Computer-based quantitative modeling of chest CT findings in pulmonary hypertension and its association with physiologic and clinical variables

Summary

The RSNA archive records this work as a scientific informal poster presentation on November 28, 2011. The study explored computer-based quantification of ground-glass and fibrotic interstitial patterns in chest CT scans and their associations with clinical and physiological measurements. The reported analyses found correlations between quantitative CT features and several physiological variables. These are associations, not evidence that the CT measures cause or clinically diagnose pulmonary hypertension. View the RSNA abstract.

• The RSNA abstract reports that a radiologist annotated 29 CT scans; the resulting regions of interest were used to train an automatic pattern-identification model. The model was then applied to 83 CT scans, and quantitative imaging measurements were compared with clinical and physiological values.
• The abstract reports correlations between quantitative CT features and mean pulmonary arterial pressure, pulmonary pressure, and pulmonary vascular resistance index. Its findings are from the reported study; the archive does not establish prospective validation or clinical readiness.