90) and high quality image identification (AUC = 0.97, 95%CI 0.96–0.99, κ = 0.81), significantly outperforming machine-reported signal strength (AUC = 0.82, 95%CI 0.77–0.86, κ= 0.52 and AUC = 0.78, 95%CI 0.73–0.83, κ = 0.27 respectively). Our study demonstrates that techniques from machine learning may be used to develop flexible and robust methods for quality control of OCTA images.
AB - Optical coherence tomography angiography (OCTA) is an emerging non-invasive technique for imaging the retinal vasculature. While there are many promising clinical applications for OCTA, determination of image quality remains a challenge. We developed a deep learning-based system using a ResNet152 neural network classifier, pretrained using ImageNet, to classify images of the superficial capillary plexus in 347 scans from 134 patients. Images were also manually graded by two independent graders as a ground truth for the supervised learning models. Because requirements for image quality may vary depending on the clinical or research setting, two models were trained—one to identify high-quality images and one to identify low-quality images. Our neural network models demonstrated outstanding area under the curve (AUC) metrics for both low quality image identification (AUC = 0.99, 95%CI 0.98–1.00, κ = 0.90) and high quality image identification (AUC = 0.97, 95%CI 0.96–0.99, κ = 0.81), significantly outperforming machine-reported signal strength (AUC = 0.82, 95%CI 0.77–0.86, κ= 0.52 and AUC = 0.78, 95%CI 0.73–0.83, κ = 0.27 respectively). Our study demonstrates that techniques from machine learning may be used to develop flexible and robust methods for quality control of OCTA images.
UR - http://www.scopus.com/inward/record.url?scp=85135835980&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85135835980&partnerID=8YFLogxK
U2 - 10.1038/s41598-022-17709-8
DO - 10.1038/s41598-022-17709-8
M3 - Article
C2 - 35962007
AN - SCOPUS:85135835980
SN - 2045-2322
VL - 12
JO - Scientific reports
JF - Scientific reports
IS - 1
M1 - 13775
ER -