Patent Application
20150305700
This disclosure relates to the field of imaging, especially to the field of intraoperative, interventional and diagnostic imaging with hand-held detectors.
In diagnostic imaging, combinations of different methods are increasingly being used, the so-called hybrid imaging systems. Above all, systems are to be combined which are complementary to one another, and in which the disadvantages of one system are compensated by the advantages of the other, or the advantages of both systems can be combined, so that the new system is better than the single application.
For example, with X-ray computed tomography (CT), morphological structures can be represented with a high spatial resolution, with the disadvantage of relatively poor soft tissue contrast, lack of functional information, the use of nephrotoxic contrast agents and the presence of harmful gamma radiation (radiation exposure).
Although a nuclear medicine system also uses harmful radiation, it can represent functional information much better, however with a rather poor spatial resolution. A combination of a CT scanner and a nuclear medicine system (such as a single-photon emission computed tomography system, SPECT, or a positron emission tomography system, PET) could thus at least provide a high resolution image with functional information. There is of course the possibility to acquire the respective image data separately and to combine subsequently (image registration). However, this may have the disadvantage of poor correlation due to an intermediate deformation or movement of the patient body, and due to a differing positioning of the patient. Thus, the accuracy of such image data is very limited. Ideally, the imaging is performed by a combined system, such as is available through the so called PET/CT or SPECT/CT systems since several years. Thereby, a nuclear image and a CT scan are performed sequentially in a combined system. Under sequential recording, one refers to a recording where the recordings of both imaging methods do not overlap in time, but one after the other is carried out with a few minutes difference.
The combination of magnetic resonance imaging (MRI) and PET in an integrated system (MR/PET system) has shown that the disadvantages of CT (limited soft tissue contrast, nephrotoxic contrast agents, gamma radiation) can be prevented. The MR system can, on the one hand, produce very high spatial resolution images with an excellent soft tissue contrast. In addition, the application of magnetic resonance imaging by the wavelength and energy of the quantums employed is harmless. With nuclear medicine, this can now be provided with additional functional information of a specific area. Specifically in oncology, this not only helps with the initial diagnosis, but also in the follow-up control of a subsequent therapy.
Another advantage of MR/PET systems is that one can perform the data acquisition of MR and PET jointly in certain implementations. This prevents a poor correlation with intervening deformation or movement of the patient, and by a varying positioning of the patient. Under a joint recording, a recording is understood in which the recording of at least two imaging modalities overlaps during a time interval. For example, in an MR/PET, the PET images can be acquired before or after the MR recording, or even simultaneously with the MR recording. Similarly, the MR recording can be carried out prior to or after the PET recording, or simultaneously with it.
The problem of currently used PET/CT, SPECT/CT, or MR/PET systems is their exclusive use for diagnostic tasks, since they cannot be used intraoperatively or interventionally in a flexible manner.
The same is true for systems such as described in US Patent Application 2010/0016765 A1, or in U.S. Pat. No. 6,455,856 where an ultrasonic system is installed in a stationary SPECT system or in the stationary gamma camera.
In a diagnosis carried out and a subsequent interventional procedure or surgery of a tumor, although the image data would be available before the intervention/operation, changes that could occur after imaging and before the intervention/operation cannot be shown. Such changes may be effected by normal shifts of organs, varying positioning (e.g. Intervention/operating table is just, bed of the diagnostic system includes various positioning aids, diagnosis is carried out with the arms above the head, but the therapy with the arms at the side, or vice versa).
In addition, it would be desirable to have, even during the intervention/operation, information about the success of the operation (tumor tissue visually differs not necessarily of normal tissue), and to ensure that all diseased tissue has been removed but then to leave as much healthy tissue in the body. This can actually only be achieved in that the imaging is also available during the intervention/operation. In addition, one can see the picture information, of course, to take advantage of the intervention/operation volume being clearly labeled and to navigate the intervention/operation tool correspondingly exactly to this volume.
Many other therapeutic methods, among others, in neurology and orthopedics have similar problems, and could benefit from such a combined system in the direct operation environment.
The lack of mobility of the equipment is in itself a disadvantage in diagnostics. It is often desired in elderly or very ill patients, that image recordings (or even intervention) is performed at the bedside. Also, in smaller hospitals, equipment is often shared, and it is thus advantageous to be able to move these.
Another problem of these hybrid systems is the high cost of the machines. This allows their use only in large institutions or specialized centers. Thus, only a subgroup of all patients can be diagnosed.
