Download PDFOpen PDF in browser

Transformer vs. CNN – A Comparison on Knee Segmentation in Ultrasound Images

6 pagesPublished: December 13, 2022

Abstract

The automated and robust segmentation of bone surfaces in ultrasound (US) images can open up new fields of application for US imaging in computer-assisted orthopedic surgery, e.g. for the patient-specific planning process in computer-assisted knee replacement. For the automated, deep learning-based segmentation of medical images, CNN-based methods have been the state of the art over the last years, while recently Transformer-based methods are on the rise in computer vision. To compare these methods with respect to US image segmentation, in this paper the recent Transformer- based Swin-UNet is exemplarily benchmarked against the commonly used CNN-based nnUNet on the application of in-vivo 2D US knee segmentation.
Trained and tested on our own dataset with 8166 annotated images (split in 7155 and 1011 images respectively), both the nnUNet and the pre-trained Swin-UNet show a Dice coefficient of 0.78 during testing. For distances between skeletonized labels and predictions, a symmetric Hausdorff distance of 44.69 pixels and a symmetric surface distance of 5.77 pixels is found for nnUNet as compared to 42.78 pixels and 5.68 pixels respectively for the Swin-UNet. Based on qualitative assessment, the Transformer-based Swin-UNet appears to benefit from its capability of learning global relationships as compared to the CNN-based nnUNet, while the latter shows more consistent and smooth predictions on a local level, presumably due to the character of convolution operation. Besides, the Swin-UNet requires generalized pre-training to be competitive.
Since both architectures are evenly suited for the task at hand, for our future work, hybrid architectures combining the characteristic advantages of Transformer-based and CNN-based methods seem promising for US image segmentation.

Keyphrases: CNN, image segmentation, machine learning, Transfomer, Ultrasound imaging

In: Ferdinando Rodriguez Y Baena, Joshua W Giles and Eric Stindel (editors). Proceedings of The 20th Annual Meeting of the International Society for Computer Assisted Orthopaedic Surgery, vol 5, pages 31--36

Links:
BibTeX entry
@inproceedings{CAOS2022:Transformer_vs._CNN,
  author    = {Peter Br\textbackslash{}"o\{\textbackslash{}ss\}ner and Benjamin Hohlmann and Klaus Radermacher},
  title     = {Transformer vs. CNN -- A Comparison on Knee Segmentation in Ultrasound Images},
  booktitle = {Proceedings of The 20th Annual Meeting of the International Society for Computer Assisted Orthopaedic Surgery},
  editor    = {Ferdinando Rodriguez Y Baena and Joshua W Giles and Eric Stindel},
  series    = {EPiC Series in Health Sciences},
  volume    = {5},
  pages     = {31--36},
  year      = {2022},
  publisher = {EasyChair},
  bibsource = {EasyChair, https://easychair.org},
  issn      = {2398-5305},
  url       = {https://easychair.org/publications/paper/Qkqd},
  doi       = {10.29007/cqcv}}
Download PDFOpen PDF in browser