The present invention relates to systems and methods for evaluating the elasticity of biological tissue. In particular, the invention relates to systems for performing elastography using CT imaging modalities.
Elastography is a noninvasive technique to estimate stiffness of soft tissue. Magnetic resonance elastography (MRE) and ultrasound elastography (UE) estimate tissue stiffness to diagnose different diseases such as staging liver fibrosis, cancerous tumors, etc. However, MRE has functional deficiencies in that it: (1) cannot be used to effectively evaluate hard tissues (e.g., bone); (2) is limited by low temporal resolution; and (3) requires long scan times. UE is similarly limited by its being inapplicable for evaluation of hard tissues, as well as by low spatial resolution precluding creation of spatial stiffness maps and by limited field-of-view for full-organ imaging.
The systems and methods described herein present a novel mechanism for performing elastography using computed tomography (CT) imaging modalities. CT elastography (CTE) is a displacement-based method that estimates local displacement of tissue, in response to a mechanical stimulus, from high resolution images. This “displacement map” can be used as a surrogate to estimate stiffness. Unlike both MRE and UE, CTE can be applied to hard structures in the body (such as bone, cartilage, and teeth), as well as soft tissues, in combination with high spatial and temporal resolution. Unlike MRE, CTE can also be used to scan patients that may not be compatible for the magnetic resonance environment from a safety standpoint (e.g., patients with pacemakers or metal implants).
In some embodiments, CTE is performed by applying external mechanical vibrations in an area of interest. In some situations (e.g., due to heart valve closures), internally induced vibration may be preferable. The induced vibratory waves are then tracked using a CT scanner over time. Finally, these images are analyzed using mathematical algorithms to estimate spatial and temporal displacement maps (a surrogate for stiffness estimate).
For the application of mechanical vibrations, CTE has advantages over MRE in that the mechanical driver/actuator used to induce tissue vibration can be constructed using metallic components for pneumatic drivers, hydraulic drivers and high-current piezo-electric drivers. Such drivers can induce high frequency vibrations beyond what is possible in an MRE environment, which are necessarily limited to constructions using non-metallic components. Furthermore, CTE can be used to develop high spatial resolution displacement maps, unlike UE.
In one embodiment, the invention provides an elastography system including a CT gantry, an x-ray source, an x-ray detector, a vibration-inducing actuator, and a controller. The x-ray tube and detector geometry are fixed opposing each other and rotate around the imaging subject. The vibration-inducing actuator is positionable in contact with the imaging subject during rotational movement of the gantry to cause vibration of tissue of the imaging subject. The controller is configured to rotate the gantry while acquiring CT data at first defined frequency and to induce vibration at a second defined frequency. The controller is also configured to synchronize the acquisition of the CT data with the induced vibrations. The first defined frequency and the second defined frequency are harmonically related such that the first defined frequency is an integer multiple of the second defined frequency or vice versa.
In some embodiments, the controller of the elastography system is further configured to acquire non-vibratory CT data by rotating the gantry without inducing any vibration and to subsequently acquire vibratory CT data by rotating the gantry while inducing vibration. For the latter acquisition, multiple temporal frames are collected to capture the different vibrational phases of the object. The non-vibratory CT data and the vibratory CT data are then compared to identify, on a frame-by-frame basis, a vibration-induced local displacement map in the CT data using, e.g., rigid or non-rigid registration. Alternatively, the vibratory CT data can be used without the non-vibratory CT data. In this case, the displacements maps are constructed by comparing the two consecutive frames of the vibratory CT data.
Other aspects of the invention will become apparent by consideration of the detailed description and accompanying drawings.
Before any embodiments of the invention are explained in detail, it is to be understood that the invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the following drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways.
In some implementations, the system controller 115 is only responsible for controlling the operation of the system components described above. In such implementations, CT image data acquired by the x-ray detector 105 is provided to the system controller 115 and stored to the memory 119 or an external memory such as, for example, a remote computer server or cloud computing environment. Analysis of the data are performed by a separate computer system and viewed by a medical professional. However, in other implementations, the data processing is performed by the system controller 115 using instructions stored on the memory 119.
The example of
Elastography is an imaging technique that maps the elastic properties of various tissues of an imaging subject. Vibration is induced in the imaging subject while image data are captured. The captured image data are then processed to detect changes in the tissue due to the induced vibration.
The vibration-inducing actuator device 113 is coupled to a vibration control system 111 that includes a pump 311 and a pump controller 317. The pump 311 can include, for example, a hydraulic pump or a pneumatic pump. In a hydraulic pump-based system, the pump 311 provides fluid to the upper portion 307 of the cylinder 301 through a first hydraulic line 313 while at the same time drawing fluid from the lower portion 309 of the cylinder 301 through a second hydraulic line 315. This causes the piston plate 303 to move linearly in a first direction (i.e., downward in the example of
To induce vibration in tissue of a subject, the vibration-inducing actuator 113 is placed in contact with the skin of an imaging subject 207 laying on the CT table 209 (see,
Further details of examples of vibration-inducing actuators for elastography are described in U.S. Publication No. 2013/0303882, entitled “Hydraulically-Powered System and Method for Achieving Magnetic Resonance Elastography,” and International Patent Application No. PCT/US2015/060717, filed Nov. 13, 2015 and entitled “Hydraulically-Powered and Hybrid Hydraulic-Pneumatic Systems and Methods for Achieving Magnetic Resonance Elastography,” the entire contents of both of which are incorporated herein by reference.
