Patent US5084675A successfully refreshed from Google Patents.
US5084675A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method for improving the signal-to-noise ratio in a nuclear magnetic resonance tomography apparatus
Assignee: Siemens AG
•
Filing: 1990-08-10
•
Granted: 1992-01-28
•
Expires: 2010-08-10
Abstract
Abstract In the echo planar imaging (EPI) method of operating a nuclear magnetic resonance (magnetic resonance imaging) tomography apparatus, the necessary, rapid switching of gradients having a high amplitude usually cannot be achieved with square-wave gradients, and therefore a read-out gradient having a sinusoidal shape is generally used. Given a chronologically equidistant sampling of the nuclear magnetic resonance signal in the time domain, however, a sampling in the k-space with different rates is present. The bandwidth of the signal is proportional to the sampling rate in the k-space. In the method disclosed herein, bandwidth of a bandpass filter or a low-pass filter in the signal processing circuit for the nuclear magnetic resonance signal is adapted to the time dependent bandwith of the nuclear magnetic resonance signal, i.e., to the sampling rate in the k-space. The bandwidth is increased with increasing sampling rate. The signal-to-noise ratio is noticeably improved.
Inventors
- 👤 Hans-Erich Reinfelder
- 👤 Hubertus Fischer
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-07 15:19
CPC Classifications
G; G01; G01R; G01R33/00; G01R33/20; G01R33/44; G01R33/48; G01R33/54; G01R33/56; G01R33/543; G01R33/5608; G01R33/561; G01R33/5615; G01R33/5616
Claims Analysis
Total Claims: 9
Claim 1
Independent Claim
203 words
A method for improving the signal-to-noise ratio in a nuclear magnetic resonance tomography apparatus comprising the steps of: generating an RF excitation pulse for exciting nuclear spins in an examination subject; simultaneously generating a gradient magnetic field in a first direction thereby making said RF pulse slice-selective; generating a phase-coding magnetic field gradient in a second direction; generating a read-out magnetic field gradient pulse sequence alternating in polarity from pulse-to-pulse; acquiring, in the presence of said read-out gradient, nuclear magnetic resonance signal arising from said nuclear spins; phase-demodulating said nuclear magnetic resonance signal to obtain a real signal part and an imaginary signal part; subjecting said real and imaginary signal parts to respective bandpass filtration, said bandpass filtration having a bandwidth; digitizing said real and imaginary signal parts at a varying sampling rate; adapting said bandwidth to said sample rate by increasing said bandwidth as said sampling rate increases; entering the digitized values of said real and imaginary signal parts per read-out gradient pulse in respective rows of a raw data matrix; subjecting said raw data matrix to a two-dimensional Fourier transformation to obtain an image matrix; and producing a visual image of the selected slice of said subject from said image matrix.
Claim 2
Independent Claim
203 words
A method for improving the signal-to-noise ratio in a nuclear magnetic resonance tomography apparatus comprising the steps of: generating an RF excitation pulse for exciting nuclear spins in an examination subject; simultaneously generating a gradient magnetic field in a first direction thereby making said RF pulse slice-selective; generating a phase-coding magnetic field gradient in a second direction; generating a read-out magnetic field gradient pulse sequence alternating in polarity from pulse-to-pulse; acquiring, in the presence of said read-out gradient, nuclear magnetic resonance signal arising from said nuclear spins; phase-demodulating said nuclear magnetic resonance signal to obtain a real signal part and an imaginary signal part; subjecting said real and imaginary signal parts to respective bandpass filtration, said bandpass filtration having a bandwidth; digitizing said real and imaginary signal parts at a varying sampling rate; adapting said bandwidth to said sample rate by increasing said bandwidth as said sampling rate increases; entering the digitized values of said real and imaginary signal parts per read-out gradient pulse in respective rows of a raw data matrix; subjecting said raw data matrix to a two-dimensional Fourier transformation to obtain an image matrix; and producing a visual image of the selected slice of said subject from said image matrix.
