Back

Computer-Aided Reading of Chest Radiographs for Pediatric Tuberculosis: Current Status and Future Directions

DuPont, M.; Castro, R.; Kik, S. V.; Palmer, M.; Seddon, J. A.; Jaganath, D.

2024-10-09 infectious diseases
10.1101/2024.10.08.24314837 medRxiv
Show abstract

Computer-aided detection (CAD) systems for automated reading of chest X-rays (CXRs) have been developed and approved for tuberculosis (TB) triage in adults, but not in children. However, CXR is frequently the only adjunctive tool to clinical assessment for the evaluation of pediatric TB in primary health settings, and children would benefit from CAD models that can detect their unique clinical and radiographic features. The current literature on CAD CXR algorithms for detecting TB and other pulmonary infections in children is limited, but highlights promising models and approaches. To advance CAD for childhood TB, large, diverse, pediatric CXR datasets are required that are linked to standardized clinical and radiographic TB classifications. These datasets could be used to train or fine-tune existing algorithms for TB screening, diagnosis and severity stratification. It is critical to include children in CAD models to increase equity and reduce the global burden of TB disease.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.