When an imaging subject 207 is properly positioned on the CT table 209, the system begins rotating the imaging gantry 109 (step 401). The CT system begins capturing non-vibratory CT data from the x-ray detector 105 at a first frequency fd (step 403). In some constructions, the frequency of CT data acquisition is defined by a first waveform generator or based on the hardware limits of the gantry rotation. The peak signal or some percentage of the peak signal is used to trigger the data acquisition at multiple time points (e.g., data can be captured after the prescribed delay of detected peak signal).
After sufficient non-vibratory CT data are acquired, the system activates the vibration-inducing system and induces vibration into the subject tissue at a second defined frequency fv (step 405). The system then captures vibratory CT image data while vibration is being induced (step 407). In some implementations, the vibration frequency fv is defined based on the first frequency fd such that fv is an integer multiple of fd(e.g., fd=X Hz; fv=X*Y Hz where Y is an integer). For example, if CT data are collected at a frequency of 2 Hz, vibration of the actuator may be set at 2 Hz (Y=1), 4 Hz (Y=2), 6 Hz (Y=3), and so forth. Alternatively, the vibration frequency fv and the first frequency fd may be defined such that fd is an integer multiple of fv. The specific frequency multiple of vibration frequency fv may be defined based on the type of tissue being examined or may be varied to evaluate how the elastic response of the imaging subject's tissue changes at various vibration frequencies. The vibration of the actuator is carefully controlled to synchronize the vibration frequency f with the data collection frequency fd and, in some constructions, external triggering (such as, for example, cardiac triggering) is used to regulate both data collection and induced vibration. As a result, multiple offsets of the vibrations in the tissue are captured. Because the vibration frequency is defined as a multiple of the rotation frequency (or vice versa), the period of each vibration and each rotation are automatically aligned, allowing segmented acquisition, which spans multiple gantry rotations.
After sufficient vibratory and non-vibratory CT data are captured and pre-processing is performed (e.g., reconstruction of three-dimensional CT models from the captured CT data). The vibratory and non-vibratory CT data are compared to each other to identify and evaluate the elastic response of the tissue of the imaging subject (step 409). In some cases, the comparison of the vibratory and non-vibratory CT data is used to isolate the vibrational component of the CT image data (step 411). After the vibrational component is identified, elastographic analysis can be performed either automatically by the system or with the aid and intervention of a medical professional (step 413). Furthermore, in some implementations, image processing techniques are performed to filter out vibrational noise from the vibratory CT data (step 415).
In some constructions, the “elastographic analysis” (step 413 of
The location of the tissue point is identified in both the non-vibratory CT image and in each vibratory CT image using two- or three-dimensional Cartesian coordinates (i.e., an X, Y and Z value). A displacement for the tissue point is then calculated for each vibratory CT image using the equations:
2D-Displacement2=Δx2+Δy2
3D-Displacement2=Δx2+Δy2+Δz2
where Δx is the difference between the x-coordinate of the tissue point location in the non-vibratory CT image (or of a reference point in the vibratory CT data) and in one of the vibratory CT images, Δy is the difference between the y-coordinate of the tissue point location in the non-vibratory CT image (or of a reference point in the vibratory CT data) and in one of the vibratory CT images, Δz is the difference between the z-coordinate of the tissue point location in the non-vibratory CT image (or of a reference point in the vibratory CT data) and in one of the vibratory CT images.
The displacement is calculated for each of a plurality of tissue points in the image data (step 507). In some embodiments, displacement is calculated for each pixel in the non-vibratory CT image. In other embodiments, an analysis resolution is defined such that only a sampling of tissue points is identified and analyzed. Once a displacement is calculated for each tissue point, a two- or three-dimensional displacement map is generated based on the non-vibratory CT image (step 509) where the color or amplitude of each point on the CT image is defined by the calculated displacement of the corresponding tissue point, which is used as a surrogate for stiffness measurement.
The specific techniques and methods discussed above are only some examples and other implementations are possible. For example, in some other embodiments the displacement map may be generated based on other techniques including estimating maximum displacement just from the raw CT data.
Furthermore, as noted above, some tissues (and some specific types of tissue irregularities) are better detected at different vibrational frequencies. Although the examples above discuss calculating displacement at only a single vibrational frequency, in some other implementations, the system may be configured to automatically and/or controllably vary the frequency of both data acquisition and induced vibration. By doing so, the system can identify a frequency where the displacement of different tissue points exhibits greater variation (i.e., a greater standard deviation) and generate a displacement map corresponding to a frequency that better differentiates the displacement of the tissue points.
Additionally, the examples above discuss synchronizing the frequency of the induced vibrations with a data acquisition frequency. In some implementations, the vibrational frequency and the data acquisition frequency can be synchronized without any adjustment to the speed or rotational frequency of the CT system gantry. However, in other implementations where the speed and rotational frequency of the gantry can be controllably regulated by the controller, the frequency of the induced vibration and the rotation of the gantry may be synchronized in order to provide the vibrational CT data as discussed above.
Thus, the invention provides, among other things, a system and method for performing elastographic imaging and analysis using CT imaging data by synchronizing the data acquisition frequency of a CT system with a vibrational frequency of a vibration-inducing actuator device. Various features and advantages of the invention are set forth in the following claims.
This application claims the benefit of U.S. Provisional Application No. 62/154,497, entitled “COMPUTED TOMOGRAPHY (CT)-BASED ELASTOGRAPHY, filed Apr. 29, 2015, the entire contents of which are incorporated herein by reference.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/US2016/027311 | 4/13/2016 | WO | 00 |
Number | Date | Country | |
---|---|---|---|
62154497 | Apr 2015 | US |