Abstract
BackgroundRenal cancer is one of the 10 most common cancers in human beings. The laparoscopic partial nephrectomy (LPN) is an effective way to treat renal cancer. Localization and delineation of the renal tumor from pre-operative CT Angiography (CTA) is an important step for LPN surgery planning. Recently, with the development of the technique of deep learning, deep neural networks can be trained to provide accurate pixel-wise renal tumor segmentation in CTA images. However, constructing the training dataset with a large amount of pixel-wise annotations is a time-consuming task for the radiologists. Therefore, weakly-supervised approaches attract more interest in research.MethodsIn this paper, we proposed a novel weakly-supervised convolutional neural network (CNN) for renal tumor segmentation. A three-stage framework was introduced to train the CNN with the weak annotations of renal tumors, i.e. the bounding boxes of renal tumors. The framework includes pseudo masks generation, group and weighted training phases. Clinical abdominal CT angiographic images of 200 patients were applied to perform the evaluation.ResultsExtensive experimental results show that the proposed method achieves a higher dice coefficient (DSC) of 0.826 than the other two existing weakly-supervised deep neural networks. Furthermore, the segmentation performance is close to the fully supervised deep CNN.ConclusionsThe proposed strategy improves not only the efficiency of network training but also the precision of the segmentation.
Highlights
Renal cancer is one of the 10 most common cancers in human beings
According to the weak annotations used for convolutional neural network (CNN) training, these approaches can be divided into four main categories: bounding box [3,4,5,6], scribble [7, 8], points [9, 10] and image-level labels [11,12,13,14,15,16,17]
DeepCut [18] adopted an iterative optimization method to train CNNs for brain and lung segmentation with the bounding-box labels which are determined by two corner coordinates, and the target object is inside the bounding box
Summary
Renal cancer is one of the 10 most common cancers in human beings. The laparoscopic partial nephrectomy (LPN) is an effective way to treat renal cancer. With the development of the technique of deep learning, deep neural networks can be trained to provide accurate pixel-wise renal tumor segmentation in CTA images. DeepCut [18] adopted an iterative optimization method to train CNNs for brain and lung segmentation with the bounding-box labels which are determined by two corner coordinates, and the target object is inside the bounding box In another weakly-supervised scenario [19], fetal brain MR images were segmented using a fully convolutional network (FCN) trained by superpixel annotations [20] which refer to an irregular region composed of adjacent pixels with similar texture, color, brightness or other features. To the best of our knowledge, there is no weakly-supervised deep learning technique reported for renal tumor segmentation
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
