No New-Net
By Fabian Isensee, P. Kickingereder, Wolfgang Wick, Martin Bendszus, Klaus Hermann Maier-Hein. 2018
444 citations recorded by the source corpus.
In this paper we demonstrate the effectiveness of a well trained U-Net in the context of the BraTS 2018 challenge. This endeavour is particularly interesting given that researchers are currently besting each other with architectural modifications that are intended to improve the segmentation performance. We instead focus on the training process, argue that a well trained U-Net is hard to beat and intend to back up this assumption with a strong participation in this years BraTS challenge. Our baseline U-Net, which has only minor modifications and is trained with a large patch size and a dice loss function already achieves competitive dice scores on the BraTS2018 validation data. By incorporating region based training, additional training data and a simple postprocessing technique, we obtain dice scores of 81.01, 90.83 and 85.44 and Hausdorff Distances (95th percentile) of 2.54, 4.97 and 7.
Subfields: Artificial Intelligence, Biophysics, Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/4926fb5e06c3552b57305e95ec44715f832c3b28
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Maxim Berman
- Tom Eelbode
- Klaus Hermann Maier-Hein
- Dirk Vandermeulen
- Daniel S. Marcus
- Tal Arbel
- Frederik Maes
- Shaoting Zhang
In Biophysics
- Klaus Hermann Maier-Hein
- Shaoting Zhang
- Nematollah Kayhan Batmanghelich
- Guotai Wang
- Mauricio Reyes
- Martin Bendszus
- Alex Varghese
- Fabian Isensee
In Computer Vision and Pattern Recognition
- Maxim Berman
- Dirk Vandermeulen
- Frederik Maes
- Alberto Albiol
- Tom Eelbode
- Jeroen Bertels
- Raf Bisschops
- Paul F. Jaeger
Citations in the corpus
- Optimization for Medical Image Segmentation: Theory and Practice When Evaluating With Dice Score or Jaccard Index (2020, methodology): “More specifically, we start from the No NewNet [31] architecture, which was a top-ranked implementation during BRATS 2018.”
- nnU-Net for Brain Tumor Segmentation (2020, methodology): “omit the postprocessing presented in this manuscript. Just like many other BraTS participants, we have used the aggregated Dice scores and HD95 values for model selection in previous BraTS challenges [13,23]. As discussed in this manuscript, this strategy may not be ideal because the ranking that is”
- Deep learning based brain tumor segmentation: a survey (2020, background): “The vanilla structure of U-Net used for brain tumor segmentation is shown in Fig.”
- A Noise-Robust Framework for Automatic Segmentation of COVID-19 Pneumonia Lesions From CT Images (2020, methodology): “2D NN U-N ET [41] W ITHOUT U SING S ELF -”
- Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge (2018, background): “6, 18057 Rostock, Germany 19 Tata Memorial Centre, Homi Bhabha National Institute, Mumbai, India 20 Shri Guru Gobind Singhji Institute of Engineering and Technology, Nanded, India 21 NVIDIA, Santa Clara, USA 22 Division of Medical Image Computing, German Cancer Research Center (DKFZ), Heidelberg, Ge”