Skip to main content

Table 2 Repeatability and diagnostic performance of the ADC parameter

From: Diffusion-weighted imaging of rectal cancer on repeatability and cancer characterization: an effect of b-value distribution study

No

b-value distribution

ICC (95% CI)

CV (%)

bias (%, ±1.96SD)

AUC (95% CI)

ρ (95% CI)

α

2

0,1000

0.984(0.967–0.992)

3.840

−1.1 (− 11.0–8.9)

0.901 (0.743–0.978)

− 0.745(− 0.868 - -0.535)

2.581

2

0,1500

0.970(0.938–0.985)

4.063

0.8 (− 10.1–11.6)

0.829 (0.655–0.939)

− 0.629(− 0.802 - -0.359)

2.379

2

0,2000

0.968(0.933–0.984)

3.729

1.2 (− 8.7–11.2)

0.837 (0.664–0.943)

− 0.629(− 0.802 - -0.359)

2.385

2

50,1000

0.978(0.955–0.989)

3.333

0.6 (− 8.7–10.0)

0.877 (0.713–0.966)

− 0.710(− 0.849 - -0.480)

2.526

2

50,1500

0.977(0.954–0.989)

3.531

− 0.1 (− 10.2–10.0)

0.813 (0.637–0.929)

− 0.618(− 0.795 - -0.343)

2.372

2

100,1000

0.985(0.970–0.993)

3.514

− 1.8 (− 10.4–6.8)

0.730 (0.545–0.871)

−0.494(− 0.719 - -0.175)

2.156

2

100,2000

0.984(0.966–0.992)

3.762

0.4 (− 9.7–10.5)

0.837 (0.664–0.943)

− 0.598(− 0.783 - -0.314)

2.377

2

200,1000

0.976(0.952–0.989)

3.582

−0.9 (− 10.1–8.3)

0.734 (0.549–0.874)

−0.462(− 0.698 - -0.135)

2.127

2

200,2000

0.988(0.975–0.994)

2.991

0.4 (− 8.5–9.2)

0.766(0.583–0.897)

− 0.511(− 0.730 - -0.197)

2.231

3

0,100,1000

0.990(0.979–0.995)

2.407

−0.6 (− 7.3–6.1)

0.825(0.651–0.936)

−0.635(− 0.805 - -0.368)

2.420

3

0,200,2000

0.980(0.959–0.990)

2.963

− 1.1 (− 9.0–6.7)

0.766(0.583–0.897)

−0.525(− 0.739 - -0.216)

2.230

3

50,1000,2000

0.989(0.978–0.995)

2.857

0.4 (− 7.7–8.6)

0.865(0.698–0.959)

−0.644(− 0.810 - -0.380)

2.465

3

100,1000,1500

0.958(0.914–0.979)

5.568

− 1.7 (− 16.1–12.7)

0.933(0.785–0.991)

−0.751(− 0.871 - -0.544)

2.569

3

200,500,1500

0.966(0.929–0.983)

3.613

−2.2 (− 11.3–6.9)

0.802(0.623–0.921)

−0.572(− 0.767 - -0.278)

2.282

3

500,1500,2000

0.922(0.840–0.962)

4.374

−0.3 (− 11.2–10.7)

0.937(0.791–0.992)

−0.759(− 0.876 - -0.558)

2.571

4

0,100,1000,1500

0.989(0.977–0.995)

3.077

−2.6 (− 10.4–5.2)

0.750(0.549–0.874)

−0.496(− 0.706 - − 0.149)

2.178

4

0,200,1000,2000

0.990(0.980–0.995)

2.143

-0.1 (− 6.0–5.8)

0.833(0.660–0.941)

−0.595(− 0.781 - -0.310)

2.396

4

0,500,1000,1500

0.983(0.965–0.992)

3.703

− 1.4 (− 10.0–7.1)

0.829(0.655–0.939)

−0.574(− 0.769 - -0.282)

2.335

4

100,200,500,1000

0.994(0.988–0.997)

2.554

−2.7 (− 7.9–2.6)

0.804(0.625–0.922)

−0.567(− 0.765 - -0.272)

2.312

4

500,1000,1500, 2000

0.920(0.835–0.961)

4.892

− 1.9 (− 13.5–9.7)

0.938(0.793–0.993)

−0.748(− 0.870 - -0.540)

2.538

5

0,50,200,500,1000

0.998(0.996–0.999)

1.475

−1.6 (− 4.5–1.2)

0.714(0.528–0.859)

−0.407(− 0.662 - -0.068)

2.088

5

0,200,500,1000,1500

0.995(0.990–0.998)

1.933

−1.6 (− 6.0–2.9)

0.774(0.592–0.902)

−0.496(− 0.720 - -0.179)

2.230

5

0,500,1000,1500,2000

0.984(0.968–0.992)

3.349

−1.4 (− 9.7–6.8)

0.871(0.705–0.963)

−0.664(− 0.822 - -0.410)

2.472

5

50,200,500,1500,2000

0.982(0.962–0.991)

2.735

−0.9 (− 7.9–6.0)

0.865(0.698–0.959)

−0.664(− 0.810 - -0.380)

2.475

5

100,200,500,1000,2000

0.990(0.979–0.995)

2.263

−1.6 (− 6.8–3.6)

0.861(0.693–0.957)

−0.626(− 0.800 - -0.355)

2.438

6

0,100,200,500,1000,1500

0.996(0.992–0.998)

1.784

−1.4 (− 5.6–2.8)

0.829 (0.655–0.939)

−0.609(− 0.790 - -0.331)

2.402

6

50,200,500,1000,1500,2000

0.988(0.976–0.994)

2.520

−1.7 (− 7.6–4.3)

0.825(0.651–0.936)

−0.589(− 0.778 - -0.302)

2.360

7

0,100,200,500,1000,1500,2000

0.993(0.985–0.996)

2.137

−1.5 (−6.4–3.5)

0.821(0.646–0.934)

−0.597(− 0.783 - -0.314)

2.375

8

0,50,100,200,250,500,1000,1500

0.995(0.991–0.998)

1.966

−1.6 (− 6.2–3.0)

0.817(0.641–0.931)

−0.597(− 0.783 - -0.314)

2.373

9

0,50,100,150,200,500,1000,1500,2000

0.988(0.975–0.994)

2.811

−1.9 (−8.6–4.8)

0.857(0.688–0.955)

−0.618(− 0.795 - -0.343)

2.416

10

All b values

0.991(0.981–0.996)

2.566

−2.0 (− 7.7–3.6)

0.774(0.592–0.902)

−0.471(− 0.704 - -0.146)

2.190

  1. ICC intraclass correlation coefficient, SD standard deviation, CI confidence interval, CV coefficient of variation, AUC area under receiver operating characteristic curve, ρ: Spearman correlation coefficient, α: = ICC + AUC + |ρ|- CV - |bias|