Patent Application
20080170654
The present invention relates generally to diagnostic imaging and, more particularly, to correcting motion errors in imaging data acquired from an object prone to motion.
Various imaging modalities are useful to image objects in or prone to motion, such as the heart in cardiac studies. For example, in computed tomography (CT), magnetic resonance imaging (MRI) and other imaging modalities directed to the acquisition of data from an object prone to motion, one or more motion correction techniques are generally used to reduce motion-induced artifacts in the reconstructed images. In known studies, this motion correction or compensation can add significant complexity in post processing of the images.
In one specific example, CT imaging requires measurement of more than 180 degrees of projection data to formulate an image. Because of various limitations in conventional CT scanners, the time necessary to collect a complete set of projection data is significant relative to object motion. For example, cardiac CT imaging is typically performed with the aid of an electrocardiogram (ECG) signal, which is used to synchronize data acquisition and image reconstruction with the phase of minimal cardiac motion. The ECG signal collected from the patient represents the electrical properties of the heart and is helpful in identifying the quiescent period of cardiac activity, which is preferred for data acquisition. Moreover, the ECG signal assists in identifying this quiescent period over several cardiac cycles. By synchronizing data collection with the quiescent period of the cardiac cycle, image artifacts and spatial resolution due to heart motion are reduced. Additionally, by consistently identifying this quiescent period in successive cardiac cycles, inconsistencies between images acquired at different cardiac cycles are reduced. ECG signals can be used similarly in MR and other imaging modalities. The ECG signal can gate acquisition of projection (known as prospective gating) or may be used subsequent to data acquisition(know as retrospective gating) to identify the phase of the cardiac cycle with minimum motion. Prospective gating allows dramatic reduction in dose administered to the patient since projection data is not collected during phases of the heart having significant organ motion.
The conventional ECG gating described above does not provide mechanical motion detection. That is, while an ECG signal can indicate that motion is occurring or is about to occur, it is a boundary measurement (electrical signals within the heart are measured on the surface of a patient) and does not provide accurate real-time placement data of the heart. Instead, mechanical motion of the heart must be inferred from the electrical activity measured in the ECG signal. Since actual mechanical motion, or displacement, of the heart contributes to sub-optimal image quality, cardiac images that depend solely on ECG signals often require significant post processing to correct for motion artifacts or often require a very high acquisition rate to minimize the extent of cardiac motion during acquisition.
CT reconstruction typically does not utilize a priori information on heart motion. In conventional ECG-gated cardiac CT studies, the heart is presumed to be a stationary object during the short acquisition period identified as the quiescent period in the acquired ECG signal (applicable to both prospective and retrospective gating techniques). Conventionally, half-scan weighting is used to suppress the impact of motion; however, its effectiveness is less than optimal since half-scan weighting combines CT projection data acquired at both extents the of angular range of data acquisition covering 180 degree plus the fan angle of the X-ray beam. The interpolation of projection data at both ends of the dataset remains constant and, therefore, does not change based on each particular acquisition as needed since there is no a priori information available. For data collected roughly in the center of the angular range of projection data acquisition, the data is treated in an identical manner without any weighting. Further, even with a gantry speed of 0.3 s/rotation, the central region of the projection range constitutes a slightly larger than 150 ms temporal window, which is prohibitively slow to sufficiently “freeze” cardiac motion. The data acquisition window for CT systems having dual tube-detector assemblies is typically between 70 ms and 80 ms during which heart motion may occur. Ideally, a temporal window of 20 ms to 40 ms is needed to adequately freeze cardiac motion.
It would therefore be desirable to synchronize data acquisition and image reconstruction with cardiac phase and motion data to acquire and reconstruct substantially motion-free datasets and images.
The present invention is directed to a method and apparatus for synchronizing data acquisition and image reconstruction with cardiac phase and motion data that overcome the aforementioned drawbacks.
Therefore, in accordance with one aspect of the present invention, a CT imaging system includes an ECG machine configured to output ECG data indicative of a cardiac cycle of a patient and an ultrasound machine configured to output ultrasound data indicative of measured torsional and deformational motion of a heart of the patient. The system further includes a CT imaging apparatus having a data acquisition module with a rotatable gantry having a bore therethrough, the rotatable gantry having an x-ray source and an x-ray detector disposed therein to emit one of a fan beam of x-rays and a cone beam of x-rays toward the patient and receive x-rays attenuated by the patient. The CT imaging apparatus further includes a computer programmed to receive ECG data in real-time, receive ultrasound data in real-time, acquire CT projection data, and reconstruct an image utilizing the acquired CT data, ECG data, and ultrasound data.
