Skip to main content

Table 5 Agreement between manual and automated analyses according to image quality

From: Fully automated quantification of biventricular volumes and function in cardiovascular magnetic resonance: applicability to clinical routine settings

 

Parameter

Mean Difference (SD of the Diff.)

ICC (95% CI)

CoV (%)

Good image quality (Score ≤ 1)

LV (n = 187)

LV Mass

3.0 (7.9)

0.96 (0.94–0.98)

12.8

LV EDV

3.4 (6.2)

0.99 (0.97–0.99)

6.5

LV ESV

1.7 (4.1)

0.99 (0.99–1.00)

9.0

LV SV

1.5 (5.2)

0.95 (0.93–0.96)

10.5

LV EF

−0.6 (3.5)

0.98 (0.98–0.99)

6.2

RV (n = 188)

RV EDV

7.8 (10.5)

0.93 (0.75–0.97)

12.2

RV ESV

1.0 (6.9)

0.96 (0.95–0.97)

16.9

RV SV

6.7 (8.5)

0.79 (0.40–0.90)

18.9

RV EF

3.0 (6.1)

0.88 (0.78–0.93)

11.5

Reduced image quality (Score ≥ 2)

LV (n = 113)

LV Mass

1.3 (11.2)

0.94 (0.91–0.96)

16.7

LV EDV

7.5 (9.7)

0.95 (0.82–0.98)

10.8

LV ESV

8.9 (16.3)

0.87 (0.74–0.93)

37.9

LV SV

−1.6 (9.6)

0.86 (0.80–0.90)

20.5

LV EF

−5.6 (7.5)

0.90 (0.67–0.96)

13.8

RV (n = 112)

RV EDV

6.7 (14.3)

0.91 (0.83–0.95)

16.8

RV ESV

−5.9 (12.1)

0.84 (0.70–0.90)

30.2

RV SV

12.7 (12.6)

0.67 (0.02–0.86)

27.9

RV EF

10.6 (12.1)

0.56 (0.03–0.77)

22.5

  1. Biventricular volumes and LV mass were indexed to body surface area. SD standard deviation, ICC intraclass correlation coefficient, CoV coefficient of variation, LV left ventricular, RV right ventricular, EDV/ESV end-diastolic/systolic volume, SV stroke volume, EF ejection fraction