MedNeXt: Transformer-driven Scaling of ConvNets for Medical Image Segmentation

By Saikat Roy, Gregor Koehler, Constantin Ulrich, Michael Baumgartner, Jens Petersen, Fabian Isensee, Paul F. Jaeger, Klaus Hermann Maier-Hein. International Conference on Medical Image Computing and Computer-Assisted Intervention, 2023

419 citations recorded by the source corpus.

There has been exploding interest in embracing Transformer-based architectures for medical image segmentation. However, the lack of large-scale annotated medical datasets make achieving performances equivalent to those in natural images challenging. Convolutional networks, in contrast, have higher inductive biases and consequently, are easily trainable to high performance. Recently, the ConvNeXt architecture attempted to modernize the standard ConvNet by mirroring Transformer blocks. In this work, we improve upon this to design a modernized and scalable convolutional architecture customized to challenges of data-scarce medical settings. We introduce MedNeXt, a Transformer-inspired large kernel segmentation network which introduces - 1) A fully ConvNeXt 3D Encoder-Decoder Network for medical image segmentation, 2) Residual ConvNeXt up and downsampling blocks to preserve semantic richness across scales, 3) A novel technique to iteratively increase kernel sizes by upsampling small kernel networks, to prevent performance saturation on limited medical data, 4) Compound scaling at multiple levels (depth, width, kernel size) of MedNeXt. This leads to state-of-the-art performance on 4 tasks on CT and MRI modalities and varying dataset sizes, representing a modernized deep architecture for medical image segmentation. Our code is made publicly available at: https://github.com/MIC-DKFZ/MedNeXt.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/b90a84f364d97c7ca9e94a1e1384fb4b280a8de9

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

1. Fahad Shahbaz Khan
2. Salman Hameed Khan
3. Ming-Hsuan Yang
4. Abdelrahman M. Shaker
5. Klaus Hermann Maier-Hein
6. Hanoona Abdul Rasheed
7. Saikat Roy
8. Muhammad Maaz

In Computer Vision and Pattern Recognition

1. Fahad Shahbaz Khan
2. Ming-Hsuan Yang
3. Salman Hameed Khan
4. Luc Van Gool
5. Martin Danelljan
6. Philip H. S. Torr
7. Dacheng Tao
8. Radu Timofte

In Radiology, Nuclear Medicine and Imaging

1. Fabian Isensee
2. Klaus Hermann Maier-Hein
3. Salman Hameed Khan
4. Ming-Hsuan Yang
5. Fahad Shahbaz Khan

Citations in the corpus

• nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentation (2024, methodology): “Despite this, numerous studies claim methodological superiority while at least 50% of the benchmark is made up by either BTCV [37,11,9] or BraTS [14,34,31,28,16].”
• SegMamba: Long-range Sequential Modeling Mamba For 3D Medical Image Segmentation (2024, methodology): “[12] 45.73 49.73 MedNeXt [21] 35.93 52.54 UNETR [7] 33.70 61.51 SwinUNETR [6] 38.36 55.05 SwinUNETR-V2 [8] 41.”
• UNETR++: Delving Into Efficient and Accurate 3D Medical Image Segmentation (2022, methodology): “In this section, we provide additional details regarding: • Implementation Details (Appendix A) • Qualitative Results (Appendix B) • Ablations (Appendix C) • Discussion (Appendix D)”