In accordance with another aspect of the present invention, a method of imaging includes the steps of acquiring ECG data, ultrasound data, and CT data from a subject. The method further includes the steps of categorizing the acquired CT data for a CT image reconstruction into one of an allowable CT data set and a non-allowable CT data set, and performing the CT image reconstruction using the CT data categorized into the allowable CT data set.
In accordance with yet another aspect of the present invention, a computer readable storage medium includes a computer program to control a CT imaging process. The computer program represents a set of instructions, that when executed by a computer, causes the computer to receive ECG data from an ECG machine, receive ultrasound data from an ultrasound machine, and acquire CT projection data by way of an x-ray source in a CT scanner. The set of instructions further causes the computer to and reconstruct a cardiac image utilizing the CT projection data, ECG data, and ultrasound data.
Various other features and advantages of the present invention will be made apparent from the following detailed description and the drawings.
The drawings illustrate one preferred embodiment presently contemplated for carrying out the invention.
In the drawings:
The present invention is directed to a method and apparatus for acquiring heart motion and heart phase data, and using this data to prospectively acquire CT imaging data of the heart. The heart motion data and heart phase data are also used for motion correction or compensation of the imaging data and for constraining a CT image reconstruction.
The present invention will be described with respect to a “third generation” CT scanner, but is equally applicable to other generations of CT system, as well as with other imaging modalities. Moreover, the present invention will be described with respect to an imaging system that includes a CT scanner that acquires image data, an ECG machine that acquires cardiac electrical data, and an ultrasound machine that acquires cardiac motion data from a patient. The CT scanner, ECG machine, and ultrasound machine are stand-alone devices that can be used independently from one another, but, as will be described, are configured to operate in tandem to acquire CT data, ECG data, and ultrasound data substantially simultaneously. It is also contemplated that the present invention is applicable with an integrated ECG/ultrasound/CT system.
Referring to
Rotation of gantry 12 and the operation of x-ray source 14 are governed by a control mechanism 26 of CT system 10. Control mechanism 26 includes an x-ray controller 28 that provides power and timing signals to an x-ray source 14 and a gantry motor controller 30 that controls the rotational speed and position of gantry 12. A data acquisition system (DAS) 32 in control mechanism 26 samples analog data from detectors 20 and converts the data to digital signals for subsequent processing. An image reconstructor 34 receives sampled and digitized x-ray data from DAS 32 and performs high-speed reconstruction. The reconstructed image is applied as an input to a computer 36, which stores the image in an electronic mass storage device 38. The image reconstructor 34 may be a separate entity as shown in
Computer 36 also receives commands and scanning parameters from an operator via console 40 that has a keyboard. An associated cathode ray tube display 42 allows the operator to observe the reconstructed image and other data from computer 36. The operator supplied commands and parameters are used by computer 36 to provide control signals and information to DAS 32, x-ray controller 28 and gantry motor controller 30. In addition, computer 36 operates a table motor controller 44, which controls a motorized table 46 to position patient 22 within gantry 12. Particularly, table 46 moves portions of patient 22 through a gantry opening 48.
Still referring to
Also shown in
Computer 36 is programmed to analyze acquired ECG data, ultrasound data, and CT image data, as well as reconstruct the data to achieve an optimal cardiac image. In
When analyzed, the acquired ultrasound data and ECG data provide several types of information. First, the ultrasound data is analyzed to determine a mean heart motion 75. That is, the ultrasound data acquired is analyzed over a period of time to examine heart position at each time during the cardiac phase and over a period of many beats. As the position of the heart varies according to both a positional shift, rotation, and/or torsional motion, a mean heart motion is helpful in determining a characteristic behavior that later can be useful to aid in the acquisition timing and data processing. Additionally, the ultrasound data and the ECG data are analyzed to determine 76 a real-time phase of the cardiac cycle.