As an alternative to this type of expensive permanently installed/stationary hybrid imaging systems, ultrasound systems have been upgraded to load and display PET/CT or SPECT/CT data in recent times. Current ultrasound systems are often upgraded with positioning systems in order to be able to display in real-time the location and orientation of the ultrasound probe. Via registration methods such as point-based registration or intensity-based registration, the CT data is associated with the ultrasound images and thus, over the known registration of PET or SPECT with CT and the calculated registration from CT to ultrasound, equivalent PET/Ultrasound and SPECT/ultrasound images may be generated. Examples of such systems are the GE Logiq E9 system, or the system of the European patent application EP 2104919 A2.
However, this strategy still bears problems: First, it needs a CT, which means a radiation exposure to the patient just for merging the ultrasound images with the nuclear medicine images. Secondly, the functional images (PET or SPECT) are often taken a few days before the ultrasound images, so that a fusion of the data is only partially valid. Thirdly, with the step of registering, these systems necessarily incorporate correlation errors that render a 100% fusion of the data impossible.
Relevant background to this invention is the freehand SPECT technology (T. Wendler, A. Hartl, T. Lasser, J. Traub, F. Daghighian, S. I. Ziegler, N. Navab; Towards intraoperative 3D nuclear imaging: reconstruction of 3D radioactive distributions using tracked gamma probes; Proceedings of Medical Image Computing and Computer-Assisted Intervention (MICCAI 2007), Brisbane, Australia, Oct. 29-Nov. 2, 2007, LCNS 4792 (2), pp. 252-260//T. Wendler, K. Herrmann, A. Schnelzer, T. Lasser, J. Traub, 0. Kutter, A. Ehlerding, K. Scheidhauer, T. Schuster, M. Kiechle, M. Schwaiger, N. Navab, S. I. Ziegler, A. K. Buck; First demonstration of 3-D lymphatic mapping in breast cancer using freehand SPECT; European Journal of Nuclear Medicine and Molecular Imaging, Springer Berlin/Heidelberg, 2010 August; 37(8):1452-61). It allows to generate SPECT equivalent images, but is based on hand-held detectors, such as a gamma probe rather than large gamma cameras that are used in conventional SPECT. These detectors can be located by a positioning system, so that their measurements are complemented by their position and orientation. From the measurements of the detectors, their respective positions and orientations, SPECT images are generated. The concept of freehand SPECT can also benefit from the use of coincidence detectors, wherein at least one of those is hand-held, and be extended to a freehand PET imaging. Freehand SPECT and PET systems are less expensive than their conventional stationary alternatives.
In 2009, a freehand SPECT system has been upgraded for additionally tracking an ultrasound probe (T. Wendler, T. Lasser, J. Traub, S. I. Ziegler, N. Navab; Freehand SPECT/ultrasound fusion for hybrid image-guided resection; Proceedings of Annual Congress of the European Association of of Nuclear Medicine—EANM 2009, Barcelona, Spain, October 2009). By calibration, it was possible to demonstrate the plausibility of a freehand SPECT/ultrasound imaging.
This prototype had the problem of sequentiality of data collection: first a freehand SPECT image was generated, and only then the ultrasound images were taken. The image quality was not particularly good, since deformation occurred and the patient moved between the two shots. Furthermore, the demands on the image quality of the freehand SPECT were high, so that an image reconstruction only based on the gamma-probe measurements and the position and orientation of the gamma probe was able to resolve resolutions of >7 mm, and could only make great contrasts visible.
Similar ideas were embodied in U.S. Pat. No. 6,512,943. Therein, an ultrasound system was mechanically coupled with two nuclear radiation detectors. From measurements of the nuclear radiation detectors, the depth of a radioactive source can be determined. With the ultrasound image that delivers the ultrasound system, one can then make an allocation which structures are radioactive, and even carry out biopsies. The main problem of such a system is that it can only be used in locating individual radioactive point sources, since it covers the 3D position of only two values.
In the U.S. Pat. No. 4,995,396, an endoscope integrates an optical camera, an ultrasound device, and a gamma camera. The system should be able to also represent the images of the gamma camera along with the images of the endoscope and ultrasound. This is in practice not possible, because the ultrasound images are sectional views in depth, and the gamma camera images are projected images.
