Skip to main content

Table 2 OLP regression results for CIMT measurements

From: Carotid intima-medial thickness measured on multiple ultrasound frames: evaluation of a DICOM-based software system

Comparison a' 95% CI b' 95% CI Fixed Bias Proportional Bias
Six-frame measurement protocol
Inter-Sonographer 0.048 -0.03, 0.12 0.964 0.87, 1.07 No No
Intra-Sonographer 1 0.013 -0.08, 0.10 1.002 0.89, 1.13 No No
Intra-Sonographer 2 -0.025 -0.12, 0.06 1.025 0.92, 1.15 No No
Inter-Observer -0.024 -0.06, 0.01 1.053 1.01, 1.10 No Yes
Intra-Observer 1 0.004 -0.02, 0.03 0.992 0.96, 1.03 No No
Intra-Observer 2 -0.031 -0.07, 0.003 1.056 1.01, 1.11 No Yes
Sixty-frame measurement protocol
Inter-Sonographer 0.001 -0.10, 0.09 1.027 0.91, 1.16 No No
Intra-Sonographer 1 -0.008 -0.11, 0.08 1.017 0.90, 1.15 No No
Intra-Sonographer 2 -0.009 -0.10, 0.07 0.995 0.89,1.11 No No
Inter-Observer -0.106 -0.17, -0.05 1.159 1.08,1.24 Yes Yes
Intra-Observer 1 -0.020 -0.04, -0.003 1.031 1.01,1.05 Yes Yes
  1. OLP indicates ordinary least products; CIMT, carotid intima-medial thickness; CI, confidence interval.
  2. The ordinary least products regression equation is y = a' + b'x. 95% CIs are for the population parameters α ' and β '. Fixed bias is present if the 95% CI for α ' does not contain zero. Proportional bias is present if 95% CI for β ' does not contain one.