An international challenge to use artificial intelligence to define the state of the art in kidney and kidney tumor segmentation in CT imaging

Heller, Nicholas; Sweeney, S.T. Mc; Peterson, Matthew; Peterson, Stephanie; Rickman, Jack; Stai, Bethany; Tejpaul, Resha; Oestreich, Makinna; Blake, Paul; Rosenberg, Joel; Moore, Keenan; Edward, Wong; Rengel, Zachary; Edgerton, Zach; Vasdev, Ranveer; Kalapara, Arveen; Sathianathen, Niranjan; Papanikolopoulos, N.; Weight, Christopher J.

doi:10.1016/s2666-1683(20)33076-7

Cited by 8 publications

(10 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Abdomen CT Dataset. We collected 663 abdomen CT scans from 6 public datasets, including BTCV [38], CHAOs [39], KiTS [40], Pancreas-CT, and two subsets (i.e., Pancreas and Spleen) of the Medical Segmentation Decathlon (MSD) challenge [41]. We extracted liver ROI of all the volumes with an off-the-shelf model based on nnU-Net [20], and our self-supervised training only uses these ROIs.…”

Section: A Pretraining Datasetsmentioning

confidence: 99%

Self-supervised Tumor Segmentation through Layer Decomposition

Zhang¹,

Xie²,

Huang³

et al. 2021

Preprint

View full text Add to dashboard Cite

In this paper, we propose a self-supervised approach for tumor segmentation. Specifically, we advocate a zero-shot setting, where models from self-supervised learning should be directly applicable for the downstream task, without using any manual annotations whatsoever. We make the following contributions. First, with careful examination on existing self-supervised learning approaches, we reveal the surprising result that, given suitable data augmentation, models trained from scratch in fact achieve comparable performance to those pre-trained with selfsupervised learning. Second, inspired by the fact that tumors tend to be characterized independently to the contexts, we propose a scalable pipeline for generating synthetic tumor data, and train a self-supervised model that minimises the generalisation gap with the downstream task. Third, we conduct extensive ablation studies on different downstream datasets, BraTS2018 for brain tumor segmentation and LiTS2017 for liver tumor segmentation. While evaluating the model transferability for tumor segmentation under a low-annotation regime, including an extreme case of zero-shot segmentation, the proposed approach demonstrates state-of-the-art performance, substantially outperforming all existing self-supervised approaches, and opening up the usage of self-supervised learning in practical scenarios.

show abstract

Section: A Pretraining Datasetsmentioning

confidence: 99%

Self-supervised Tumor Segmentation through Layer Decomposition

Zhang¹,

Xie²,

Huang³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The dataset used of FLARE2021 is adapted from MSD [4] (Liver [5], Spleen, Pancreas), NIH Pancreas [6,7,8], KiTS [9,10], and Nanjing University under the license permission. For more detail information of the dataset, please refer to the challenge website and [11].…”

Section: Datasetmentioning

confidence: 99%

Efficient Context-Aware Network for Abdominal Multi-organ Segmentation

Zhang¹,

Wang²

2021

Preprint

View full text Add to dashboard Cite

The contextual information, presented in abdominal CT scan, is relative consistent. In order to make full use of the overall 3D context, we develop a whole-volumebased coarse-to-fine framework for efficient and effective abdominal multi-organ segmentation. We propose a new efficientSegNet network, which is composed of encoder, decoder and context block. For the decoder module, anisotropic convolution with a k×k×1 intra-slice convolution and a 1×1×k inter-slice convolution, is designed to reduce the computation burden. For the context block, we propose strip pooling module to capture anisotropic and long-range contextual information, which exists in abdominal scene. Quantitative evaluation on the FLARE2021 validation cases, this method achieves the average dice similarity coefficient (DSC) of 0.895 and average normalized surface distance (NSD) of 0.775. The average running time is 9.8 s per case in inference phase, and maximum used GPU memory is 1017 MB.

show abstract

“…The dataset used in the FLARE 2021 Challenge is adapted from MSD [12] (Liver [13], Spleen, Pancreas), NIH Pancreas [14,15,16], KiTS [17,18], and Nanjing University under the license permission. For more detailed information on the dataset, please refer to the challenge website and [19].…”

Section: Datasetmentioning

confidence: 99%

Efficient Multi-Organ Segmentation Using SpatialConfiguration-Net with Low GPU Memory Requirements

Thaler¹,

Payer²,

Bischof³

et al. 2021

Preprint

View full text Add to dashboard Cite

Even though many semantic segmentation methods exist that are able to perform well on many medical datasets, often, they are not designed for direct use in clinical practice. The two main concerns are generalization to unseen data with a different visual appearance, e.g., images acquired using a different scanner, and efficiency in terms of computation time and required Graphics Processing Unit (GPU) memory. In this work, we employ a multi-organ segmentation model based on the SpatialConfiguration-Net (SCN), which integrates prior knowledge of the spatial configuration among the labelled organs to resolve spurious responses in the network outputs. Furthermore, we modified the architecture of the segmentation model to reduce its memory footprint as much as possible without drastically impacting the quality of the predictions. Lastly, we implemented a minimal inference script for which we optimized both, execution time and required GPU memory.

show abstract

An international challenge to use artificial intelligence to define the state of the art in kidney and kidney tumor segmentation in CT imaging

Cited by 8 publications

References 0 publications

Self-supervised Tumor Segmentation through Layer Decomposition

Self-supervised Tumor Segmentation through Layer Decomposition

Efficient Context-Aware Network for Abdominal Multi-organ Segmentation

Efficient Multi-Organ Segmentation Using SpatialConfiguration-Net with Low GPU Memory Requirements

Contact Info

Product

Resources

About