BRAX, Brazilian labeled chest x-ray dataset

BRAX, Brazilian labeled chest x-ray dataset

Chest radiographs allow for the meticulous examination of a patient’s chest but demands specialized training for proper interpretation. Automated analysis of medical imaging has become increasingly accessible with the advent of machine learning (ML) algorithms. Large labeled datasets are key element...

Full description

Saved in:
Bibliographic Details
Published in:Scientific data Vol. 9; no. 1; pp. 487 - 8
Main Authors: Reis, Eduardo P., de Paiva, Joselisa P. Q., da Silva, Maria C. B., Ribeiro, Guilherme A. S., Paiva, Victor F., Bulgarelli, Lucas, Lee, Henrique M. H., Santos, Paulo V., Brito, Vanessa M., Amaral, Lucas T. W., Beraldo, Gabriel L., Haidar Filho, Jorge N., Teles, Gustavo B. S., Szarf, Gilberto, Pollard, Tom, Johnson, Alistair E. W., Celi, Leo A., Amaro, Edson
Format: Journal Article
Language:English
Published: London Nature Publishing Group UK 10-08-2022
Nature Publishing Group
Nature Portfolio
Subjects:
692/53/2421
692/699/1785
692/700/1421/1770
Algorithms
Brazil
Chest
Data Descriptor
Datasets
Humanities and Social Sciences
Humans
Language
multidisciplinary
Natural Language Processing
Patients
Radiography
Radiography, Thoracic
Radiology
Science
Science (multidisciplinary)
X-Rays
Brazil
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Chest radiographs allow for the meticulous examination of a patient’s chest but demands specialized training for proper interpretation. Automated analysis of medical imaging has become increasingly accessible with the advent of machine learning (ML) algorithms. Large labeled datasets are key elements for training and validation of these ML solutions. In this paper we describe the Brazilian labeled chest x-ray dataset, BRAX: an automatically labeled dataset designed to assist researchers in the validation of ML models. The dataset contains 24,959 chest radiography studies from patients presenting to a large general Brazilian hospital. A total of 40,967 images are available in the BRAX dataset. All images have been verified by trained radiologists and de-identified to protect patient privacy. Fourteen labels were derived from free-text radiology reports written in Brazilian Portuguese using Natural Language Processing. Measurement(s) Chest Radiography Technology Type(s) natural language processing Factor Type(s) radiological findings/labels Sample Characteristic - Organism Homo sapiens Sample Characteristic - Environment chest organ Sample Characteristic - Location Brazil
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Undefined-1
ObjectType-Feature-3
content type line 23
ISSN:2052-4463
2052-4463
DOI:10.1038/s41597-022-01608-8