Skip to main content

Table 4 Summary of input parameters used in FSI models and prediction methods

From: Predicting plaque vulnerability change using intravascular ultrasound + optical coherence tomography image-based fluid–structure interaction models and machine learning methods with patient follow-up data: a feasibility study

Image

Resolution (μm)

Image size

Field of view (mm2)

Pixel size (mm)

Image

    

 IVUS

150–200

512*512

9*9

0.01752

 OCT

10–20

704*704

7.01*7.01

0.00996

 Angiography

 > 200

512*512

152*152

0.29688

Model parameters

c1 (kPa)

c2 (kPa)

D1 (kPa)

D2

K (kPa)

K1 (kPa)

K2

Material

       

 Tissue

− 278.7

24.35

133.7

2

13,157

7.19

23.5

 Lipid

0.5

0

0.5

1.5

1250

 Calcification

92

0

36

2

164,000

Pressures

Maximum = 136 mmHg, Minimum = 88 mmHg

Prediction methods

Setting

Data processing

Validation

SVM

Kernel function: Gaussian radial basis function

Synthetic minority oversampling technique (SMOTE)

Fivefold cross-validation

RF

Number of tree: 20

EL

Number of ensemble learning cycles: 100

DA

Discriminant type: linear