Thyroid ultrasound diagnosis improvement via multi-view self-supervised learning and two-stage pre-training

• URL: http://arxiv.org/abs/2402.11497v1
• Date: Sun, 18 Feb 2024 07:56:29 GMT
• Title: Thyroid ultrasound diagnosis improvement via multi-view self-supervised learning and two-stage pre-training
• Authors: Jian Wang, Xin Yang, Xiaohong Jia, Wufeng Xue, Rusi Chen, Yanlin Chen, Xiliang Zhu, Lian Liu, Yan Cao, Jianqiao Zhou, Dong Ni, Ning Gu
• Abstract summary: We proposed a multi-view contrastive self-supervised method to improve thyroid nodule classification and segmentation performance with limited manual labels. Our method aligns the transverse and longitudinal views of the same nodule, thereby enabling the model to focus more on the nodule area. We also adopted a two-stage pre-training to exploit the pre-training on ImageNet and thyroid ultrasound images.
• Score: 14.852699885616218
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Thyroid nodule classification and segmentation in ultrasound images are crucial for computer-aided diagnosis; however, they face limitations owing to insufficient labeled data. In this study, we proposed a multi-view contrastive self-supervised method to improve thyroid nodule classification and segmentation performance with limited manual labels. Our method aligns the transverse and longitudinal views of the same nodule, thereby enabling the model to focus more on the nodule area. We designed an adaptive loss function that eliminates the limitations of the paired data. Additionally, we adopted a two-stage pre-training to exploit the pre-training on ImageNet and thyroid ultrasound images. Extensive experiments were conducted on a large-scale dataset collected from multiple centers. The results showed that the proposed method significantly improves nodule classification and segmentation performance with limited manual labels and outperforms state-of-the-art self-supervised methods. The two-stage pre-training also significantly exceeded ImageNet pre-training.

Related papers

• Real-time guidewire tracking and segmentation in intraoperative x-ray [52.51797358201872]
  We propose a two-stage deep learning framework for real-time guidewire segmentation and tracking. In the first stage, a Yolov5 detector is trained, using the original X-ray images as well as synthetic ones, to output the bounding boxes of possible target guidewires. In the second stage, a novel and efficient network is proposed to segment the guidewire in each detected bounding box.
  arXiv  Detail & Related papers  (2024-04-12T20:39:19Z)
• Ultrasound Image Segmentation of Thyroid Nodule via Latent Semantic Feature Co-Registration [12.211161441766532]
  The present paper proposes ASTN, a framework for thyroid nodule segmentation achieved through a new type co-registration network. By extracting latent semantic information from the atlas and target images, this framework can ensure the integrity of anatomical structure. This paper also provides an atlas selection algorithm to mitigate the difficulty of co-registration.
  arXiv  Detail & Related papers  (2023-10-13T16:18:48Z)
• DINO-CXR: A self supervised method based on vision transformer for chest X-ray classification [0.9883261192383611]
  We propose a self-supervised method, DINO-CXR, which is a novel adaptation of a self-supervised method, DINO, based on a vision transformer for chest X-ray classification. A comparative analysis is performed to show the effectiveness of the proposed method for both pneumonia and COVID-19 detection.
  arXiv  Detail & Related papers  (2023-08-01T11:58:49Z)
• Less is More: Adaptive Curriculum Learning for Thyroid Nodule Diagnosis [50.231954872304314]
  We propose an Adaptive Curriculum Learning framework, which adaptively discovers and discards the samples with inconsistent labels. We also contribute TNCD: a Thyroid Nodule Classification dataset.
  arXiv  Detail & Related papers  (2022-07-02T11:50:02Z)
• Harmonizing Pathological and Normal Pixels for Pseudo-healthy Synthesis [68.5287824124996]
  We present a new type of discriminator, the segmentor, to accurately locate the lesions and improve the visual quality of pseudo-healthy images. We apply the generated images into medical image enhancement and utilize the enhanced results to cope with the low contrast problem. Comprehensive experiments on the T2 modality of BraTS demonstrate that the proposed method substantially outperforms the state-of-the-art methods.
  arXiv  Detail & Related papers  (2022-03-29T08:41:17Z)
• A New Semi-supervised Learning Benchmark for Classifying View and Diagnosing Aortic Stenosis from Echocardiograms [4.956777496509955]
  We develop a benchmark dataset to assess semi-supervised approaches to two tasks relevant to cardiac ultrasound (echocardiogram) interpretation. We find that a state-of-the-art method called MixMatch achieves promising gains in heldout accuracy on both tasks. We pursue patient-level diagnosis prediction, which requires aggregating across hundreds of images of diverse view types.
  arXiv  Detail & Related papers  (2021-07-30T21:08:12Z)
• Towards Unbiased COVID-19 Lesion Localisation and Segmentation via Weakly Supervised Learning [66.36706284671291]
  We propose a data-driven framework supervised by only image-level labels to support unbiased lesion localisation. The framework can explicitly separate potential lesions from original images, with the help of a generative adversarial network and a lesion-specific decoder.
  arXiv  Detail & Related papers  (2021-03-01T06:05:49Z)
• Improved Slice-wise Tumour Detection in Brain MRIs by Computing Dissimilarities between Latent Representations [68.8204255655161]
  Anomaly detection for Magnetic Resonance Images (MRIs) can be solved with unsupervised methods. We have proposed a slice-wise semi-supervised method for tumour detection based on the computation of a dissimilarity function in the latent space of a Variational AutoEncoder. We show that by training the models on higher resolution images and by improving the quality of the reconstructions, we obtain results which are comparable with different baselines.
  arXiv  Detail & Related papers  (2020-07-24T14:02:09Z)
• Multi-label Thoracic Disease Image Classification with Cross-Attention Networks [65.37531731899837]
  We propose a novel scheme of Cross-Attention Networks (CAN) for automated thoracic disease classification from chest x-ray images. We also design a new loss function that beyond cross-entropy loss to help cross-attention process and is able to overcome the imbalance between classes and easy-dominated samples within each class.
  arXiv  Detail & Related papers  (2020-07-21T14:37:00Z)
• 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.