TY - JOUR
T1 - Deep Neural Network Regression for Automated Retinal Layer Segmentation in Optical Coherence Tomography Images
AU - Ngo, Lua
AU - Cha, Jaepyeong
AU - Han, Jae Ho
N1 - Funding Information:
Manuscript received October 10, 2018; revised May 16, 2019; accepted July 23, 2019. Date of publication August 1, 2019; date of current version September 23, 2019. This work was supported in part by the National Research Foundation of Korea (NRF) through the Korean Government (MSIT) under Grant NRF-2017R1A2B2003808 and in part by the LG Yonam Foundation of Korea. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Christophoros Nikou. (Corresponding author: Jae-Ho Han.) L. Ngo and J.-H. Han are with the Brain and Cognitive Engineering Department, Korea University, Seoul 02841, South Korea (e-mail: hanjaeho@ korea.ac.kr).
Funding Information:
This work was supported in part by the National Research Foundation of Korea (NRF) through the Korean Government (MSIT) under Grant NRF-2017R1A2B2003808 and in part by the LG Yonam Foundation of Korea.
Publisher Copyright:
© 1992-2012 IEEE.
PY - 2020
Y1 - 2020
N2 - Segmenting the retinal layers in optical coherence tomography (OCT) images helps to quantify the layer information in early diagnosis of retinal diseases, which are the main cause of permanent blindness. Thus, the segmentation process plays a critical role in preventing vision impairment. However, because there is a lack of practical automated techniques, expert ophthalmologists still have to manually segment the retinal layers. In this paper, we propose an automated segmentation method for OCT images based on a feature-learning regression network without human bias. The proposed deep neural network regression takes the intensity, gradient, and adaptive normalized intensity score (ANIS) of an image segment as features for learning, and then predicts the corresponding retinal boundary pixel. Reformulating the segmentation as a regression problem obviates the need for a huge dataset and reduces the complexity significantly, as shown in the analysis of computational complexity given here. In addition, assisted by ANIS, the method operates robustly on OCT images containing intensity variances, low-contrast regions, speckle noise, and blood vessels, yet remains accurate and time-efficient. In the evaluation of the method conducted using 114 images, the processing time was approximately 10.596 s per image for identifying eight boundaries, and the training phase for each boundary line took only 30 s. Further, the Dice similarity coefficient used for assessing accuracy gave a computed value of approximately 0.966. The absolute pixel distance of manual and automatic segmentation using the proposed scheme was 0.612, which is less than a one-pixel difference, on average.
AB - Segmenting the retinal layers in optical coherence tomography (OCT) images helps to quantify the layer information in early diagnosis of retinal diseases, which are the main cause of permanent blindness. Thus, the segmentation process plays a critical role in preventing vision impairment. However, because there is a lack of practical automated techniques, expert ophthalmologists still have to manually segment the retinal layers. In this paper, we propose an automated segmentation method for OCT images based on a feature-learning regression network without human bias. The proposed deep neural network regression takes the intensity, gradient, and adaptive normalized intensity score (ANIS) of an image segment as features for learning, and then predicts the corresponding retinal boundary pixel. Reformulating the segmentation as a regression problem obviates the need for a huge dataset and reduces the complexity significantly, as shown in the analysis of computational complexity given here. In addition, assisted by ANIS, the method operates robustly on OCT images containing intensity variances, low-contrast regions, speckle noise, and blood vessels, yet remains accurate and time-efficient. In the evaluation of the method conducted using 114 images, the processing time was approximately 10.596 s per image for identifying eight boundaries, and the training phase for each boundary line took only 30 s. Further, the Dice similarity coefficient used for assessing accuracy gave a computed value of approximately 0.966. The absolute pixel distance of manual and automatic segmentation using the proposed scheme was 0.612, which is less than a one-pixel difference, on average.
KW - Artificial intelligence
KW - biomedical optical imaging
KW - image segmentation
KW - neural network
KW - optical coherence tomography
UR - http://www.scopus.com/inward/record.url?scp=85072758442&partnerID=8YFLogxK
U2 - 10.1109/TIP.2019.2931461
DO - 10.1109/TIP.2019.2931461
M3 - Article
AN - SCOPUS:85072758442
VL - 29
SP - 303
EP - 312
JO - IEEE Transactions on Image Processing
JF - IEEE Transactions on Image Processing
SN - 1057-7149
M1 - 8784409
ER -