Schematic timing diagram of the SAP-T1 pulse sequence. The first five measurements are shown, in which all data for one slice are acquired. The slice acquisition is repeated with four different preparation pulse delays (PD) and also without the 90° preparation pulse (equivalent to ). This set of measurements is repeated with different trigger delays (TD) which are unique for each slice. : slice selection gradient, : phase encoding gradient, : frequency encoding gradient.
(a) Plots of theoretical calculation of the -magnetization evolution for various preparation pulse delays for a T1 value of . The 12th pulses which acquire the lines are indicated with a black dot. The time from the center of the preparation pulse to the line is the PD. The calculation takes account of the flip angle sweep during the first 12 rf pulses as described in the text. (b) Each point from (a) is normalized here to a theoretical calculation of the measurement. This measurement is not a “true” measurement of the fully relaxed magnetization because it also is affected by the multiple readout pulses and also approaches a steady state. For this reason the points on the graph associated with the filling of the lines, when fitted to a monoexponential function, do not exactly trace the theoretical monoexponential T1 recovery line (also plotted). This is seen more clearly in the enlarged portion. Use of in Eq. (4) allows the fitted curve to move along the axis. The concave shapes of the dotted lines are a consequence of the normalization process.
(a) Accuracy of the T1 relaxation rate of IR-SE (used for T1 reference values) and the SAP-T1 sequence for one of the three slices (slice 3), in a phantom containing a range of concentrations of Gd-DTPA expected to be found in the myocardium during the first-pass of Gd-DTPA in a myocardial perfusion study. Graphs for slices 2 and 1 are similar (not shown). (b) The linearity of relaxation rate of IR-SE (squares, used for T1 reference values) and the SAP-T1 method using the sequence (triangles) with the concentration of Gd-DTPA for slice 3 (in vitro). Graphs for slices 2 and 1 are similar (not shown). The concentration of Gd-DTPA in the myocardium during a first-pass study is typically less than . The maximum 6% error in T1 using SAP-T1 translates to a maximum error in Gd-DTPA concentration estimation of 4.5%, for ( Gd-DTPA).
(a) In vitro T1 pixel map of Gd-DTPA-doped agarose gel phantoms using SAP-T1 with . (b) Example in vivo source image (mid-short-axis slice, , ), from patient 1. (c) In vivo SAP-T1 map from patient 1, precontrast. (d) In vivo SAP-T1 map from patient 1, postcontrast. (e) In vivo SAP-T1 map from patient 2, precontrast. (e) In vivo SAP-T1 map from patient 2, postcontrast.
The 15 single heartbeat measurements required to obtain a full data set (four points on the recovery curve for three slices). Note that the trigger delays remain unaltered for each slice. The measurements may be made in a single breath hold if the measurements are set as measurements 1–3. The order thereafter is not significant.
T1 relaxation times estimated by SAP-T1 technique in vitro for three slices compared with reference values obtained using a multiple-point IR-SE technique.
Article metrics loading...
Full text loading...