Radiomics-Guided Global-Local Transformer for Weakly Supervised Pathology Localization in Chest X-Rays

  title={Radiomics-Guided Global-Local Transformer for Weakly Supervised Pathology Localization in Chest X-Rays},
  author={Yan Han and Greg Holste and Ying Ding and Ahmed Tewfik and Yifan Peng and Zhangyang Wang},
  journal={IEEE transactions on medical imaging},
Before the recent success of deep learning methods for automated medical image analysis, practitioners used handcrafted radiomic features to quantitatively describe local patches of medical images. However, extracting discriminative radiomic features relies on accurate pathology localization, which is difficult to acquire in real-world settings. Despite advances in disease classification and localization from chest X-rays, many approaches fail to incorporate clinically-informed domain-specific… 

Figures and Tables from this paper



Anatomy-Guided Weakly-Supervised Abnormality Localization in Chest X-rays

An Anatomy-Guided chest X-ray Network (AGXNet) is proposed that consists of a cascade of two networks, one responsible for identifying anatomical abnormalities and the second responsible for pathological observations, and uses Positive Unlabeled (PU) learning to account for the fact that lack of mention does not necessarily mean a negative label.

Align, Attend and Locate: Chest X-Ray Diagnosis via Contrast Induced Attention Network With Limited Supervision

A Contrast Induced Attention Network (CIA-Net) is proposed, which exploits the highly structured property of chest X-ray images and localizes diseases via contrastive learning on the aligned positive and negative samples.

Pneumonia Detection On Chest X-Ray Using Radiomic Features And Contrastive Learning

This study proposed a novel framework that leverages radiomics features and contrastive learning to detect pneumonia in chest X-ray and achieves superior results to several state-of-the-art models and increases the model’s interpretability.

Knowledge Distillation with Adaptive Asymmetric Label Sharpening for Semi-supervised Fracture Detection in Chest X-rays

This work proposes a new knowledge distillation method that effectively exploits large-scale image-level labels extracted from the medical records, augmented with limited expert annotated region- level labels, to train a rib and clavicle fracture CAD model for chest X-ray (CXR), which leverages the teacher-student model paradigm and features a novel adaptive asymmetric label sharpening (AALS) algorithm.

Medical Transformer: Gated Axial-Attention for Medical Image Segmentation

A Gated Axial-Attention model is proposed which extends the existing architectures by introducing an additional control mechanism in the self-attention module and achieves better performance than the convolutional and other related transformer-based architectures.

Localization with Limited Annotation for Chest X-rays

A novel loss function is proposed for tackling the localization of thoracic diseases in chest X-ray images and a new architecture is proposed which accounts for both patch dependence and shift-invariance, through the inclusion of CRF layers and anti-aliasing filters, respectively.

Boosted Cascaded Convnets for Multilabel Classification of Thoracic Diseases in Chest Radiographs

This work experiments a set of deep learning models and presents a cascaded deep neural network that can diagnose all 14 pathologies better than the baseline and is competitive with other published methods.

TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation

It is argued that Transformers can serve as strong encoders for medical image segmentation tasks, with the combination of U-Net to enhance finer details by recovering localized spatial information, and empirical results suggest that the Transformer-based architecture presents a better way to leverage self-attention compared with previous CNN-based self-Attention methods.

CheXclusion: Fairness gaps in deep chest X-ray classifiers

It is demonstrated that TPR disparities exist in the state-of-the-art classifiers in all datasets, for all clinical tasks, and all subgroups, and that a multi-source dataset corresponds to the smallest disparities, suggesting one way to reduce bias.