Magnetic resonance tomography method with screening of signals from tissues with long transverse relaxation times

• US 5,814,991 A
• Filed: 04/14/1997
• Issued: 09/29/1998
• Est. Priority Date: 04/24/1996
• Status: Expired due to Term

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Deimling, Michael
• Primary Examiner(s): O'Shea, Sandra L.
• Assistant Examiner(s): EISENBERG, MICH
• Application Number:
  US08/843,267
• Time in Patent Office: 533 Days
• Field of Search: 324/303, 324/300, 324/307, 324/309, 324/311, 324/312, 600/410
• US Class Current: 324/309
• CPC Class Codes: G01R 33/50 based on the determination of relaxation times , e.g. T1 measurement by IR sequences; T2 measurement by multiple-echo sequences

Abstract

In a nuclear magnetic resonance tomography apparatus and a method in the form of a pulse sequence for operating the apparatus, after an excitation of spins in a subject, two groups of nuclear magnetic resonance signals are obtained in two time intervals at different chronological spacings from the excitation time. An image is obtained on the basis of the signal differences of nuclear magnetic resonance signals of the first and second groups with corresponding locus-encoding. Signal contributions from tissues with longer T2 relaxation time can thereby be screened out.

Claims

1. A method for operating a nuclear magnetic resonance tomography apparatus for examining first tissue in an examination subject having a first T2 time, the subject also containing tissue having a second T2 time which is longer than said first T2 time, said method comprising the steps of:

exciting nuclear spins in each of said first and second tissues in said subject, beginning at an excitation time, under transverse magnetization which decays with said first and second T2 time constants, obtaining a first group and a second group of locus-encoded nuclear magnetic resonance signals from said subject;

obtaining said first group of nuclear magnetic resonance signals shortly after said excitation time;

obtaining said second group of nuclear magnetic resonance signals after said first group in a time interval following said excitation time at which said second tissue having said second time constant supplies a predominant signal contribution to the nuclear magnetic resonance signals; and

reconstructing an image of said examination subject containing said first and second tissues on the basis of differences in the respective content of said first and second groups of locus-encoded nuclear magnetic resonance signals having identical locus encoding.

2. A method as claimed in claim 1 wherein the step of reconstructing an image includes digitizing each of said first and second groups of locus-encoded nuclear magnetic resonance signals, to obtain digitized first and second groups of signals, and forming a subtraction difference between said first and second digitized groups of signals having identical locus encoding.

3. A method as claimed in claim 1 wherein the step of reconstructing an image includes Fourier-transforming each of said first and second groups of locus-encoded nuclear magnetic resonance signals with spatial allocation, to obtain first and second Fourier-transformed groups of signals, and forming a subtraction difference between said first and second groups of Fourier-transformed signals having identical locus encoding.

4. A method as claimed in claim 1 wherein the step of reconstructing an image comprises Fourier-transforming each of said first and second groups of locus-encoded nuclear magnetic resonance signals to obtain first and second groups of Fourier-transformed signals, each of said first and second groups of Fourier-transformed signals having a real part and an imaginary part, forming respective absolute values of the real part and the imaginary part of each of the signals of said first and second groups of Fourier-transformed signals to obtain first and second groups of absolute values, and forming a subtraction difference between said first and second groups of absolute values of the same locus allocation.

5. A method as claimed in claim 1 wherein the steps of exciting nuclear spins and reading out said nuclear magnetic resonance signals comprise:

emitting a 90°

radio-frequency pulse to said subject;

emitting a plurality of refocusing radio-frequency pulses to said subject; and

reading out said first and second groups of locus-encoded nuclear magnetic resonance signals under respective read out gradients.

6. A method as claimed in claim 5 wherein the step of reading out said nuclear magnetic resonance signals includes multiply inverting at least one readout gradient after each refocusing radio-frequency pulse for obtaining a plurality of said nuclear magnetic resonance signals under the multiply inverted readout gradient.

7. A method as claimed in claim 1 comprising the step of locus-encoding said nuclear magnetic resonance signals by emitting a phase-encoding gradient with a first amplitude-time product before each nuclear magnetic resonance signal and emitting a phase-encoding gradient with an amplitude-time product having an opposite polarity after each nuclear magnetic resonance signal.

8. A method as claimed in claim 7 wherein said phase-encoding gradient produce respective phase factors for each of said nuclear magnetic resonance signals in said first and second groups of nuclear magnetic resonance signals, and said method comprising the additional steps of entering said first group of nuclear magnetic resonance signals into a first raw data matrix according to the respective phase factors of said nuclear magnetic resonance signals in said first group, entering said second group of nuclear magnetic resonance signals into a second raw data matrix according to the respective phase factors of the nuclear magnetic resonance signals in said second group, with only a portion of the nuclear magnetic resonance signals required for a complete filling of each raw data matrix being obtained for each excitation of spins, and repeating the steps of exciting nuclear spins and reading out said first and second groups of nuclear magnetic resonance signals until each raw data matrix is completely filled.

9. A method as claimed in claim 8 wherein each of said raw data matrices represent k-space values, and wherein the step of obtaining said second group of nuclear magnetic resonance signals comprises obtaining central k-space values for said second group of nuclear magnetic resonance signals at an end of reading out of said second group of nuclear magnetic resonance signals.

10. A nuclear magnetic resonance tomography apparatus for examining first tissue in an examination subject having a first T2 time, said subject also containing second tissue having a second T2 time which is longer than said first T2 time, said apparatus comprising:

transmission means for emitting a plurality of 90°

radio-frequency excitation pulses for exciting nuclear spins in said examination subject, and for subsequently emitting a plurality of 180°

refocusing radio-frequency pulses;

reception means for receiving a first group of nuclear magnetic resonance signals at a short chronological distance from a selected 90°

radio-frequency pulse and for receiving a second group of nuclear magnetic resonance signals at a larger chronological distance from said selected 90°

radio-frequency pulse;

a first memory for storing said first group of nuclear magnetic resonance signals and a second memory for said second group of nuclear magnetic resonance signals;

subtraction means for subtracting corresponding nuclear magnetic resonance signals in said first and second groups respectively stored in said first and second memories to obtain a difference; and

image reconstruction means for reconstructing an image of said subject including said first and second tissues from said difference.