Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound

@article{He2021StatisticalDG,
  title={Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound},
  author={Shuangchi He and Zehui Lin and Xin Yang and Chaoyu Chen and Jian Wang and Xue Shuang and Ziwei Deng and Qin Liu and Yan Cao and Xiduo Lu and Ruobing Huang and Nishant Ravikumar and Alejandro F. Frangi and Yuanji Zhang and Yi Xiong and Dong Ni},
  journal={ArXiv},
  year={2021},
  volume={abs/2108.05055}
}
. Standard plane recognition plays an important role in prenatal ultrasound (US) screening. Automatically recognizing the standard plane along with the corresponding anatomical structures in US image can not only facilitate US image interpretation but also improve diagnos-tic efficiency. In this study, we build a novel multi-label learning (MLL) scheme to identify multiple standard planes and corresponding anatomical structures of fetus simultaneously. Our contribution is three-fold. First, we… 

References

SHOWING 1-10 OF 15 REFERENCES
Multi-task learning for quality assessment of fetal head ultrasound images
Ultrasound Standard Plane Detection Using a Composite Neural Network Framework
TLDR
A general framework for the automatic identification of different standard planes from US videos is presented and explores in- and between-plane feature learning with a novel composite framework of the convolutional and recurrent neural networks.
Evaluation of deep convolutional neural networks for automatic classification of common maternal fetal ultrasound planes
TLDR
Results indicate for the first time that computational models have similar performance compared to humans when classifying common planes in human fetal examination, however, the dataset leaves the door open on future research to further improve results, especially on fine-grained plane categorization.
A Generic Quality Control Framework for Fetal Ultrasound Cardiac Four-Chamber Planes
TLDR
A generic deep learning framework for automatic quality control of fetal US cardiac four-chamber plane, which consists of a basic CNN, a deeper CNN, and the aggregated residual visual block net (ARVBNet), detecting the key anatomical structures on a plane is proposed.
SonoEyeNet: Standardized fetal ultrasound plane detection informed by eye tracking
We present a novel automated approach for detection of standardized abdominal circumference (AC) planes in fetal ultrasound built in a convolutional neural network (CNN) framework, called SonoEyeNet,
Learning Spatial Regularization with Image-Level Supervisions for Multi-label Image Classification
TLDR
Analysis of the learned SRN model demonstrates that it can effectively capture both semantic and spatial relations of labels for improving classification performance, and significantly outperforms state-of-the-arts and has strong generalization capability.
Multi-Label Image Recognition With Graph Convolutional Networks
TLDR
This work proposes a multi-label classification model based on Graph Convolutional Network (GCN), and proposes a novel re-weighted scheme to create an effective label correlation matrix to guide information propagation among the nodes in GCN.
CNN-RNN: A Unified Framework for Multi-label Image Classification
TLDR
The proposed CNN-RNN framework learns a joint image-label embedding to characterize the semantic label dependency as well as the image- label relevance, and it can be trained end-to-end from scratch to integrate both information in a unified framework.
Supervised Contrastive Learning
TLDR
A novel training methodology that consistently outperforms cross entropy on supervised learning tasks across different architectures and data augmentations is proposed, and the batch contrastive loss is modified, which has recently been shown to be very effective at learning powerful representations in the self-supervised setting.
Deep Residual Learning for Image Recognition
TLDR
This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.
...
...