Datasets:

mfromm
/

AMSR

{"forum": "H1gLLOgxeE", "submission_url": "https://openreview.net/forum?id=H1gLLOgxeE", "submission_content": {"title": "SPDA: Superpixel-based Data Augmentation for Biomedical Image Segmentation", "authors": ["Yizhe Zhang", "Lin Yang", "Hao Zheng", "Peixian Liang", "Colleen Mangold", "Raquel G. Loreto", "David P. Hughes", "Danny Z. Chen"], "authorids": ["yzhang29@nd.edu", "lyang5@nd.edu", "hzheng3@nd.edu", "pliang@nd.edu", "cav154@psu.edu", "raquelgloreto@gmail.com", "dhughes@psu.edu", "dchen@nd.edu"], "keywords": ["Superpixel", "Perception-preserving transformation", "Data augmentation", "Biomedical image segmentation"], "abstract": "In biomedical image segmentation, supervised training of a deep neural network aims to \"teach\" the network to mimic human visual perception that is represented by image-and-label pairs in the training data. Superpixelized (SP) images are visually perceivable to humans, but a conventionally trained deep learning model often performs poorly when working on SP images. To better mimic human visual perception, we think it is desirable for the deep learning model to be able to perceive not only raw images but also SP images. In this paper, we propose a new superpixel-based data augmentation (SPDA) method for training deep learning models for biomedical image segmentation. Our method applies a superpixel generation scheme to all the original training images to generate superpixelized images. The SP images thus obtained are then jointly used with the original training images to train a deep learning model. Our experiments of SPDA on four biomedical image datasets show that SPDA is effective and can consistently improve the performance of state-of-the-art fully convolutional networks for biomedical image segmentation in 2D and 3D images. Additional studies also demonstrate that SPDA can practically reduce the generalization gap.", "pdf": "/pdf/8e54830ff78c94b9ff1c6257e16e4a4c9a664aa3.pdf", "code of conduct": "I have read and accept the code of conduct.", "remove if rejected": "(optional) Remove submission if paper is rejected.", "paperhash": "zhang|spda_superpixelbased_data_augmentation_for_biomedical_image_segmentation", "_bibtex": "@inproceedings{zhang:MIDLFull2019a,\ntitle={{\\{}SPDA{\\}}: Superpixel-based Data Augmentation for Biomedical Image Segmentation},\nauthor={Zhang, Yizhe and Yang, Lin and Zheng, Hao and Liang, Peixian and Mangold, Colleen and Loreto, Raquel G. and Hughes, David P. and Chen, Danny Z.},\nbooktitle={International Conference on Medical Imaging with Deep Learning -- Full Paper Track},\naddress={London, United Kingdom},\nyear={2019},\nmonth={08--10 Jul},\nurl={https://openreview.net/forum?id=H1gLLOgxeE},\nabstract={In biomedical image segmentation, supervised training of a deep neural network aims to ''teach'' the network to mimic human visual perception that is represented by image-and-label pairs in the training data. Superpixelized (SP) images are visually perceivable to humans, but a conventionally trained deep learning model often performs poorly when working on SP images. To better mimic human visual perception, we think it is desirable for the deep learning model to be able to perceive not only raw images but also SP images. In this paper, we propose a new superpixel-based data augmentation (SPDA) method for training deep learning models for biomedical image segmentation. Our method applies a superpixel generation scheme to all the original training images to generate superpixelized images. The SP images thus obtained are then jointly used with the original training images to train a deep learning model. Our experiments of SPDA on four biomedical image datasets show that SPDA is effective and can consistently improve the performance of state-of-the-art fully convolutional networks for biomedical image segmentation in 2D and 3D images. Additional studies also demonstrate that SPDA can practically reduce the generalization gap.