Claim 3
Dependent Claim
113 words
A method as claimed in claim 1 wherein the step of generating a read-out magnetic gradient field pulse sequence is further defined by generating a read-out magnetic field gradient pulse sequence having a sinusoidal chronological curve and wherein the step of subjecting said real and imaginary signal parts to respective bandpass filtration is further defined by subjecting said real and imaginary signal parts to respective bandpass filtration in respective bandpass filters each having a bandwidth Δf s having the following chronological curve: ##EQU6## wherein δt is the chronological sampling spacing, ω G is the radian frequency of said read-out gradient, and N is the number of columns and rows of said image matrix.
Claim 4
Dependent Claim
86 words
A method as claimed in claim 2 wherein the step of subjecting said real and imaginary signal parts to respective bandpass filtration is further defined by subjecting said real and imaginary signal parts to filtration in respective low-pass filters each having a variable limit frequency, and wherein the step of adapting said bandwidth to said sampling rate is further defined by adapting said bandwidth to said sampling rate by controlling the respective limit frequencies of said low-pass filters to achieve said chronological curve of said bandwidth.
Claim 5
Dependent Claim
100 words
A method as claimed in claim 2 wherein the step of subjecting said real and imaginary signal parts to respective bandpass filtration is further defined by supplying said real signal part to a first plurality of low-pass filters having fixed filter characteristics and supplying said imaginary signal part to a second plurality of low-pass filters having fixed filter characteristics, and wherein the step of adapting said bandwidth to said sampling rate is further defined by adapting said bandwidth to said sampling rate by switching said first and second pluralities of low-pass filters to approximate said chronological curve of said bandwidth.
Claim 6
Dependent Claim
73 words
A method as claimed in claim 1 wherein the step of digitizing said real and imaginary signal parts is further defined by digitizing said real and imaginary signal parts at a varying sampling rate which is higher than a sampling rate prescribed by the sampling theorem, and wherein the step of adapting said bandwidth to said sampling rate is further defined by convoluting said real and imaginary signal parts with a band-limiting function.
Claim 7
Dependent Claim
37 words
A method as claimed in claim 5 wherein the step of convoluting said real and imaginary signal parts with a band-limiting function is further defined by convoluting said real and imaginary signal parts with a sinc function.
Claim 8
Dependent Claim
51 words
A method as claimed in claim 5 wherein the step of phase-demodulating said nuclear magnetic resonance signal and all subsequent steps are conducted in an on-line signal processing system, and comprising the additional step of calculating values required for said band-limiting function off-line and storing said values in a digital memory.
Claim 9
Dependent Claim
151 words
A method as claimed in claim 1 wherein the step of digitizing said real and imaginary signal parts is further defined by digitizing said real and imaginary signal parts at a varying sampling rate which is greater than a sampling rate prescribed by the sampling theorem, and wherein the step of adapting said bandwidth to said sampling rate is further defined by entering numerical sequences corresponding to the sampled nuclear magnetic resonance signal in rows of said raw data memory, conducting a Fourier transformation of said rows to obtain Fourier functions, zeroizing said Fourier functions, back-transforming the zeroized Fourier functions in a sampling grid which is finer than the original sampling grid, subjecting the values in said finer sampling grid to a sinc interpolation to obtain interpolated signals, convoluting said interpolated signals with a band-limiting function to obtain convoluted signals, and writing said convoluted signals back into said raw data memory.
References Cited (Backward Citations)
EP0021535A1
Cited
DE3114421A1
Cited
EP0076054A1
Cited
EP0132337A2
Cited
US4825161A
Cited
US5022398A
Cited
US5038784A
Cited
US4881033A
Cited
US4973111A
Cited
US4896113A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1995-01-27 | 1995-07-29 | 1996-01-28 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1999-01-27 | 1999-07-29 | 2000-01-28 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2003-01-27 | 2003-07-29 | 2004-01-28 | $7700.00 | Window Closed / Lapsed |