LOTUS: Learning to Optimize Task-based US representations

• URL: http://arxiv.org/abs/2307.16021v1
• Date: Sat, 29 Jul 2023 16:29:39 GMT
• Title: LOTUS: Learning to Optimize Task-based US representations
• Authors: Yordanka Velikova, Mohammad Farid Azampour, Walter Simson, Vanessa Gonzalez Duque, Nassir Navab
• Abstract summary: Anatomical segmentation of organs in ultrasound images is essential to many clinical applications. Existing deep neural networks require a large amount of labeled data for training in order to achieve clinically acceptable performance. In this paper, we propose a novel approach for learning to optimize task-based ultra-sound image representations.
• Score: 39.81131738128329
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Anatomical segmentation of organs in ultrasound images is essential to many clinical applications, particularly for diagnosis and monitoring. Existing deep neural networks require a large amount of labeled data for training in order to achieve clinically acceptable performance. Yet, in ultrasound, due to characteristic properties such as speckle and clutter, it is challenging to obtain accurate segmentation boundaries, and precise pixel-wise labeling of images is highly dependent on the expertise of physicians. In contrast, CT scans have higher resolution and improved contrast, easing organ identification. In this paper, we propose a novel approach for learning to optimize task-based ultra-sound image representations. Given annotated CT segmentation maps as a simulation medium, we model acoustic propagation through tissue via ray-casting to generate ultrasound training data. Our ultrasound simulator is fully differentiable and learns to optimize the parameters for generating physics-based ultrasound images guided by the downstream segmentation task. In addition, we train an image adaptation network between real and simulated images to achieve simultaneous image synthesis and automatic segmentation on US images in an end-to-end training setting. The proposed method is evaluated on aorta and vessel segmentation tasks and shows promising quantitative results. Furthermore, we also conduct qualitative results of optimized image representations on other organs.

Related papers

• S-CycleGAN: Semantic Segmentation Enhanced CT-Ultrasound Image-to-Image Translation for Robotic Ultrasonography [2.07180164747172]
  We introduce an advanced deep learning model, dubbed S-CycleGAN, which generates high-quality synthetic ultrasound images from computed tomography (CT) data. The synthetic images are utilized to enhance various aspects of our development of the robot-assisted ultrasound scanning system.
  arXiv  Detail & Related papers  (2024-06-03T10:53:45Z)
• CathFlow: Self-Supervised Segmentation of Catheters in Interventional Ultrasound Using Optical Flow and Transformers [66.15847237150909]
  We introduce a self-supervised deep learning architecture to segment catheters in longitudinal ultrasound images. The network architecture builds upon AiAReSeg, a segmentation transformer built with the Attention in Attention mechanism. We validated our model on a test dataset, consisting of unseen synthetic data and images collected from silicon aorta phantoms.
  arXiv  Detail & Related papers  (2024-03-21T15:13:36Z)
• Cardiac ultrasound simulation for autonomous ultrasound navigation [4.036497185262817]
  We propose a method to generate large amounts of ultrasound images from other modalities and from arbitrary positions. We present a novel simulation pipeline which uses segmentations from other modalities, an optimized data representation and GPU-accelerated Monte Carlo path tracing. The proposed approach allows for fast and accurate patient-specific ultrasound image generation, and its usability for training networks for navigation-related tasks is demonstrated.
  arXiv  Detail & Related papers  (2024-02-09T15:14:48Z)
• Voice-assisted Image Labelling for Endoscopic Ultrasound Classification using Neural Networks [48.732863591145964]
  We propose a multi-modal convolutional neural network architecture that labels endoscopic ultrasound (EUS) images from raw verbal comments provided by a clinician during the procedure. Our results show a prediction accuracy of 76% at image level on a dataset with 5 different labels.
  arXiv  Detail & Related papers  (2021-10-12T21:22:24Z)
• Deep Learning for Ultrasound Beamforming [120.12255978513912]
  Beamforming, the process of mapping received ultrasound echoes to the spatial image domain, lies at the heart of the ultrasound image formation chain. Modern ultrasound imaging leans heavily on innovations in powerful digital receive channel processing. Deep learning methods can play a compelling role in the digital beamforming pipeline.
  arXiv  Detail & Related papers  (2021-09-23T15:15:21Z)
• Semantic segmentation of multispectral photoacoustic images using deep learning [53.65837038435433]
  Photoacoustic imaging has the potential to revolutionise healthcare. Clinical translation of the technology requires conversion of the high-dimensional acquired data into clinically relevant and interpretable information. We present a deep learning-based approach to semantic segmentation of multispectral photoacoustic images.
  arXiv  Detail & Related papers  (2021-05-20T09:33:55Z)
• Learning Ultrasound Rendering from Cross-Sectional Model Slices for Simulated Training [13.640630434743837]
  Computational simulations can facilitate the training of such skills in virtual reality. We propose herein to bypass any rendering and simulation process at interactive time. We use a generative adversarial framework with a dedicated generator architecture and input feeding scheme.
  arXiv  Detail & Related papers  (2021-01-20T21:58:19Z)
• Pathological Retinal Region Segmentation From OCT Images Using Geometric Relation Based Augmentation [84.7571086566595]
  We propose improvements over previous GAN-based medical image synthesis methods by jointly encoding the intrinsic relationship of geometry and shape. The proposed method outperforms state-of-the-art segmentation methods on the public RETOUCH dataset having images captured from different acquisition procedures.
  arXiv  Detail & Related papers  (2020-03-31T11:50:43Z)
• Breast lesion segmentation in ultrasound images with limited annotated data [2.905751301655124]
  We propose the use of simulated US images and natural images as auxiliary datasets in order to pre-train our segmentation network. We show that fine-tuning the pre-trained network improves the dice score by 21% compared to training from scratch.
  arXiv  Detail & Related papers  (2020-01-21T03:34:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.