In the U.S. Pat. No. 6,628,984, a hand-held gamma camera is described, which is tracked in order to reconstruct tomographic images. These images can be understood as 3D nuclear images such as freehand SPECT or PET Freehand, and can also be registered with ultrasound images, according to the inventors. This approach is basically the same that was implemented in the commercial ultrasound systems, which were discussed above. It captures the data sequentially, and suffers from problems with movements and deformations of the patient. Since this system is similar to the above-mentioned freehand SPECT system of the group of T. Wendler et al., it is also expected to provide only a low resolution and poor contrast in a real implementation.
In view of the above, a combined nuclear and ultrasound system for hybrid imaging according to claim 1, and a method for hybrid imaging according to claim 10 is proposed.
In the following, the invention will be described with reference to exemplary embodiments illustrated in figures, from which further advantages and modifications will readily occur.
In the following, various embodiments of the invention are described, some of which are exemplified in the figures. In the following description of the figures, like reference numerals refer to like or similar components. In general, only differences between the various embodiments will be described. Here, features that are described as part of one embodiment, can also readily be combined in connection with other embodiments, in order to produce still other embodiments.
In embodiments, among other things, a hybrid system of a nuclear radiation detection system and an ultrasound system are proposed, which have the benefits of nuclear medicine (functional diagnostics mainly by tumor foci, possibility of flexible and individual handling on hand-held detectors, no magnetic fields and thus use of sensitive electronic peripheral systems) and ultrasonic systems (excellent soft tissue contrast, very high spatial resolution, flexibility, low cost) combined on a common system. Both imaging modalities can be incorporated into a common data acquisition, and are connected, and thus the possibility for compensation of movement and deformation is provided, particularly in sequential recordings.
In the proposed system and method according to embodiments is the combination of a freehand nuclear medicine system (such as freehand SPECT or freehand PET) and an ultrasound system that use a common reference system. For this, a nuclear radiation detector, and an ultrasonic probe are functionally connected, wherein the position and orientation of the nuclear radiation detector and the ultrasonic probe are detected by a respective positioning system each, in real time. From the measurements (detected radiation and ultrasonic signals) of both systems and the information of the position and orientation of the radiation detector and the ultrasonic probe with respect to a common reference, 3D tomographic images of the distribution of radiation can be generated in a living being or object. These images are also visualized together with the ultrasonic signals in the form of a hybrid image.
A combination of nuclear detection and ultrasonic detection as outlined above allows functional data (such as freehand SPECT or freehand PET images) and anatomical images (such as generated from the ultrasonic signals) to acquire, reconstruct and visualize together in real time or quasi-real time. With such a system, one can perform the data acquisition and visualization in a single step, i.e. the images of different modalities can be recorded jointly within a few seconds of one another, or even simultaneously, when one ultrasonic probe and nuclear radiation detector can be used simultaneously or when both are incorporated in a probe. Furthermore, it is expected that the cost of implementing such a system are significantly lower than the cost of a PET/CT, SPECT/CT, or MR/PET.
Embodiments of the present invention are useful, by the included 3D image reconstruction (in the sense of freehand SPECT or freehand PET) to solve the problem, which is to display images of a gamma camera together or combined with the images of the endoscope and ultrasonic probe. In embodiments, the nuclear-image is a 3D image, which makes it possible to merge the ultrasonic sectional image and the otherwise projective nuclear information into a fusion or overlay. Fixed or stationary 3D nuclear imaging devices (such as SPECT, PET or their combinations with CT and MR) are generally less suitable for use in embodiments, since they are not flexible for intraoperative or interventional use, and also do not allow for common data acquisition of ultrasound and nuclear detector measurements due to their construction (as gantry).
In the context of this publication, the term “image” is understood to include Information on the distribution of a nuclear source and information from an ultrasonic signal which are included as a function of position for at least one image segment of the image. That is, an assignment H (x, y, z), wherein H is assigned to an area in space for the coordinates x, y, z, H comprises at least two values, and thereby at least one radioactivity density and echogenicity (a value of an ultrasound image).
In particular, the following aspects have to be considered in the implementation of embodiments of the invention. On the one hand the quality of the images of freehand nuclear imaging is highly dependent on the user. It requires therefore generally an evaluation system which is preferably continuously calculates the quality of the nuclear-image and, based on deciding whether the current database to display sufficient in accordance with defined or desired quality. Alternatively, and recommended is the use of a (e.g. visual or auditory) system instruction for the user, so that the user can optimize the quality of the resulting nuclear-image based on the instruction. To this end, a desired position and orientation of the nuclear radiation detector (20) are used for calculating at least one nuclear-image-quality-value, preferably in a continuous manner. Furthermore, based thereon an instruction may be provided/issued to the user, or optionally to a robot that guides the detector, and preferably continuously—in order to achieve a movement of the nuclear radiation detector to the desired position(s) and orientation(s), which permit/obtain the optimization of the quality value.
