A Study of Domain Generalization on Ultrasound-based Multi-Class Segmentation of Arteries, Veins, Ligaments, and Nerves Using Transfer Learning

Identifying landmarks in the femoral area is crucial for ultrasound (US) -based robot-guided catheter insertion, and their presentation varies when imaged with different scanners. As such, the performance of past deep learning-based approaches is also narrowly limited to the training data distribution; this can be circumvented by fine-tuning all or part of the model, yet the effects of fine-tuning are seldom discussed. In this work, we study the US-based segmentation of multiple classes through transfer learning by fine-tuning different contiguous blocks within the model, and evaluating on a gamut of US data from different scanners and settings. We propose a simple method for predicting generalization on unseen datasets and observe statistically significant differences between the fine-tuning methods while working towards domain generalization.

[1]  Nassir Navab,et al.  Deep Residual Learning for Instrument Segmentation in Robotic Surgery , 2017, MLMI@MICCAI.

[2]  Tejas Sudharshan Mathai,et al.  Segmentation of Vessels in Ultra High Frequency Ultrasound Sequences Using Contextual Memory , 2019, MICCAI.

[3]  Alexey Shvets,et al.  TernausNet: U-Net with VGG11 Encoder Pre-Trained on ImageNet for Image Segmentation , 2018, Computer-Aided Analysis of Gastrointestinal Videos.

[4]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[5]  Jon Kleinberg,et al.  Transfusion: Understanding Transfer Learning for Medical Imaging , 2019, NeurIPS.

[6]  J. Ricke,et al.  A Randomized Trial Assessing the Value of Ultrasound-Guided Puncture of the Femoral Artery for Interventional Investigations , 2004, The International Journal of Cardiovascular Imaging.

[7]  Alvin I. Chen,et al.  Deep learning robotic guidance for autonomous vascular access , 2020, Nat. Mach. Intell..

[8]  Vishal Patel,et al.  A framework for secure and decentralized sharing of medical imaging data via blockchain consensus , 2019, Health Informatics J..

[9]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[10]  Tejas Sudharshan Mathai,et al.  Fast Vessel Segmentation and Tracking in Ultra High-Frequency Ultrasound Images , 2018, MICCAI.

[11]  Matthias Bethge,et al.  ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness , 2018, ICLR.

[12]  Senthil Mani,et al.  Coverage Testing of Deep Learning Models using Dataset Characterization , 2019, ArXiv.

[13]  Rupert Brooks,et al.  Fine Tuning U-Net for Ultrasound Image Segmentation: Which Layers? , 2019, DART/MIL3ID@MICCAI.

[14]  Gang Hua,et al.  Learning Discriminative Reconstructions for Unsupervised Outlier Removal , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).