Patent Application
20130274590
1. Field of the Invention
The present invention concerns the generation of a signal indicative of motion of a subject in a magnetic resonance (MR) apparatus, and in particular to the generation of such a signal for triggering an event in an MR-guided procedure.
2. Description of the Prior Art
Many types of procedures are implemented by interaction with an examination subject while the examination subject is located in a magnetic resonance apparatus, namely while the subject is located inside the data acquisition unit of such a magnetic resonance apparatus. One example of such a procedure is treatment with high intensity focused ultrasound (HIFU).
The principle of HIFU treatment is to concentrate a high acoustic intensity within a focal spot having a size of a few millimeters, in order to produce sharply localized mechanical or thermal effects. This treatment is implemented using an external source of energy, and a propagating ultrasound beam. When such a treatment is implemented to apply HIFU to a focal spot located in moving tissue, such as movement caused by respiration, such movement or motion of the organ in which the focal spot is located must be taken into account when dynamically steering the HIFU beam.
As described by Fischer et al. in “Focused Ultrasound as a Local Therapy for Liver Cancer,” Cancer Journal, Vol. 16, No. 2 (2010) pgs. 118-124, the primary challenges to HIFU therapy in the abdomen are to manage the complex motion of abdominal organs, and to prevent or avoid the risk of collateral heating at bone interfaces.
If the tissue motion management during HIFU sonication is not accurate, this may result in under-treatment of the target tissue, and/or unwanted collateral damage to healthy or critical surrounding anatomical structures.
It is known to guide a HIFU treatment procedure based on acquired MR images, which means that the HIFU therapy must then be administered to the subject while the subject is located in the examination volume of an MR data acquisition unit of the MR apparatus. In general, such MR-guided HIFU therapy requires motion encoding or motion monitoring using a detection device that is compatible with the radio-frequency fields and magnetic fields that are generated in the examination volume of such an MR data acquisition unit, and real time processing of motion information and feedback to the beam-steering system. In principle, this motion monitoring can be achieved by analyzing the magnetic resonance images themselves, but this requires compromises with respect to certain parameters (temporal resolution v. signal-to-noise ratio (SNR)), or image contrast.
One straightforward approach in this context is the use of respiratory gating. This means that HIFU sonication takes place periodically, during each quite or rest phase of the respiratory cycle, i.e., during exhalation. Respiratory gating generally increases the treatment time. Conventionally, some type of belt or respiratory cushion has been used to detect and monitor the respiration curve, or a volume and/or pressure signal from a mechanical ventilator can be used to determine the on/off sonication periods, with the patient being under general anesthesia.
Another known approach is to generate an atlas of motion fields during an initial learning phase of a control unit, based on magnitude data of temperature-sensitive GRE acquisition. This procedure is disclosed by deSenneville et al., in “Real-time Adaptive Methods for Treatment of Mobile Organs by MRI-Controlled High-Intensity Focused Ultrasound,” Magnetic Resonance in Medicine, Vol. 57, No. 2 (2007) pgs. 319-330. In this approach, the motion field of the most similar image in the atlas is used to correct the target position. Under the hypothesis of periodic motion, the focal point position for the next cycle is then estimated. This procedure, however, can only manage liver deformations caused by the periodic breathing cycle, and is not capable of handling the non-rigid liver deformations that are caused by intestinal activity or muscle relaxation, as noted by von Siebenthal et al. in “4D MR Imaging of Respiratory Organ Motion and its Variability,” Phys. Med. Biol., Vol. 52, No. 6 (2007) pgs. 1547-1564. In general, T2*-weighted MR magnitude data from a gradient echo sequence dedicated to fast MR thermometry generally lacks anatomical contrast. Another approach is to make use of a pencil-beam navigator, as described by Hardy et al. in “Rapid NMR Cardiography with a Half-Echo Mode-Method,” Journal of Computer Assisted Tomography, Vol. 15, No. 5 (1991) pgs. 868-874, and this could be used for motion compensation of thermometry imaging, and to provide information about the target motion to the HIFU system for focal spot adjustment. MR information-based real-time motion compensation, however, generally comprises spatial resolution, geometric distortion, and the precision of the MR thermometry, as reported by Ries et al., “Real-time 3D Target Tracking in MRI Guided Focused Ultrasound Ablations in Moving Tissues,” Magnetic Resonance in Medicine, Vol. 64, No. 6 (2010) pgs. 1704-1712.
