Preparing a collection of radiology examinations for distribution and retrieval
By Dina Demner-Fushman, M. Kohli, M. Rosenman, Sonya E. Shooshan, Laritza M. Rodriguez, Sameer Kiran Antani, George R. Thoma, Clement J. McDonald. 2015
1,302 citations recorded by the source corpus.
OBJECTIVE Clinical documents made available for secondary use play an increasingly important role in discovery of clinical knowledge, development of research methods, and education. An important step in facilitating secondary use of clinical document collections is easy access to descriptions and samples that represent the content of the collections. This paper presents an approach to developing a collection of radiology examinations, including both the images and radiologist narrative reports, and making them publicly available in a searchable database. MATERIALS AND METHODS The authors collected 3996 radiology reports from the Indiana Network for Patient Care and 8121 associated images from the hospitals' picture archiving systems. The images and reports were de-identified automatically and then the automatic de-identification was manually verified. The authors coded the key findings of the reports and empirically assessed the benefits of manual coding on retrieval. RESULTS The automatic de-identification of the narrative was aggressive and achieved 100% precision at the cost of rendering a few findings uninterpretable. Automatic de-identification of images was not quite as perfect. Images for two of 3996 patients (0.05%) showed protected health information. Manual encoding of findings improved retrieval precision. CONCLUSION Stringent de-identification methods can remove all identifiers from text radiology reports. DICOM de-identification of images does not remove all identifying information and needs special attention to images scanned from film. Adding manual coding to the radiologist narrative reports significantly improved relevancy of the retrieved clinical documents. The de-identified Indiana chest X-ray collection is available for searching and downloading from the National Library of Medicine (http://openi.nlm.nih.gov/).
Subfields: Computer Vision and Pattern Recognition, Pulmonary and Respiratory Medicine, Radiology, Nuclear Medicine and Imaging, 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 Computer Vision and Pattern Recognition
- Luc Van Gool
- Ming-Hsuan Yang
- Alan L. Yuille
- Dacheng Tao
- Xiaogang Wang
- Fahad Shahbaz Khan
- Trevor Darrell
- Bernt Schiele
In Pulmonary and Respiratory Medicine
- Matthew P. Lungren
- Ronald M. Summers
- Xiaosong Wang
- Marzyeh Ghassemi
- Pranav Rajpurkar
- Andrew Y. Ng
- Matthew B. A. McDermott
- Le Lu
In Radiology, Nuclear Medicine and Imaging
- Matthew P. Lungren
- Xiaosong Wang
- Ronald M. Summers
- Eric P. Xing
- Marzyeh Ghassemi
- Daguang Xu
- Baoyu Jing
- Zeya Wang
In Medical Imaging
- Matthew P. Lungren
- Xiaosong Wang
- Ronald M. Summers
- Eric P. Xing
- Curtis P. Langlotz
- Yuhao Zhang
- Yasuhide Miura
- Ahmed Tahseen Minhaz
Citations in the corpus
- Contrastive Learning of Medical Visual Representations from Paired Images and Text (2020, background): “Medical experts such as radiologists produce textual descriptions of images as part of their routine workflow, some of which are also made publicly available (Demner-Fushman et al., 2016; Johnson et al., 2019).”
- ChestX-Ray8: Hospital-Scale Chest X-Ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases (2017, methodology): “Using the OpenI API [1], we retrieve a total of 3,851 unique radiology reports where each OpenI report is assigned with its key findings/disease names by human annotators [9].”
- TieNet: Text-Image Embedding Network for Common Thorax Disease Classification and Reporting in Chest X-Rays (2018, methodology): “OpenI [7] is a publicly available radiography dataset collected from multiple institutes by Indiana University.”
- High-Throughput Classification of Radiographs Using Deep Convolutional Neural Networks (2016, methodology): “To serve as an independent test set, we obtained a publicly available set of anonymized chest radiographs in PNG format [17].”
- MIMIC-CXR, a de-identified publicly available database of chest radiographs with free-text reports (2019, background): “The Open-I Indiana University Chest X-ray dataset contains 8,121 images associated with 3,996 de-identified radiology reports 31 .”
- Abnormality Detection and Localization in Chest X-Rays using Deep Convolutional Neural Networks (2017, background): “Detection of car-diomegaly is done only for the frontal CXR images from the Indiana Dataset.”
- Bimodal Network Architectures for Automatic Generation of Image Annotation from Text (2018, methodology): “On the text side, the findings and diagnoses in radiology reports have been manually coded with the standard Medical Subject Headings (MeSH) terms by domain experts [1].”
- Translation between Molecules and Natural Language (2022, background): “, 2019), a close cousin of the chemistry domain, where tasks tend to be focused on diagnosis of various image types such as x-rays (Demner-Fushman et al., 2016).”
- NegBio: a high-performance tool for negation and uncertainty detection in radiology reports (2017, background): “One is a public benchmarking dataset, OpenI 23 .”
- PadChest: A large chest x-ray image dataset with multi-label annotated reports (2019, methodology): “They validated the image labeling procedure against 3,800 annotated reports of x-ray images from OpenI - Indiana DB- [16].”
- When Radiology Report Generation Meets Knowledge Graph (2020, methodology): “We evaluate our work using the publicly accessible IU-RR dataset (Demner-Fushman et al. 2015).”
- Self-supervised Image-text Pre-training With Mixed Data In Chest X-rays (2021, background): “OpenI-CXR [9] is a publicly shared chest X-ray dataset collected from multiple institutes by Indiana University.”
- Domain Adaptation for Medical Image Analysis: A Survey (2021, neutral): “[56] propose a two-step DA framework for ultrasound image classification.”
- CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison (2019, methodology): “The Indiana Network for Patient Care hosts the OpenI dataset (Demner-Fushman et al. 2015) consisting of 7,470 frontal-view radio-graphs and radiology reports which have been labeled with key findings by human annotators .”