Table 4:

Per-patient and per-lesion sensitivity and specificity using GEEs for target neurologic findings accounting for acute neurologic deficits

Dose-Kernel ConfigurationPer-Patient Sensitivity for CT Findings Accounting for Acute Neurologic Deficits (GEE) (%) (95% CI) (Range) (%)Per-Patient Specificity for CT Findings Accounting for Acute Neurologic Deficits (GEE) (%) (95% CI) (Range) (%)Target Lesion Sensitivity for CT Findings Accounting for Acute Neurologic Deficits (GEE) (%) (95% CI) (Range) (%)
250-eff. mAs IR81.7 (71.1–92.3) (78.6–83.3)93.5 (88.9–98.1) (85.4–100.0)
200-eff. mAs FBP79.4 (68.2–90.6) (76.2–83.3)91.9 (87.5–96.3) (80.5–100.0)68.6 (62.3–74.9) (61.4–72.9)
100-eff. mAs IR77.0 (65.5–88.5) (73.8–81.0)88.6 (82.8–94.4) (73.2–95.1)68.1 (61.8–74.4) (64.3–71.4)
100-eff. mAs FBP74.6 (62.4–86.8) (69.0–78.6)87.0 (81.1–92.9) (75.6–95.1)62.9 (56.3–69.4) (57.1–65.7)
50-eff. mAs IR73.8 (62.3–85.4) (66.7–78.6)88.6 (82.8–94.4) (75.6–97.6)60.5 (53.9–67.1) (52.9–65.7)
50-eff. mAs FBP72.2 (60.6–83.8) (71.4–73.8)83.7 (77.7–89.8) (69.8–86.0)61.0 (54.4–67.6) (60.0–62.9)
25-eff. mAs IR65.9 (53.3–78.4) (61.9–71.4)88.6 (82.4–94.8) (85.4–92.7)53.3 (46.6–60.1 (45.7–62.9)a
  • a The 95% confidence interval does not overlap the routine dose, so the dose-reconstruction configuration is significantly worse.