Summary of Learning Low-rank Feature For Thorax Disease Classification, by Rajeev Goel et al.
Learning Low-Rank Feature for Thorax Disease Classification
by Rajeev Goel, Utkarsh Nath, Yancheng Wang, Alvin C. Silva, Teresa Wu, Yingzhen Yang
First submitted to arxiv on: 14 Feb 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper presents a novel approach to thorax disease classification on radiographic images, leveraging deep neural networks such as Convolutional Neural Networks (CNNs) and Visual Transformers (ViT). The authors highlight the importance of effectively extracting features for disease areas, as noise and background can significantly impact classification accuracy. To address this challenge, they propose Low-Rank Feature Learning (LRFL), a universally applicable method that reduces the adverse effect of non-disease areas on radiographic images. LRFL is motivated by the low frequency property observed in medical datasets and theoretically grounded by a sharp generalization bound for neural networks with low-rank features. The authors demonstrate the effectiveness of their approach using pre-trained ViT or CNN models, achieving better classification results in terms of multiclass area under the receiver operating curve (mAUC) and classification accuracy. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about improving how computers can identify diseases from X-ray images. Right now, these systems are not very good at ignoring things that aren’t relevant to the disease, like background noise or other parts of the image. The authors propose a new way to make these systems better by reducing the impact of non-disease areas on the image. This method is called Low-Rank Feature Learning (LRFL). It’s based on some interesting properties found in medical datasets and helps neural networks like Visual Transformers (ViT) or Convolutional Neural Networks (CNNs) classify diseases more accurately. |
Keywords
» Artificial intelligence » Classification » Cnn » Generalization » Vit
