Skip to main content

Table 4 Demographic and clinical characteristics of the datasets

From: Automated assessment of cardiac pathologies on cardiac MRI using T1-mapping and late gadolinium phase sensitive inversion recovery sequences with deep learning

Participant characteristics

Training set

(nā€‰=ā€‰130)

Validation set (nā€‰=ā€‰30)

Test set

(nā€‰=ā€‰40)

Sex

Ā 

ā€ƒWomen

48 (36.9)

7 (23.3)

13 (32.5)

ā€ƒMen

82 (63.1)

23 (76.7)

27 (67.5)

Baseline EF (%)

57.3ā€‰Ā±ā€‰13.3

58.5ā€‰Ā±ā€‰12.9

57.9ā€‰Ā±ā€‰15.2

No. of normal MRIs

41 (31.5)

9 (30.0)

13 (32.5)

No. of abnormal MRIs

89 (68.5)

21 (70.0)

27 (67.5)

ā€ƒAcute myocardial infarction

6 (6.7)

2 (9.5)

2 (7.4)

ā€ƒChronic myocardial infarction

11 (12.4)

1 (4.8)

2 (7.4)

ā€ƒMyocarditis (new and old)

24 (27.0)

7 (33.3)

5 (18.5)

ā€ƒDCM

10 (11.2)

4 (19.0)

5 (18.5)

ā€ƒHCM

5 (5.6)

0 (0)

1 (3.7)

ā€ƒOthers

33 (37.1)

7 (33.3)

12 (44.4)

  1. Note. ā€” Data are numbers of patients with percentage in parentheses. Ejection fraction (EF) is mean dataā€‰Ā±ā€‰standard deviation. HCM is hypertrophic cardiomyopathy. DCM is dilated cardiomyopathy. Others includes Pericarditis, aortic stenosis, storage disease, systemic sclerosis, cardiomyopathy, amyloidosis, pericardial effusion, unspecific load, exercise-induced myocardial ischemia, arrhythmogenic right ventricular dysplasia. Training, validation, and test set were spilt 65/15/20% retrospectively.