Selected article for: "cut point and sensitivity specificity"
Author: Cereda, Maurizio; Xin, Yi; Hamedani, Hooman; Bellani, Giacomo; Kadlecek, Stephen; Clapp, Justin; Guerra, Luca; Meeder, Natalie; Rajaei, Jennia; Tustison, Nicholas J; Gee, James C; Kavanagh, Brian P; Rizi, Rahim R
Title: Tidal changes on CT and progression of ARDS
  • Document date: 2017_6_20
    • ID: sncded7z_27
    Snippet: After superimposing all sequential images, evolving density of individual voxels was tracked over time. Injury progression was a binary outcome and was defined by an increase of both EI and EE densities to above the −300 HU cut point ('severe injury') in subsequent images (figure 1). In the earlier images, voxels with higher probability of later progression were considered 'high risk'. We measured such probability by analysing rangewise receive.....
    Document: After superimposing all sequential images, evolving density of individual voxels was tracked over time. Injury progression was a binary outcome and was defined by an increase of both EI and EE densities to above the −300 HU cut point ('severe injury') in subsequent images (figure 1). In the earlier images, voxels with higher probability of later progression were considered 'high risk'. We measured such probability by analysing rangewise receiver operating characteristic (ROC) 16 between consecutive PRMs: values of sensitivity and specificity for observed progression were calculated for voxels of each bin (online supplementary figure 5 ). Using multivariate linear regression in a two-stage approach (see online Supplementary material), we created a statistical model to predict the future distribution of lung CT densities and the worsening of C dyn , starting from the percentage of voxels which, in earlier images, fell into the highrisk and the 'severe injury' density ranges.

    Search related documents:
    Co phrase search for related documents
    • binary outcome and high risk consider: 1
    • binary outcome and multivariate linear: 1, 2
    • binary outcome and multivariate linear regression: 1, 2
    • binary outcome and observed progression: 1
    • binary outcome and online supplementary material: 1, 2, 3
    • binary outcome and specificity sensitivity: 1
    • binary outcome and supplementary material: 1, 2, 3
    • cut point and high probability: 1
    • cut point and high risk: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31
    • cut point and online supplementary material: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10