A first attempt at ultrasound-based motion tracking during MR-guided HIFU was reported in phantoms undergoing periodic and rigid motion of a small amplitude, deOliveira et al., “Rapid Motion Correction in MR-Guided High-Intensity Focused Ultrasound Heating Using Realtime Ultrasound Echo Information,” NMR Biomed., Vol. 23, No. 9 (2010) pgs. 1103-1108. Continuous one-dimensional ultrasound echo detection along a direction parallel to the main axis of motion was used. This setup is not suitable for clinical application, because the external ultrasound imaging probe cannot emit ultrasound parallel to the axis of respiratory motion. Moreover, the local motion in liver is spatially dependent, so as one-dimensional projection would not be sufficient.
By contrast to images obtained with ultrasound, external images of the subject do not provide tomographic information from the interior of the body. External images, however, have the advantages of being acquired with a relatively simply implementation, with no need for finding an additional acoustic window as is the case for ultrasound imaging, and there is no sensitivity to the HIFU emission. An optical camera is not influenced by the ultrasonic waves, and is virtually insensitive to electromagnetic radiation from the HIFU hardware.
In the context of brain MRI, it has been recently suggested to track the patient using a camera located inside the patient-receiving opening (bore or tunnel) of the MR data acquisition unit using an in-bore camera and a checkerboard marker attached to the forehead of the patient. Such an approach is described in Forman et al., “Self-Encoded Marker for Optical Prospective Head Motion Correction in MRI,” Med. Image Anal., Vol. 15, No. 5 (2011) pgs. 708-719. This article describes the use of a self-encoded marker with each feature on the pattern being augmented with a 2D barcode, tracked by a single analog in-bore camera attached to the head MR-coil. Outside of the scanner room, the analog video signal is converted to a digital signal using a frame grabber. This technology has been used for the correction of fMRI data, but has not been used in the context of image-guided therapy. Motion correction with this approach encompassed a rotation of 18° around the principle axis of the cylindrical phantom in between two scans. After rigid registration of the resulting volumes, a maximum error of 0.39 mm and 0.15° in translation and rotation were measured, respectively.
An object of the present invention is to provide a method for taking motion of an examination subject into account in an MR-guided intervention procedure, wherein the effective administration of the therapy is dependent on the therapy occurring at a desired point in time with respect to motion exhibited by the subject
A further object of the present invention is to provide an MR apparatus to implement such a method.
The first object is achieved in accordance with the present invention by a method for MR-guided therapy administration wherein the therapy is administered to a patient inside of a magnetic resonance data acquisition unit, the patient exhibiting extracorporeally detectable motion, such as periodic motion due to respiration. While the examination subject is located inside the MR data acquisition unit, MR image data are acquired from the subject from which MR images are reconstructed that are used to guide the administration of the therapy in terms of appropriately identifying an intracorporeal site at which the therapy is to be administered. The site is located in an organ of the patient that is subject to the aforementioned motion. In accordance with the invention, the motion is detected by placing a digital camera inside the patient-receiving opening of the MR apparatus, and obtaining digital images with the camera that are then analyzed in a computerized processor to identify information therefrom indicative of the motion. This information is then used to generate an electrical signal at an output of the processor in a form that allows the therapy to be implemented at a desired point in time with respect to the motion.
The signal that is generated that is indicative of the motion may be a continuous signal, such as a continuous signal representing respiratory movement, or may be a trigger signal that is emitted upon the detected motion exhibiting a particular characteristic, such as a spatially-dependent characteristic.
The analysis of the motion represented in the digital images can take place using a suitable pattern recognition algorithm, comparison algorithm, amplitude detection algorithm, or any other appropriate image processing algorithm that is able to detect and track motion from an analysis of the successive digital images.
The therapy procedure may be, for example, the administration of HIFU, with the on-off times of HIFU being controlled dependent on the identified motion. Another example is MR-guided Acoustic Radiation Force Imaging (ARFI) wherein the data acquisition is triggered dependent on the respiratory cycle of the patient, as indicated by the detected motion, so that data acquisition takes place when the patient is exhibiting the least amount of movement, such as at the end of exhalation.