The motion of the heart can be represented as the three dimensional trajectory in time of strategic points of the heart anatomy. Of particular significance is the trajectory of each point along the coronary vessels, the outer walls of the myocardium and the inner walls lining the ventricles and atria. Alternately the heart motion can be represented by the parameterization of a numerical model of the heart. The simplest heart model would capture features like the angular positions of the major axes of the heart (treating it like an ellipsoidal solid). In this way the positional, rotational, and torsional motions are captured by the parameterized motions of the angular positions of these axes. The mean motion of axes 75 is captured in the periodic angular variations and the real-time phase 76 of the cardiac cycle is represented by the real-time location along these mean variations. More realistic numerical models of the heart employ parameterized surfaces and volumetric solids that change shape and position over time with constraints representing the elastic and connective properties of the heart chambers, muscles and vessels. In either case the free parameters of the model are determined by comparison with the measured ultrasound data.
The ultrasound measurement generates multi-dimensional data of the imaged object over time which can be used to fit the heart model. There is a correspondence between the heart model at each phase location and the expected ultrasound data. This correspondence will be maximized when the model parameters are optimum. A maximum likilihood estimation is a method used in the art to perform this optimization (or fit) between the model and the data. Similarly, the ECG data represents an electrical voltage signal trace over time that is periodic. The mean periodic variation averaged over many heart beats is taken to indicate the mean motion 75 and the real-time voltage to indicate the real-time phase 76.
From the determination of the mean heart motion 75 and the real-time phase of the cardiac cycle 76, a CT data acquisition window is determined 77. The CT data acquisition window is representative of the quiescent periods in the determined real-time phase of the cardiac cycle that exhibit heart motion where the heart is relatively stationary. The magnitude of motion correction required is reduced if the CT acquisition window occurs during this “minimal-motion” phase of the cardiac cycle. During the CT data acquisition window, x-ray source 14 is enabled to administer x-ray beams 16 to the patient 22 while CT data is acquired 78. The acquired CT data, the ultrasound data, and the ECG data are stored to an electronic mass storage device 38 for later retrieval. After CT data has been stored, a determination is then made of whether the CT data acquisition window is still open 79. If it is 80, the process returns to acquire additional requisite CT data 78 required for CT image reconstruction. If the acquisition window is closed 81, a determination is then made whether additional CT data is required for image reconstruction 82.
The need for a complete CT data set before reconstruction is determined by whether a full range of gantry angles have been acquired. For a half-scan reconstruction, the angular range may be 240 degrees. Each acquisition window generates some data which can fill some of this angular region. When the full angular region is acquired upon several heart beats, then reconstruction can commence. This procedure can be implemented in software as an internal table of zero-valued entries, each entry for one degree of the angular range. Upon completion of a portion of the entire angular range, the corresponding portion of the table is filled with unity value. The full population of the table or a sufficient sum over the table would trigger the passage to a reconstruction mode. This approach is typical of acquisition protocols used with multi-sector reconstruction techniques. However, the acquisition window can be prescribed such that projection data suitable for image reconstruction is acquired during one acquisition, which is typical of acquisition protocols utilized with segment reconstruction techniques.
If more CT data (either to complete the projection data required for the current acquisition, or to acquired data for another section of cardiac anatomy) is required 83, the process returns to the start state 70 and repeats as described above. If it has been determined that sufficient data has been acquired 84, CT scanning is terminated, and image reconstruction, described below with respect to
As stated above, the reconstruction process 85 compares 90 a local section of an acquired heart motion with the mean periodic cardiac motion in the modeled motion curve and determines 92 whether the local section of an acquired heart motion falls within a pre-determined threshold of the modeled motion curve. In this manner, atypical heart beats characterized by atypical heart motion and/or an atypical cardiac phase may be identified and removed from image reconstruction.
An atypical heart motion is shown in
Referring again to
As a further constraint on image reconstruction, it is also envisioned that the phase of the cardiac cycle for an acquired heart beat is also compared 90 to the modeled curve. Referring again to
For those beats that are marked to be included in image reconstruction 100, the acquired ultrasound data and associated fit model corresponding to those heart beats is also used as a consistency condition for the CT image reconstruction. The ultrasound data acquired over these beats measuring heart motion and position are compared to a mean heart motion and position as determined by the ultrasound data acquired over many beats. Referring now to
Following displacement 103 of the CT data or if displacement is not needed 104, process 85 then determines 105 whether all acquired heart motions have been compared to the motion curve modeling mean periodic heart motion. If not all heart motions have been compared 106, process 85 returns to step 90 to compare additional acquired heart beats. If all heart motions have been compared to the motion curve 107, a CT image is reconstructed 108 from the marked heart motion data according to known image reconstruction techniques, where the reconstruction grid is deformed as needed on a view-by-view basis to properly capture the measured motion information.
