Skip to main content

Table 5 Reproducibility calculated as rn in percent for 54 pairs of consecutive spectra

From: Reproducibility of brain metabolite concentration measurements in lesion free white matter at 1.5 T

Protocol

6

10

12

13a

14

14 ref

18

18 ref

19a

Naa

4.4

2.3

2.5

2.5

3.1

4.5

2.2

4.2

2.3

Cre3

4.9

4.7

4.3

4.5

4.4

0

4.3

0

4.4

Cho

5.3

5.3

4.7

4.7

4.1

4.9

4.8

5.2

4.9

Myo

12.9

23.7

10.0

11.3

9.3

9.1

9.0

9.0

9.7

Glx

7.7

5.7

12.9

13.5

4.6

6.0

5.2

5.7

5.5

Sum of all met.

35.1

41.8

34.4

36.4

25.5

24.5

25.5

24.2

26.8

Naa, Cre3, Cho

14.5

12.4

11.5

11.7

11.6

(9.4)

11.3

(9.4)

11.6

all–Myo

22.2

18.2

24.4

25.2

16.2

15.4

16.5

15.1

17.1

all–Glx

27.4

36.1

21.5

23.0

20.9

18.5

20.3

18.5

21.3

amp.(Naa)

190.8

206.9

181.0

180.4

200.0

1.9

208.2

2.0

209.0

amp.(Cre3)

96.5

95.9

97.3

97.5

104.5

1

105.4

1

106.5

amp.(Cho)

28.2

28.5

29.4

29.3

31.5

0.3

31.7

0.3

32.0

amp. (Myo)

55.3

66.9

57.1

57.9

68.4

0.6

71.5

0.7

71.4

amp. (Glx)

226.5

279.7

155.9

159.7

322.3

3.1

261.8

2.5

255.7

±% of mean

         

Naa

13.1

6.1

6.4

6.3

8.01

12.8

6.3

11.1

6.3

Cre3

11.2

13.2

11.7

11.5

10.9

0.0

11.4

0.0

11.1

Cho

11.9

13.8

13.8

14.4

10.0

15.2

13.9

18.3

13.3

Myo

57.2

88.2

35.7

50.2

34.0

33.4

28.2

28.2

28.9

Glx

21.2

16.2

29.2

30.8

11.9

14.9

13.3

12.0

13.5

  1. The minimal values for each of the five main metabolites are marked bold. The values of the referenced metabolites are in italic. Protocol 18 shows up together with protocol 14 the best reproducibility rated by the sum of the five rn’s of the main metabolites. Protocol 18 is preferable, because
  2. 1. it needs less fit parameters for identical reproducibility results
  3. 2. the rn sum of the major metabolites is minimal for 18 ref, because rn of Cre3 is minimal in protocol 18
  4. Note: In all cases the rn calculated with 54 pairs is similar as in the case of 11 pairs (compare Table 3 with Table 4). This legitimates the missing calculations with all 54 pairs for the protocols not enlisted in Table 4
  5. “ref” indicates that the amplitude of a metabolite is referenced to Cre3, which is the calculation of relative concentrations i.e. evaluatedas ampref(met) = amp(met)/amp(Cre3)
  6. “±% of mean” for each metabolite (five last lines) is the maximal percentage deviation from the mean value for a pair of amplitudes