Surgery – Diagnostic testing – Detecting nuclear – electromagnetic – or ultrasonic radiation
Reexamination Certificate
2001-05-18
2003-03-25
Lateef, Marvin M. (Department: 3737)
Surgery
Diagnostic testing
Detecting nuclear, electromagnetic, or ultrasonic radiation
Reexamination Certificate
active
06539246
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to an apparatus and a method for the time-resolved and location-resolved evaluation of nuclear magnetic resonance signals acquired with the aid of nuclear magnetic resonance in order to detect activity changes in a life form, whereby physiological processes are stimulated in the life form by using at least two stimulation functions.
2. Description of the Prior Art
It has been determined that brain activities caused by stimulation can be detected in the cerebral cortex of human beings with a nuclear magnetic resonance tomography apparatus. A procedure producing and registering such brain activity is known as a stimulation experiment. Such stimulation experiments have been carried out with visual stimulation and with stimulation around the primary motor cortex, for example, by finger movement. Functional brain examinations also can be carried out by different techniques such as PET (positron emission tomography) or EEG. A significantly better topical resolution, however, can be obtained by nuclear magnetic resonance tomography.
An experiment that is typically carried out using nuclear magnetic resonance tomography is known as an BOLD-fMRI experiment. BOLD stands for Blood-Oxygen-Level Dependent (dependent on the oxygen content of the blood). Activity due to a stimulation in a tissue generates a temporary lack of oxygen in the blood surrounding the tissue. The organism detects this lack of oxygen. New oxygen is supplied via the surrounding blood vessels. Given a sudden activity in the tissue, the oxygen content is initially, temporarily, slightly reduced, followed by a long-term-decay overshoot of the oxygen content in the blood as a result of the automatic controller action of the organism. This chronological modification of the local oxygen content in the blood is measured and localized in BOLD-fMRI experiments (fMRI stands for functional magnetic resonance tomography) using a nuclear magnetic resonance tomography apparatus. As an example,
FIG. 4
herein shows the chronological course of such a local oxygen concentration in the blood.
Given the acquisition of data in realtime, the time resolution is also limited as a result of the limited pickup speed of the nuclear magnetic resonance technique. Therefore, there are suggestions to trigger the acquisition of data for the functional imaging by stimulations. Only one part of the raw data required for a complete image dataset is acquired per stimulation. Therefore, it has been suggested to synchronize the data acquisition with a periodic repetition of a task triggering brain activities. A similar method has already been used for “film pickups” of the heart movement.
A problem in functional imaging is to separate signal changes of other signal changes arising from stimulations or brain activities—caused by movements, for example. For this purpose, it has also been proposed to calculate a correlation coefficient for each pixel between the stimulation function and the received chronological signal curve. Periodically repeated stimulations that are separated by pauses are used as a stimulation function. Periodic stimulation functions, however, have a number of disadvantages:
Periodic disturbance processes (e.g heartbeat, respiration) cannot be separated from the activity signal and appear as “physiological noise.” Processes showing a delay of integral multiples of the repetition period cannot be correctly recognized either. Prolonging this experiment does not lead to a better noise suppression in any of these cases.
Periodic stimulation functions have a nonuniform frequency spectrum. Therefore, specific spectral components are not or only weakly excited by the stimulation. This introduces a systematic error into the system identification, i.e.; the determination of the parameters of a mathematical model.
German Patent 195 29 639 suggests a method for the time-resolved and location-resolved representation of functional brain activities of a patient. A stimulation function stimulates physiological processes in a patient. The stimulation function is nonperiodic and has as a few as possible secondary maxima. On the basis of a pulse sequence for exciting and reading out nuclear magnetic resonance signals, time-resolved and location-resolved nuclear magnetic resonance signals are acquired and are converted into bits of image information. Time-resolved and location-resolved activity modifications in the patient are detected by the chronological correlation of the so acquired bits of information with the stimulation function.
The required MR data must be acquired as fast as possible with respect to the time and location resolution. Fast pulse sequences therefore are primarily used. The fastest currently available MR imaging sequence is referred to as EPI (echo planar imaging) sequence. Other fast pulse sequences such as turbo spin echo sequences, FISP or FLASH sequences are possible as well.
A high-frequency pulse is initially emitted in the EPI sequence. A slice selection gradient is simultaneously generated, so that only one slice of the examination subject is excited, dependent on the frequency spectrum of the high-frequency pulse and on the intensity of the slice selection gradient. A positive sub-pulse of the slice selection gradient is followed by a negative sub-pulse, with which the dephasing caused by the positive sub-pulse is reversed again.
At the same time as the negative sub-pulse of the slice selection gradient, two prephase pulses are emitted in a phase encoding direction or a readout direction.
The readout gradient, with alternating polarity, is subsequently activated. As a result of the alternating sign of the readout gradient, the nuclear magnetic resonance signals are repeatedly rephrased and a signal arises under each sub-pulse of the readout gradient.
The signals are respectively differently phase-encoded because the phase continues to switch from signal to signal by small phase encoding pulses between the signals.
The signals are phase-sensitively demodulated and are digitized in a matrix or grid. The received digital values are entered into a row of a raw data matrix for each signal. In the fastest version of the EPI method, referred to as “single-shot EPI,” a sufficient number of signals are acquired after one single excitation in order to prepare a complete raw data set for an image. In a known way, the image can be acquired by two-dimensional Fourier transformation from the raw data matrix.
Not only spatial resolution but also a time resolution of the signals must occur for the functional imaging. For this purpose, the represented sequence is repeated as fast as possible, so that image data, which are allocated to different points in time, are successively received.
The smallest element of an image dataset is referred to as a pixel. A coarser resolution is generally sufficient for the functional imaging compared to conventional nuclear spin tomography images, such as a typical resolution of 256×256 pixels.
FIG. 5
herein shows the schematic process of the known method according to German Patent 195 29 639. The pulse sequence for exciting and reading out nuclear magnetic resonance signals AMR (k) and the stimulation function f
S1
(k) run independently of one another. Both have an effect on the system
51
comprised of a human being and a magnetic resonance tomograph apparatus, and are clocked by a control computer (not shown); the pulse sequence AMR (k), however, is not triggered by the stimulation function f
S1
(k). On the basis of the pulse sequence AMR (k), raw data sets SMR (k) are acquired and image data sets B (t), in turn, are acquired from said raw data sets by two-dimensional Fourier transformation
52
. A chronological signal curve is received for each element in the raw data matrix SMR (k), or for each pixel in the image data matrix B (t). A cross correlation
54
subsequently ensues between this signal curve SMR (k), or B (t) and the stimulation function f
S1
(k). The stimulation function f
S1
(k) was subjected to
Lateef Marvin M.
Merlader Eleni Mantis
Schiff & Hardin & Waite
LandOfFree
Method and apparatus for conducting an fMRI-BOLD experiment... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for conducting an fMRI-BOLD experiment..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for conducting an fMRI-BOLD experiment... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3062470