The CT imaging process described above utilizes real-time acquisition of ECG and ultrasound data to prospectively gate CT data acquisition and identify typical and atypical heart beats for reconstruction purposes. Furthermore, the CT imaging process is able to deform and displace acquired CT data, i.e. deform the reconstruction grid used during backprojection process of CT image reconstruction, by comparing acquired ultrasound data representative of heart motion for an acquired local section to the mean periodic heart motion, thus providing a high resolution reconstructed CT image that minimizes image artifacts.
Additionally, the ultrasound data acquired in the imaging process described above may be used to produce a 3D image of the heart (i.e., the myocardial surface). In this manner, motion displacement and deformation of the CT data is obtainable for the entire heart surface. That is, for each heart wall/section, ultrasound data regarding heart position can be compared to the mean heart position for that wall/section. Upon combination thereof, a determination of a position variation of the entire heart surface from one beat to a mean heart surface position can be made. Likewise, the corresponding displacement of CT data for each of these heart walls/sections results in a CT image of the entire myocardial surface using both the mean motion signal and the temporally localized motion signal. Accurate displacement of the acquired CT data allows for reconstruction of an optimal CT cardiac image having fewer image artifacts and higher resolution of the heart surface.
Ultrasound is a real-time imaging modality and provides accurate and near-instant information on the mechanical state of an object in motion, such as the heart. In the context of cardiac imaging, ultrasound can provide real-time information as to the size, shape, and location of the heart when it is in diastole, systole, or other phases of the cardiac cycle. Moreover, using Doppler imaging techniques that are well known in the art, displacement information over the volume of the heart can be measured. When combined with simultaneous acquisition of ultrasound data and CT data, the ultrasound data provides information on the shape, location, and deformation, of the heart allowing motion induced artifacts to be reduced in the CT imaging reconstruction. Ultrasound can further provide data regarding a mean heart position for each phase of the heart and, when used in combination with ECG, provide synchronized mechanical motion data with the heart phase data and information on the quiescent period of cardiac activity. In this manner, the combination of ultrasound data and ECG data makes it possible to reduce x-ray dosage to a patient by prospectively gating CT data acquisition to those heart beats that are within certain pre-determined thresholds of heart motion and heart phase. X-ray dosage to a patient is further reduced by configuring the CT imaging system to extrapolate and displace acquired CT data, thus allowing for the CT imaging system to integrate a wider range of acquired CT data into image reconstruction.
Therefore, in accordance with one embodiment of the present invention, a CT imaging system includes an ECG machine configured to output ECG data indicative of a cardiac cycle of a patient and an ultrasound machine configured to output ultrasound data indicative of measured torsional and deformational motion of a heart of the patient. The system further includes a CT imaging apparatus having a data acquisition module with a rotatable gantry having a bore therethrough, the rotatable gantry having an x-ray source and an x-ray detector disposed therein to emit one of a fan beam of x-rays and a cone beam of x-rays toward the patient and receive x-rays attenuated by the patient. The CT imaging apparatus further includes a computer programmed to receive ECG data in real-time, receive ultrasound data in real-time, acquire CT projection data, and reconstruct an image from the acquired CT data, ECG data, and ultrasound data.
In accordance with another embodiment of the present invention, a method of imaging includes the steps of acquiring ECG data, ultrasound data, and CT data from a subject. The method further includes the steps of categorizing the acquired CT data for a CT image reconstruction into one of an allowable CT data set and a non-allowable CT data set, and performing the CT image reconstruction using the CT data categorized into the allowable CT data set.
In accordance with yet another embodiment of the present invention, a computer readable storage medium includes a computer program to control a CT imaging process. The computer program represents a set of instructions, that when executed by a computer, causes the computer to receive ECG data from an ECG machine, receive ultrasound data from an ultrasound machine, and acquire CT projection data by way of an x-ray source in a CT scanner. The set of instructions further causes the computer to and reconstruct a cardiac image utilizing the CT projection data, ECG data, and ultrasound data.
The present invention has been described in terms of the preferred embodiment, and it is recognized that equivalents, alternatives, and modifications, aside from those expressly stated, are possible and within the scope of the appending claims. Moreover, the present invention has been described in terms of medical imaging; however, the techniques described herein apply equally to imaging of inanimate objects.
