CAD for pulmonary infections: automatic detection of tree-in-bud opacities
Radiological Society of North America (RSNA), 97th Scientific Assembly and Annual Meeting
Abstract
An abstract is not available in the lab archive. View the paper at its source.
A 2011 RSNA meeting presentation described a computer-assisted method for detecting tree-in-bud patterns on chest CT. The retrospective evaluation used scans from patients with human parainfluenza virus and normal volunteers.
Summary
The presentation described a computer-assisted detection (CAD) method intended to identify tree-in-bud (TIB) opacities—branching patterns associated with small-airway abnormalities—in chest computed tomography (CT). The abstract reports retrospective analysis of 39 CT scans from patients with human parainfluenza virus confirmed by laboratory testing and 21 normal volunteer scans. The method used local appearance and size information to select candidate regions, then combined shape and texture features to classify them. The reported accuracy was 90.96%.
The evaluation was retrospective and used a limited dataset; the abstract does not establish prospective performance or clinical readiness. The RSNA meeting record provides the available abstract.
Cite this work
U. Bagci, J. Yao, J. Caban, T. N. Palmore, A. F. Suffredini, and D. J. Mollura (2011). CAD for pulmonary infections: automatic detection of tree-in-bud opacities. Radiological Society of North America (RSNA), 97th Scientific Assembly and Annual Meeting https://www.researchgate.net/publication/266106060_CAD_for_Pulmonary_Infections_Automatic_Detection_of_Tree-in-Bud_Opacities
BibTeX