Self-learning to detect and segment cysts in lung CT images without manual annotation

Image segmentation is a fundamental problem in medical image analysis. In recent years, deep neural networks achieve impressive performances on many medical image segmentation tasks by supervised learning on large manually annotated data. However, expert annotations on big medical datasets are tedious, expensive or sometimes unavailable. Weakly supervised learning could reduce the effort for annotation but still required certain amounts of expertise. Recently, deep learning shows a potential to produce more accurate predictions than the original erroneous labels. Inspired by this, we introduce a very weakly supervised learning method, for cystic lesion detection and segmentation in lung CT images, without any manual annotation. Our method works in a self-learning manner, where segmentation generated in previous steps (first by unsupervised segmentation then by neural networks) is used as ground truth for the next level of network learning. Experiments on a cystic lung lesion dataset show that the deep learning could perform better than the initial unsupervised annotation, and progressively improve itself after self-learning.

[1]  Geoffrey E. Hinton,et al.  Who Said What: Modeling Individual Labelers Improves Classification , 2017, AAAI.

[2]  Jianhua Yao,et al.  Sustained effects of sirolimus on lung function and cystic lung lesions in lymphangioleiomyomatosis. , 2014, American journal of respiratory and critical care medicine.

[3]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  E.E. Pissaloux,et al.  Image Processing , 1994, Proceedings. Second Euromicro Workshop on Parallel and Distributed Processing.

[5]  W. Marsden I and J , 2012 .

[6]  Aaas News,et al.  Book Reviews , 1893, Buffalo Medical and Surgical Journal.

[7]  Saining Xie,et al.  Holistically-Nested Edge Detection , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[8]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[9]  Vladimir Kolmogorov,et al.  An experimental comparison of min-cut/max- flow algorithms for energy minimization in vision , 2001, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[10]  Trevor Darrell,et al.  Fully convolutional networks for semantic segmentation , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[11]  Gebräuchliche Fertigarzneimittel,et al.  V , 1893, Therapielexikon Neurologie.

[12]  Jianping Yin,et al.  Cytoplasm and nucleus segmentation in cervical smear images using Radiating GVF Snake , 2012, Pattern Recognit..

[13]  Bernt Schiele,et al.  Simple Does It: Weakly Supervised Instance and Semantic Segmentation , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  Lin Yang,et al.  Suggestive Annotation: A Deep Active Learning Framework for Biomedical Image Segmentation , 2017, MICCAI.

[15]  Zhipeng Jia,et al.  Constrained Deep Weak Supervision for Histopathology Image Segmentation , 2017, IEEE Transactions on Medical Imaging.

[16]  Demis Hassabis,et al.  Mastering the game of Go without human knowledge , 2017, Nature.