TY - GEN
T1 - An integrated framework for automatic clinical assessment of diabetic retinopathy grade using spectral domain OCT images
AU - Eltanboly, Ahmed
AU - Ghazaf, Mohammed
AU - Khalil, Ashraf
AU - Shalaby, Ahmed
AU - Mahmoud, Ali
AU - Switala, Andy
AU - El-Azab, Magdi
AU - Schaal, Shlomit
AU - El-Baz, Ayman
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2018/5/23
Y1 - 2018/5/23
N2 - Diabetic retinopathy (DR) is a progressive disease and its detection at an early stage is crucial for saving a patient's vision. In this paper, an enhanced computer-assisting diagnostic (CAD) system is developed for the discovery and grading of non-proliferative DR from optical coherence tomography (OCT) images. The proposed CAD system elaborates three sequential stages. Initially, 12 distinct retina layers are localized using our previously developed segmentation approach based on an integrated joint model that combines shape, intensity, and spatial information. Secondly, three features, namely the reflectivity, curvature, and thickness are quantitatively measured from the segmented layers. Finally, a two-stage deep fusion classification network (DFCN), trained by stacked non-negativity constraint autoencoder (SNCAE), is used first to classify the subject as normal or DR, then assess the grade of DR as either early stage or mild/moderate. Using 'leave-one-subject-out' experiments on a dataset of 74 OCT images, the CAD system distinguished between normal and DR subjects with a 93% accuracy (sensitivity =91%, specificity =97%) and achieved a 98% correct classification between early stage and mild/moderate DR. These results confirm the proposed framework as a reliable non-invasive diagnostic tool.
AB - Diabetic retinopathy (DR) is a progressive disease and its detection at an early stage is crucial for saving a patient's vision. In this paper, an enhanced computer-assisting diagnostic (CAD) system is developed for the discovery and grading of non-proliferative DR from optical coherence tomography (OCT) images. The proposed CAD system elaborates three sequential stages. Initially, 12 distinct retina layers are localized using our previously developed segmentation approach based on an integrated joint model that combines shape, intensity, and spatial information. Secondly, three features, namely the reflectivity, curvature, and thickness are quantitatively measured from the segmented layers. Finally, a two-stage deep fusion classification network (DFCN), trained by stacked non-negativity constraint autoencoder (SNCAE), is used first to classify the subject as normal or DR, then assess the grade of DR as either early stage or mild/moderate. Using 'leave-one-subject-out' experiments on a dataset of 74 OCT images, the CAD system distinguished between normal and DR subjects with a 93% accuracy (sensitivity =91%, specificity =97%) and achieved a 98% correct classification between early stage and mild/moderate DR. These results confirm the proposed framework as a reliable non-invasive diagnostic tool.
KW - DFCN
KW - Early DR
KW - Mild/moderate DR
KW - NPDR
KW - OCT
KW - SNCAE
UR - http://www.scopus.com/inward/record.url?scp=85048124468&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85048124468&partnerID=8YFLogxK
U2 - 10.1109/ISBI.2018.8363841
DO - 10.1109/ISBI.2018.8363841
M3 - Conference contribution
AN - SCOPUS:85048124468
T3 - Proceedings - International Symposium on Biomedical Imaging
SP - 1431
EP - 1435
BT - 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 15th IEEE International Symposium on Biomedical Imaging, ISBI 2018
Y2 - 4 April 2018 through 7 April 2018
ER -