},\n}"}, "submission_cdate": 1544714318417, "submission_tcdate": 1544714318417, "submission_tmdate": 1561398656753, "submission_ddate": null, "review_id": ["Bke77eclVN", "rklhgDV8mV", "SkgdRYICXN"], "review_url": ["https://openreview.net/forum?id=H1gLLOgxeE&noteId=Bke77eclVN", "https://openreview.net/forum?id=H1gLLOgxeE&noteId=rklhgDV8mV", "https://openreview.net/forum?id=H1gLLOgxeE&noteId=SkgdRYICXN"], "review_cdate": [1548947482628, 1548269299808, 1548802511698], "review_tcdate": [1548947482628, 1548269299808, 1548802511698], "review_tmdate": [1548947482628, 1548856723187, 1548856680824], "review_readers": [["everyone"], ["everyone"], ["everyone"]], "review_writers": [["MIDL.io/2019/Conference/Paper55/AnonReviewer3"], ["MIDL.io/2019/Conference/Paper55/AnonReviewer1"], ["MIDL.io/2019/Conference/Paper55/AnonReviewer2"]], "review_reply_count": [{"replyCount": 0}, {"replyCount": 0}, {"replyCount": 0}], "review_replyto": ["H1gLLOgxeE", "H1gLLOgxeE", "H1gLLOgxeE"], "review_content": [{"pros": "This paper proposes a novel data augmentation approach based on the\nsuperpixel representation of the image. In particular, the authors\ngenerate superpixel parcelation of the training images and add a term\nto the cost function that penalizes classifier (segmentor) errors when\napplied to the superpixelized image. The authors evaluate their\napproach on several biomedical image data sets and demonstrate robust\nimprovement in the segmentation accuracy. The paper offers an\ninteresting idea that others in the community might find useful to\nimprove the robustness of their models.\n", "cons": "The innovation is relatively minor.", "rating": "3: accept", "confidence": "2: The reviewer is fairly confident that the evaluation is correct"}, {"pros": "This paper proposed a new data augmentation method using superpixels (SPDA) for training deep learning models for biomedical image segmentation. \n\nThe proposed method can effectively improve the performance of deep learning models for biomedical image segmentation tasks.\nThe experimental results are detailed and solid.", "cons": "Some technical details are missing.", "rating": "3: accept", "confidence": "3: The reviewer is absolutely certain that the evaluation is correct and very familiar with the relevant literature"}, {"pros": "This paper proposed to use super pixel/voxel to do data augmentation. Authors conducted comprehensive experiments and evaluation to show the effectiveness of the method. ", "cons": "Technical novelty not very significant, but still it is a good study overall.", "rating": "3: accept", "confidence": "3: The reviewer is absolutely certain that the evaluation is correct and very familiar with the relevant literature"}], "comment_id": ["rklQke874N", "rkg_-gUQV4", "BkgRd18X4E"], "comment_cdate": [1549127642838, 1549127679663, 1549127541605], "comment_tcdate": [1549127642838, 1549127679663, 1549127541605], "comment_tmdate": [1555946042519, 1555946042301, 1555946042089], "comment_readers": [["everyone"], ["everyone"], ["everyone"]], "comment_writers": [["MIDL.io/2019/Conference/Paper55/Authors", "MIDL.io/2019/Conference"], ["MIDL.io/2019/Conference/Paper55/Authors", "MIDL.io/2019/Conference"], ["MIDL.io/2019/Conference/Paper55/Authors", "MIDL.io/2019/Conference"]], "comment_reply_content": [{"replyCount": 0}, {"replyCount": 0}, {"replyCount": 0}], "comment_content": [{"title": "Thank you.", "comment": "We would like to thank reviewer #2 for his/her comments. "}, {"title": "Thank you.", "comment": "We would like to thank reviewer #3 for his/her comments. "}, {"title": "Thank you. More technical details will be added to the final version of the paper.", "comment": "We would like to thank reviewer #1 for his/her comments. We will add more technical details in the final version of the paper. Thank you."}], "comment_replyto": ["SkgdRYICXN", "Bke77eclVN", "rklhgDV8mV"], "comment_url": ["https://openreview.net/forum?id=H1gLLOgxeE&noteId=rklQke874N", "https://openreview.net/forum?id=H1gLLOgxeE&noteId=rkg_-gUQV4", "https://openreview.net/forum?id=H1gLLOgxeE&noteId=BkgRd18X4E"], "meta_review_cdate": 1551356590007, "meta_review_tcdate": 1551356590007, "meta_review_tmdate": 1551881976679, "meta_review_ddate ": null, "meta_review_title": "Acceptance Decision", "meta_review_metareview": "The authors explore a method to augment the training dataset with superpixel-based representations of the input data. The method is well-explained and the set of experiments is properly motivated in the text, and well designed to demonstrate the effectiveness of the proposed approach. While the level of innovation is relatively minor (superpixels have been around from some time now), the authors do compare the performance of adding the SPDA to different network designs, and the results seem promising.", "meta_review_readers": ["everyone"], "meta_review_writers": ["MIDL.io/2019/Conference"], "meta_review_reply_count": {"replyCount": 0}, "meta_review_url": ["https://openreview.net/forum?id=H1gLLOgxeE&noteId=HylU2fISLV"], "decision": "Accept"}