AI Article Synopsis
- Chest radiographs provide detailed examination capabilities but require specialized training for accurate interpretation.
- The paper introduces the BRAX dataset, which consists of almost 25,000 chest x-ray studies from a major Brazilian hospital, aiding in the development and validation of machine learning models for medical imaging.
- The dataset includes over 40,000 images, all checked for accuracy by radiologists, and employs Natural Language Processing to extract fourteen labels from radiology reports in Brazilian Portuguese.
Article Abstract
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.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9364309 | PMC |
| http://dx.doi.org/10.1038/s41597-022-01608-8 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Strangers in a strange land: how diversity professionals navigate their marginal leadership identity.
Front Psychol
December 2024
Fundação Getúlio Vargas, Rio de Janeiro, Rio de Janeiro, Brazil.
The purpose of the present study was to understand how executives responsible for Diversity and Inclusion construct their identities while occupying positions of "Marginal Leadership," and how they deal with threats to such identities. We conducted qualitative and inductive research with Brazilian executives in 66 organizations, focusing on their experiences as leaders. In our resulting model, we theorized that leaders in positions of Marginal Leadership demonstrate little role clarity, resources, and confidence for the exercise of their activities, suffering threats to their identities, which are not observed in executives occupying positions in more traditional functions such as Finance, Production, and Marketing.
View Article and Find Full Text PDFThe more fun, the more dangerous: The nutritional quality of foods with child-directed marketing.
Food Res Int
December 2024
Food Labeling Observatory, Nutrition and Food Service Research Center (CPPNAC), Federal University of São Paulo (UNIFESP), Santos, SP, Brazil. Electronic address:
Article Synopsis
- The study analyzed the nutritional quality of foods and beverages that employ child-directed marketing on their labels in Brazil, assessing data from over 8,900 products collected before new labeling laws were introduced.
- More than 10% of the products used child-targeted marketing, with over half having high sugar and fat content, indicating poor overall nutritional quality.
- The findings showed low agreement among different nutritional classification systems, with a recommendation for stricter regulations to limit marketing unhealthy foods to children.
Authentication of indigenous Brazilian specialty canephora coffees using smartphone image analysis.
Food Res Int
November 2024
Universidade Federal do Oeste da Bahia, Programa de Pós-Graduação em Química Pura e Aplicada, 47810-047 Barreiras, Bahia, Brazil. Electronic address:
The prevention of coffee fraud through the use of digital and intelligence-based technologies is an analytical challenge because depending on the adulterant, visual inspection is unreliable in roasted and ground coffee due to the similarity in color and texture of the materials used. In this work, a 3D-printed apparatus for smartphone image acquisiton is proposed. The digital images are used to authenticate the geographical origin of indigenous canephora coffees produced at Amazon region, Brazil, against canephora coffees from Espírito Santo, Brazil, and to capture the adulteration of indigenous samples.
View Article and Find Full Text PDFAlignment of countries in the Americas with the latest WHO guidelines for hepatitis B virus (HBV) infection: a review.
Lancet Reg Health Am
November 2024
Evandro Chagas National Institute of Infectious Diseases (INI), Oswaldo Cruz Foundation (FIOCRUZ), Rio de Janeiro, RJ, Brazil.
Evidence is lacking on alignment of current guidance from the Region of the Americas (AMR) countries with the new guidelines for people with hepatitis B virus (HBV) infection published by the World Health Organization (WHO) in March 2024. We gathered the most updated guidance on HBV infection from organisations/societies and seven countries from AMR. Most guidelines were aligned with the new WHO recommendation to treat persons with elevated ALT and HBV-DNA levels ≥2,000 IU/ml or with HIV-coinfection, hepatocellular carcinoma family history, extra-hepatic manifestations, or immunosuppression.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!