The preliminary exploration of 64-slice volume computed tomography in the accurate measurement of pleural effusion

Abstract

The 64-slice CT volume-rendering technique can accurately measure the volume in pleural effusion patients, and a linear regression equation can be used to estimate the volume of the free pleural effusion.

Deep Learning for Detection of Pneumothorax and Pleural Effusion on Chest Radiographs

“…CT-derived PEf volume is proven to be a close surrogate for true PEf volume 22 . CT volumetry enabled us to address the effects of size and radiographic view (ie, PA vs AP) on pleural pathology detection.…”

“…Many FDA-cleared algorithms for identifying pleural pathology provide nominal sensitivity and specificity > 90% for Ptx and PEf detection; however, these validation studies feature CXR labels as ground truth, which may render lower sensitivity and modest interrater agreement compared with CT. [9][10][11] For instance, Yousefifard et al 20 found a pooled sensitivity of 0.51 (95% CI: 0.33, 0.68) among 10 studies comparing the detection of PEfs, whereas Ebrahimi et al 21 a pooled sensitivity of 0.46 (95% CI: 0.36, 0.56) among 21 studies comparing the detection of Ptx on CXR against CT. CT-derived PEf volume is proven to be a close surrogate for true PEf volume. 22 CT volumetry enabled us to address the effects of size and radiographic view (ie, PA vs AP) on pleural pathology detection. Though both PEf and Ptx volumes were correlated with Likert and AI probability scores, only PEf volume and radiographic view influenced PEf detection.…”