Catalysing Artificial Intelligence for Paediatric Tuberculosis Research (CAPTURE): protocol for a global multicentre study establishing a paediatric chest X-ray repository to evaluate computer-aided detection algorithms
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Developing a large dataset of child tuberculosis (TB) chest radiographs (CXRs) and associated data will enable the evaluation and optimization of artificial intelligence (AI) computer-aided detection (CAD) tools for improved pediatric TB diagnosis.
Area Of Science
- Medical Imaging
- Artificial Intelligence
- Pediatric Tuberculosis
Background
- Childhood tuberculosis (TB) diagnosis faces challenges due to inadequate tools, with chest radiographs (CXRs) being crucial but often hampered by a lack of expert interpretation.
- Existing artificial intelligence (AI) computer-aided detection (CAD) software for CXR interpretation, primarily developed for adults, shows suboptimal performance in children due to differing TB presentation.
- The Catalysing Artificial Intelligence for Paediatric Tuberculosis Research (CAPTURE) initiative aims to address this gap by creating a specialized data repository.
Purpose Of The Study
- To establish a comprehensive repository (CAPTURE) of pediatric TB CXR images and associated clinical data.
- To evaluate the diagnostic performance of current adult-developed CAD products in children with presumptive TB.
- To facilitate the optimization and development of novel pediatric-specific CAD algorithms through data sharing.
Main Methods
- A repository was created by pooling approximately 11,000 CXRs from ~20 high-quality child TB diagnostic studies.
- CXRs and metadata, including consensus radiological interpretations and TB case classifications, were centrally collated.
- Existing CAD products will be evaluated against clinical, microbiological, and radiological reference standards, with a subset of images designated for training and validation of optimized algorithms.
Main Results
- The CAPTURE repository now houses a substantial collection of pediatric TB CXRs and associated data, with expert radiological interpretations.
- Initial evaluations will benchmark the performance of existing adult CAD algorithms against established reference standards.
- A training and validation dataset will be provided to developers for optimizing CAD tools for pediatric use.
Conclusions
- The CAPTURE initiative provides a vital resource for advancing AI-based diagnostic tools in pediatric TB.
- Optimizing CAD algorithms using pediatric-specific data is essential for improving diagnostic accuracy and addressing the case detection gap.
- Dissemination of findings and optimized tools aims to guide policy and improve global access to effective pediatric TB diagnostics.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
Related Concept Videos
Tuberculosis, more commonly referred to as TB, is an infectious disease stemming from Mycobacterium tuberculosis. While it primarily impacts the lungs, TB can also affect other body areas. Given its severity and global impact, timely and accurate diagnosis is crucial for controlling its spread and improving patient outcomes.
Several diagnostic approaches are used to detect TB. The conventional method is the Tuberculin Skin Test (TST), also known as the Mantoux test. However, this method has...
Radiological investigations, including X-rays and computed tomography (CT) scans, are critical for diagnosing and evaluating various medical conditions. These imaging techniques provide valuable insights into the body's internal structures, aiding in the detection of abnormalities, assessment of disease progression, and development of treatment strategies. This article delves into two primary radiological investigations, chest X-rays and CT scans, outlining their purpose, procedures, and...