Table 2 SNR and CNR results of conventional FSE, accelerated FSE without DNN reconstruction, and accelerated FSE with DNN reconstruction (FSE-DNN).

From: Highly accelerated knee magnetic resonance imaging using deep neural network (DNN)–based reconstruction: prospective, multi-reader, multi-vendor study

 

Philips (Ellition CX)

GE (Discovery MR750)

Siemens (Prisma Fit)

p-value

FSE-conventional

Accelerated FSE

FSE-DNN

FSE-conventional

Accelerated FSE

FSE-DNN

FSE-conventional

Accelerated FSE

FSE-DNN

SNR

Ax T2W FS

4.8 ± 1.3

4.3 ± .9

7.7 ± 2.9

14.2 ± 5.4

8.6 ± 4.4

18.4 ± 12.8

2.0 ± 0.2

2.1 ± 0.4

7.3 ± 2.1

 < 0.001

Ax T1W

22.8 ± 4.0

13.7 ± 3.5

28.1 ± 6.3

90.9 ± 51.5

57.8 ± 31.9

101.2 ± 53.6

12.2 ± 1.9

8.0 ± 1.3

34.6 ± 6.5

0.002

Sag T2W

22.4 ± 4.4

18.8 ± 3.4

38.9 ± 7.0

33.7 ± 15.5

30.5 ± 20.6

41.9 ± 34.8

9.9 ± 1.7

8.0 ± 1.5

39.3 ± 15.8

 < 0.001

Cor T2W FS

5.0 ± 1.7

4.2 ± .9

11. ± 4.3

9.9 ± 3.1

6.7 ± 2.6

11.9 ± 3.6

3.0 ± 0.5

2.6 ± 0.4

12.5 ± 4.2

 < 0.001

CNR

BM-muscle on T2W FS

9.5 ± 6.5

6.0 ± 6.0

10.1 ± 12.0

14.3 ± 10.2

9.5 ± 10.0

17.7 ± 14.1

8.1 ± 3.1

6.5 ± 2.9

37.5 ± 28.2

0.001

BM-muscle on T1W

54.8 ± 16.4

31.7 ± 16.1

65.1 ± 31.6

165.1 ± 76.8

123.1 ± 55

210.9 ± 88.4

31.6 ± 10.3

21.3 ± 7.7

101.2 ± 43.5

 < 0.001

BM-tendon on T2W

69.4 ± 22.7

56.3 ± 13.3

112.4 ± 29.1

172.7 ± 120.8

150.6 ± 121.3

194.1 ± 154.2

32.4 ± 7.8

25.0 ± 5.5

118.9 ± 54.6

 < 0.001

BM-meniscus on T2W FS

12.1 ± 9.9

8.6 ± 4.6

22.9 ± 18.0

29.2 ± 17.7

15.9 ± 11.8

31.8 ± 22.8

8.3 ± 5.0

6.3 ± 3.9

31.9 ± 27.1

 < 0.001

  1. Ax T2W FS Axial T2-weighted fat-saturated FSE, Ax T1W Axial T1-weighted FSE, Sag T2W Sag T2-weighted FSE, Cor T2W FS Cor T2-weighted fat-saturated FSE, p-values of paired t-test (Accelerated FSE and FSE-DNN).
Back to article page