Deep learning-based models have been extensively used in computer vision and image analysis to automatically segment the region of interest (ROI) in an image. Optical coherence tomography (OCT) is used to obtain the images of the kidney’s proximal convoluted tubules (PCTs), which can be used to quantify the morphometric parameters such as tubular density and diameter. However, the large image dataset and patient movement during the scan made the pattern recognition and deep learning task to be difficult. Another challenge is a large number of non-ROIs compared to ROI pixels which caused data imbalanced and low network performance. This paper aims at developing a soft Attention-based UNET model for automatic segmentation of tubule lumen kidney images. Attention-UNET can extract features based on the ground truth structure and hence the irrelevant feature maps are not contributed during training. The performance of the soft-Attention-UNET is compared with standard UNET, Residual UNET (Res-UNET), and fully convolutional neural network (FCN). The original dataset contains 14403 OCT images from 169 transplant kidneys for training and testing. The results have shown that soft-Attention-UNET can achieve the dice score of 0.78±0.08 and intersection over union (IOU) of 0.83 which was as accurate as the manual segmentation results (dice score = 0.835±0.05) and the best segmentation scores among Res-UNET, regular UNET, and FCN networks. The results show that CLAHE contrast enhancement can improve the segmentation metrics of all models significantly (p < 0.05). Experimental results of this paper have proven that the soft Attention-based UNET is highly powerful for tubule lumen identification and localization and can improve clinical decision-making on a new transplant kidney as fast and accurately as possible.
Traditional super-resolution techniques are generally presented as optimization problems with variations in the choice of optimization methods and cost functions. Even for the overdetermined cases, the problem is ill-conditioned. The situation is worsened when considering underdetermined cases with unknown regions due to occlusions or lack of data. Deep learning-based methods have shown promise in solving a similar problem. One recent advancement has come in the form of partial convolutions, which were developed to perform infilling of holes in images. When used in an appropriate deep neural network, this particular variant of the convolutional filter has shown great promise in approximating missing spatial information. The method described is formulated as a two-stage process. Lower resolution images are first registered and placed on a high-resolution grid. The problem is then treated as an in-painting task where the missing regions are reconstructed using a deep neural network with partial convolutional filters. We compare our method against deep learning-based single image super-resolution methods and classical multi-image super-resolution techniques using two similarity metrics and show that our method is more robust to occlusions and errors in registration while also producing higher quality outputs.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.