Augmenting the National Institutes of Health Chest Radiograph Dataset with Expert Annotations of Possible Pneumonia.

By G. Shih, Carol C. Wu, S. Halabi, M. Kohli, L. Prevedello, T. Cook, Arjun Sharma, J. Amorosa, V. Arteaga, M. Galperin-Aizenberg and 9 more. 2019

412 citations recorded by the source corpus.

This dataset is intended to be used for machine learning and is composed of annotations with bounding boxes for pulmonary opacity on chest radiographs which may represent pneumonia in the appropriate clinical setting.

Subfields: Computer Vision and Pattern Recognition, Pulmonary and Respiratory Medicine, Radiology, Nuclear Medicine and Imaging, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Pulmonary and Respiratory Medicine

In Radiology, Nuclear Medicine and Imaging

In Evaluation & Benchmarks

Citations in the corpus

Contrastive Learning of Medical Visual Representations from Paired Images and Text (2020, methodology): “We evaluate our pretrained image encoders on four representative medical image classiﬁcation tasks: 1) RSNA Pneumonia Detection (Wang et al., 2017; Shih et al., 2019), which involves binary classiﬁcation of a chest radiograph image into either a pneumonia or a normal category; 2) CheXpert image…”
MedCLIP: Contrastive Learning from Unpaired Medical Images and Text (2022, positive): “RSNA Pneumonia (Shih et al., 2019) is a collection of pneumonia cases found in the database of chest x-rays made public by the National Institutes of Health.”
Assessing the (Un)Trustworthiness of Saliency Maps for Localizing Abnormalities in Medical Imaging (2020, methodology): “In this study, we comprehensively evaluate popular saliency maps for CNNs trained on the SIIM-ACR Pneumothorax Segmentation and RSNA Pneumonia Detection datasets[11, 37, 12, 36] in terms of 4 key criteria for trustworthiness: 1.”
LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day (2023, background): “Domain-specific pretraining has been shown to be effective for biomedical natural language processing (NLP) applications [18, 15, 11, 31] and biomedical vision-language (VL) tasks [16, 7, 41, 51, 9].”
BiomedCLIP: a multimodal biomedical foundation model pretrained from fifteen million scientific image-text pairs (2023, background): “Surpris-ingly, BiomedCLIP even outperformed radiology-specific state-of-the-art models such as BioViL [18] on radiology-specific tasks such as RSNA pneumonia detection [19], highlighting the potential of positive transfer in pretraining from diverse image categories.”
Making the Most of Text Semantics to Improve Biomedical Vision-Language Processing (2022, methodology): “…evaluation, we use RadNLI [54] to assess the proposed CXR-BERT text model in isolation, the new MS-CXR assesses joint image–text understanding via phrase grounding, and the RSNA Pneumonia dataset [66,76] to test zero-shot segmentation, as well as zero-shot and fine-tuned classification performance.”
Pneumonia Detection On Chest X-Ray Using Radiomic Features And Contrastive Learning (2021, methodology): “For example, Tang et al., [16] introduces the task-oriented unsupervised adversarial network, which consists of a cyclic I2I translation framework for RSNA Pneumonia Detection Challenge and a pediatric pneumonia diagnosis dataset.”
Learning to Exploit Temporal Structure for Biomedical Vision-Language Processing (2023, methodology): “Table 4 reports results on zero-and few-shot pneumonia classiﬁcation from single images [60], where BioViL-T establishes a new SOTA compared to prior work [9,32].”
RadImageNet: An Open Radiologic Deep Learning Research Dataset for Effective Transfer Learning (2022, background): “The pneumonia dataset consists of 26 685 chest radiographs (26).”
GLoRIA: A Multimodal Global-Local Representation Learning Framework for Label-efficient Medical Image Recognition (2021, methodology): “We use CheXpert and the RSNA Pneumonia datasets [14] for supervised classification and the CheXpert 5x200 dataset for zero-shot classification.”
Joint Learning of Localized Representations from Medical Images and Reports (2021, methodology): “– RSNA Pneumonia Detection[81,69] (Licensed following the competition rules(13)): We download the dataset from its Kaggle page(14) but use only their training set which we randomly split into our training, validation, and test set resulting in 16010/5337/5337 training/validation/test samples, respec”