On the other hand, a non-one-dimensional motion and/or deformation of the object or living being (80) during the measurement act potentially has very negative effect on the quality of the obtained nuclear image. The present invention implements in some embodiments one or more fixing devices (82), so that the object or living being (80) fixedly remains stable during the measurement. In this way, the quality of the nuclear-image can be ensured.
The position and orientation of the nuclear detector (20), i.e. the nuclear detector coordinates (41) are detected by the nuclear-tracking system (40). These are usually a vector having a 3D position and 3 Euler angles. Alternatively, one can use a 4D quaternion, to describe the 3 Euler angles in a manner more numerically stable.
Both the detector measurements (21) and nuclear detector coordinates (41) are recorded by the data acquisition module (60). In an embodiment of the present invention the data acquisition module (60) can synchronize this data. Thus, each detector measurement can be assigned to a nuclear detector coordinate. A possible implementation of this is the use of tables, where all data will be stored and then mapping algorithms assign the closest nuclear detector coordinate to each detector measurement.
The nuclear detector measurements (21) and nuclear-detector coordinates (41) can be assigned to one another in a further embodiment of the invention by own timestamps of a common clock, of its own clocks with known timing differences, or after adoption of a known transmission delay. Alternatively, this data can be stored in a so-called ring buffer, with their own timestamps or new timestamps given by the data acquisition module (60), and are stored and assigned as needed. The “ring” in the name comes from the fact that old measurements are overwritten after a certain time period.
In a further embodiment of the invention, interpolation algorithms such as linear interpolation, or cubic interpolation algorithms, or algorithms using filters in the time domain, such as Kalman filter or particle filters, are employed for a better assignment of detector measurements (21) and nuclear detector coordinates (41).
On the side of the ultrasound, the ultrasound probe (30) delivers ultrasonic signals (31), such as, amongst others, linear measurements (A-mode ultrasound), 2D images (B-mode ultrasound), 2D Doppler images (normal Doppler, power Doppler, etc), Elastographic images or 3D images.
The position and orientation of the ultrasound probe (30) is also detected by a tracking system, namely, the ultrasonic tracking system (50).
The ultrasonic signals (31) and the ultrasonic probe coordinate (51) are also sent to the data acquisition module (60) and, in one embodiment of the invention, are synchronized with the nuclear detector measurements (21) and the nuclear detector coordinates (41).
In a further embodiment, the data acquisition module (60) preprocesses all the data from the data acquisition module (60), this data being designated together as a “full data” (61), such as by the use of filters to clear “Outliers” or eliminate various known noise signals from the complete data (61).
The complete data (61), whether preprocessed, synchronized or untouched, are then sent to the image reconstruction module (70). This module has in embodiments, multiple tasks:
1. It can determine a volume for image reconstruction. This may be predetermined, but can also be determined from the nuclear detector coordinates (41) and the ultrasonic probe coordinates (51), by accumulating which 3D positions were most often recorded by the nuclear detector (20) and/or the ultrasonic probe (30). For this, a calibration of the two hand-held parts is necessary in order to assign to where the field of vision of the two hand-held parts is directed, each relative to the elements tracked by the respective trackers. Other methods for image reconstruction can be found in the German application 102011053708.2 of one of the inventors of this invention.
2. It can use the information in the ultrasound signals (31) and the ultrasonic probe coordinates (51), e.g. for determining, in an embodiment, an attenuation map (or weakening card). A weakening card can be used later in the image reconstruction. The image reconstruction module (70) can in another embodiment calculate a scattering map from n ultrasonic signals (31) and the ultrasonic probes coordinates (51). Other tasks of the image reconstruction module (70) can be the segmentation of organs or parts thereof, the compensation of movements, etc.
3. It can be a nuclear image (71) from the nuclear detector measurements (21) and the nuclear detector coordinates (41) in consideration of the of the ultrasonic signals (31) and ultrasonic probe coordinates (51) information reconstruct (such as attenuation maps, scattering cards, segmentation of organs, estimated movements and/or deformations, etc.). To calculate the nuclear-image (71) the the nuclear image reconstruction module (70) can use conventional methods of image reconstruction, such as iterative image reconstruction methods. A pre-processing of the input data or a reworking of the reconstructed image data can also be implemented in the nuclear image reconstruction module (70). Examples of such preprocessing methods can be plausibility methods detecting non-plausible measurements, filtering techniques which smoothen potential noise in the recordings, information calculation methods that calculate the area where the calculating the nuclear-image has sufficient information, etc. Details on how such as a nuclear image can be calculated and how such filters can be applied, can be found in the German applications 102008025151 of a subgroup of the inventors of this invention. More details are also in the German application 102011053708.2 of one of the inventors of this invention.
