Skip to main content

Table 4 Nodule-Based Estimates of the Detection of Lung-RADS Category 3 or 4 Nodules

From: 75% radiation dose reduction using deep learning reconstruction on low-dose chest CT

 

AUFROC

  

Sensitivity

  

FP rate

  
 

QLD-DLIR

LD-IR

P-value*

QLD-DLIR

LD-IR

P-value*

QLD-DLIR

LD-IR

P-value*

Reader 1

0.72

(0.62–0.82)

0.75

(0.66–0.85)

0.38

77.1%

(37/48)

72.9%

(35/48)

0.73

0.29

(29/100)

0.36

(36/100)

0.29

Reader 2

0.86

(0.78–0.94)

0.82

(0.74–0.91)

0.36

79.2%

(38/48)

68.8%

(33/48)

0.30

0.10

(10/100)

0.10

(10/100)

1.00

Reader 3

0.72

(0.63–0.82)

0.75

(0.66–0.84)

0.52

72.9%

(35/48)

75.0%

(36/48)

1.00

0.63

(63/100)

0.56

(56/100)

0.31

Pooled readers

0.77

(0.70–0.83)

0.78

(0.71–0.85)

0.68

76.4%

(110/144)

72.2%

(104/144)

0.35

0.34

(102/300)

0.34

(102/300)

1.00

  1. Note.—AUFROC values are presented with 95% confidence intervals. FP rates were calculated as the total number of FP nodules divided by the total number of patients (n = 100)
  2. AUFROC = area under the jackknife free-response receiver operating characteristic curve, DLIR = deep-learning image reconstruction, FP = false positive, IR = iterative reconstruction, LD = low dose, Lung-RADS = lung imaging reporting and data system, QLD = quarter of the low dose
  3. *P-values were calculated using jackknife free-response receiver operating characteristic curve analysis (for the AUFROC), the McNemar test (for the sensitivity of individual radiologists), the chi-square test (for the FP rate of individual radiologists), or generalized estimating equations (for the sensitivity and specificity of pooled radiologists)
Back to article page

BMC Medical Imaging

ISSN: 1471-2342

Contact us