ChestX-Ray8: Hospital-Scale Chest X-Ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases
By Xiaosong Wang, Yifan Peng, Le Lu, Zhiyong Lu, Mohammadhadi Bagheri, Ronald M. Summers. 2017
3,535 citations recorded by the source corpus.
The chest X-ray is one of the most commonly accessible radiological examinations for screening and diagnosis of many lung diseases. A tremendous number of X-ray imaging studies accompanied by radiological reports are accumulated and stored in many modern hospitals Picture Archiving and Communication Systems (PACS). On the other side, it is still an open question how this type of hospital-size knowledge database containing invaluable imaging informatics (i.e., loosely labeled) can be used to facilitate the data-hungry deep learning paradigms in building truly large-scale high precision computer-aided diagnosis (CAD) systems. In this paper, we present a new chest X-ray database, namely ChestX-ray8, which comprises 108,948 frontal-view X-ray images of 32,717 unique patients with the text-mined eight disease image labels (where each image can have multi-labels), from the associated radiological reports using natural language processing. Importantly, we demonstrate that these commonly occurring thoracic diseases can be detected and even spatially-located via a unified weakly-supervised multi-label image classification and disease localization framework, which is validated using our proposed dataset. Although the initial quantitative results are promising as reported, deep convolutional neural network based reading chest X-rays (i.e., recognizing and locating the common disease patterns trained with only image-level labels) remains a strenuous task for fully-automated high precision CAD systems.
Subfields: Pulmonary and Respiratory Medicine, Radiology, Nuclear Medicine and Imaging, Evaluation & Benchmarks, Medical Imaging
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Pulmonary and Respiratory Medicine
- Matthew P. Lungren
- Ronald M. Summers
- Rameswar Panda
- Quanfu Fan
- Chun-Fu Chen
- Xiaosong Wang
- Marzyeh Ghassemi
- Ahmed Hosny
In Radiology, Nuclear Medicine and Imaging
- Ronald M. Summers
- Matthew P. Lungren
- Xiaosong Wang
- Marzyeh Ghassemi
- Eric P. Xing
- Dinggang Shen
- Daniel Rueckert
- B. van Ginneken
In Evaluation & Benchmarks
- Chunhua Shen
- Alan L. Yuille
- Yi Yang
- Eric P. Xing
- Mohit Bansal
- Antonio Torralba
- Sanja Fidler
- Richard Socher
In Medical Imaging
- Matthew P. Lungren
- Xiaosong Wang
- Eric P. Xing
- Ronald M. Summers
- Mingxia Liu
- Curtis P. Langlotz
- Ahmed Hosny
- Hugo J. W. L. Aerts
Citations in the corpus
- Florence: A New Foundation Model for Computer Vision (2021, neutral): “…leaf images, 38 disease states over 14 plant species), EuroSAT (Helber et al., 2019) (RGB satellite images, 10 categories), ISIC 2018 (Codella et al., 2019; Tschandl et al., 2018) (dermoscopic images of skin lesions, 7 disease states), and ChestX (Wang et al., 2017) (Chest X-rays, 16 conditions).”
- Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: A cross-sectional study (2018, methodology): “Three datasets were obtained from different hospital groups: National Institutes of Health Clinical Center (NIH; 112,120 radiographs from 1992-2015), Indiana University Network for Patient Care (IU; 7,470 radiographs, date range not available), and Mount Sinai Hospital (MSH; 48,915 radiographs from ”
- What is being transferred in transfer learning? (2020, background): “In some cases there is nontrivial difference in visual forms between the source and the target domain [42, 37].”
- CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning (2017, methodology): “%) sets, following previous work on ChestX-ray14 (Wang et al., 2017; Yao et al., 2017).”
- Computer-aided detection in chest radiography based on artificial intelligence: a survey (2018, methodology): “Figure 1 shows the eight most common types of diseases observed in the chest radiograph [23], which are the disease of infiltration, atelectasis, cardiac hypertrophy, effusion, lumps, nodules, pneumonia, and pneumothorax, respectively.”
- Augmenting the National Institutes of Health Chest Radiograph Dataset with Expert Annotations of Possible Pneumonia. (2019, background): “Although machine learning algorithms can be trained with categorical labels (eg, “pneumonia” vs “no pneumonia”), accurate object-class detectors typically require a large set of images in which objects (eg, pneumonia) have been annotated manually with bounding boxes (3).”
- An Efficient Deep Learning Approach to Pneumonia Classification in Healthcare (2019, methodology): “Statistical dependency between labels [19] was studied to arrive at more precise predictions, thereby outperforming other techniques on given 13 images selected from 14 classes [20].”
- Domain Generalization using Causal Matching (2020, methodology): “We introduce a harder real-world dataset based on Chest X-ray images from three different sources: NIH (Wang et al., 2017), ChexPert (Irvin et al., 2019) and RSNA (rsn, 2018).”
- The Effect of Image Resolution on Deep Learning in Radiography. (2020, methodology): “In this study, when each selected label is considered separately, the AUCs have similar or higher performance compared with the multilabel DenseNet121 model in Rajpurkar et al (1) and other ChestX-ray14 deep learning approaches (3,4).”
- When Radiology Report Generation Meets Knowledge Graph (2020, methodology): “We also included Several previous works for comparison, i.e. ChestXray8 (Wang et al. 2017) and TieNet (Wang et al. 2018).”
- Deep learning with noisy labels: exploring techniques and remedies in medical image analysis (2019, background): “Another form of label imperfection that is becoming more common in medical image datasets is when there is only image-level label, and no pixel-level annotations are available [28], [23].”
- RegionViT: Regional-to-Local Attention for Vision Transformers (2021, methodology): “Our model outperforms DeiT (Touvron et al., 2020) by 2∼3% on ChestXRay8, which has a larger domain gap from ImageNet1K than other four datasets.”
- A New Benchmark for Evaluation of Cross-Domain Few-Shot Learning (2019, methodology): “The proposed benchmark includes data from the CropDiseases [23], EuroSAT [13], ISIC2018 [32, 5], and ChestX [35] datasets, which covers plant disease images, satellite images, dermoscopic images of skin lesions, and X-ray images, respectively.”
- COVID-CT-Dataset: A CT Scan Dataset about COVID-19 (2020, methodology): “Specifically, following (Rajpurkar et al., 2017) we use the ChestX-ray14 (Wang et al., 2017) dataset released by NIH to pretrain the DenseNet (Huang et al., 2017), then fine-tune the pre-trained DenseNet on the COVID-CT dataset.”