Here, the nuclear radiation detector (20) is a conventional gamma-probe, the detected gamma radiation in the energy range 27-364 keV and has a lateral shielding, so that essentially only radiation from a narrow cone in the direction of the major axis of the gamma probe is measured. The nuclear measurements are counting rates in CPS.
The Nuclear Radiation Detector (20) is tracked by an optical passive localization system, here the design of the nuclear-tracking system (40). The nuclear tracking system (40) consists of a stationary tracking part (42), here e.g. two infrared reflectors on the nuclear detector (43) and on the ultrasonic probe (54).
This figure shows relatively clear how the different coordinates are converted to a common coordinate system, the coordinate system of the hybrid image. By calibration or mechanical drawings can be the transformation (eg., a 4×4 transformation matrix when using homogeneous coordinates) of the infrared reflectors in the nuclear detector (43) to the detector material determined by the nuclear radiation detector (24)—Transformation T1.
The transformation from the nuclear-tracking system (40) to the infrared reflectors on the nuclear detector (43)—Transformation T2—is determined in real time by the nuclear-tracking system (40).
Similar to the nuclear detector (20) from the nuclear tracking system (40), which is in this embodiment identical with the ultrasonic tracking system (50), tracked to the ultrasonic probe (30). For those with infrared reflectors (54) is upgraded. Thus, the transformation T3—are determined by the nuclear tracking system (40) to the infrared reflectors on the ultrasound probe (44).
By ultrasonic calibration methods such as a “Single-Wall-Calibration” or a grid-based calibration to transform the infrared reflectors on the ultrasound probe (54) to the plane of the ultrasound image (the ultrasonic signal (31) determine in this version)—transformation T4.
Thus, one can convert the nuclear detector measurements (21) and the ultrasonic signals (31) in any common coordinate system. There, they can then be used to reconstruct a hybrid image.
By calibration or algorithms for calculating the volume where the hybrid image is reconstructed (such as in the description of
The use of a reference (82) for the article or the animal brings advantages. For example, the object or living being (80) can move rigid without new data must be recorded. Furthermore, the tracking system also can move without losing the validity of the data collected up to that point.
The fixation device (82) can be selected from the group consisting of:
A) mechanical mounts,
B) Vacuum pillows,
C) on the surface or a part of a tailor-made rigid mask formed from the object or living being (80),
D) Materials which are fixed under electric, magnetic or electromagnetic fields, but in their absence are deformable,
E). Materials which are in a temperature or pressure range determined, but are outside of those deformable
F) or a combination of any of the above fixation devices.
In embodiments, a combined nuclear and ultrasonic system for hybrid imaging is proposed, which comprises:
In further embodiments, the nuclear tracking system and the ultrasonic tracking system are part of the same tracking system.
In further embodiments, the nuclear radiation detector and the ultrasonic probe are integrated in a hand-held probe.
Further embodiments comprise a fixation device for an object or living being.
In further embodiments the nuclear tracking system and the ultrasonic tracking system are each selected from the group consisting of:
Further embodiments comprise:
Further embodiments comprise:
Further embodiments comprise an interface for receiving:
Further embodiments comprise a surgical or interventional instrument which is tracked by the nuclear tracking system, the ultrasonic tracking system, or by both tracking systems.
A method for hybrid imaging is proposed in embodiments, comprising:
A method for hybrid imaging further comprises calculating the desired position and orientation of the nuclear radiation detector and the subsequent issuing, to a user or a robot, of an instruction to move the nuclear radiation detector to the desired position and orientation.
A method for hybrid imaging further comprises the use of at least one of the following information to reconstruct a nuclear image:
A method for hybrid imaging further comprises: tracking a surgical or interventional instrument.
A method for hybrid imaging further comprises: displaying the relation between the surgical or interventional instrument and the hybrid image, and/or navigating the surgical or interventional instrument(s) to a position in the hybrid image.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2012 111 386.6 | Nov 2012 | DE | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2013/074626 | 11/25/2013 | WO | 00 |