The second object noted above is accomplished by an MR imaging apparatus designed to implement the method described above, using a digital camera that is constructed with all of its components being formed of MR-compatible material, and being located inside the patient-receiving opening of the data acquisition unit of the MR apparatus, with appropriate RF shielding surrounding the camera inside the patient-receiving opening. The digital images obtained with the camera are communicated via a shielded cable.
The camera may be provided with a high-power light emitting diode (LED) that can be appropriately operated continuously or intermittently to provide sufficient light to obtain the optical images, when the digital camera is simultaneously operated to obtain those images.
A cylindrical gradient coil system 3 that is composed of three sub-windings is introduced into the basic field magnet 1. Each sub-winding is supplied with current by an amplifier 14 for generating a linear gradient field in the respective direction of the Cartesian coordinate system. The first sub-winding of the gradient field system generates a gradient Gx in the x-direction, the second sub-winding generates a gradient Gy in the y-direction and the third sub-winding generates a gradient Gz in the x-direction. Each amplifier 14 has a digital-to-analog converter that is driven by a sequence controller 18 for the temporally correct generation of gradient pulses.
A radio frequency antenna 4 is situated within the gradient field system 3. This antenna 4 converts the radio frequency pulse output by a radio frequency power amplifier 30 into a magnetic alternating field for exciting the nuclei and alignment of the nuclear spins of the examination subject or of the region of the subject to be examined. The antenna 4 is schematically indicated in
The radio frequency antenna 4 and the gradient coil system 3 are operated in a pulse sequence composed of one or more radio frequency pulses and one or more gradient pulses. The radio frequency antenna 4 converts the alternating field emanating from the precessing nuclear spins, i.e. the nuclear spin echo signals, into a voltage that is supplied via an amplifier 7 to a radio frequency reception channel 8 of a radio frequency system 22. The radio frequency system 22 also has a transmission channel 9 in which the radio frequency pulses for exciting the nuclear magnetic resonance are generated. The respective radio frequency pulses are digitally represented as a sequence of complex numbers in the sequence controller 18 on the basis of a pulse sequence prescribed by the system computer 20. As a real part and an imaginary part, this number sequence is supplied via an input 12 to a digital-to-analog converter in the radio frequency system 22 and from the latter to a transmission channel 9. In the transmission channel 9, the pulse sequences are modulated onto a high-frequency carrier signal having a base frequency corresponding to the resonant frequency of the nuclear spins in the measurement volume.
The switching from transmission mode to reception mode ensues via a transmission-reception diplexer 6. The radio frequency antenna 4 emits the radio frequency pulses for exciting the nuclear spins into the measurement volume M and samples resulting echo signals. The correspondingly acquired nuclear magnetic resonance signals are phase-sensitively demodulated in the reception channel 8 of the radio frequency system 22 and converted via respective analog-to-digital converters into a real part and an imaginary part of the measured signal. An image computer 17 reconstructs an image from the measured data acquired in this way. The management of the measured data, of the image data and of the control programs ensues via the system computer 20. On the basis of control programs, the sequence controller 18 controls the generation of the desired pulse sequences and the corresponding sampling of k-space. In particular, the sequence controller 18 controls the temporally correct switching of the gradients, the emission of the radio frequency pulses with defined phase and amplitude as well as the reception of the magnetic resonance signals. The time base (clock) for the radio frequency system 22 and the sequence controller 18 is made available by a synthesizer 19. The selection of corresponding control programs for generating a magnetic resonance image as well as the presentation of the generated magnetic resonance image ensue via a terminal 21 that has a keyboard as well as one or more picture screens.
The apparatus shown in
As also schematically indicated in
Although
As also indicated in
The MR-compatible digital camera 25 can be a consumer grade USB digital camera that has been made MR-compatible by removing any magnetic parts and adding the RF shielding 26.
As schematically shown in
The cable shown in
As an alternative to the arrangement shown in
The optical camera can be mounted to a non-magnetic orbital ring located in the patient receiving opening of the data acquisition unit 27, or alternatively can be connected to the HIFU platform.
Suitable triggers based on motion analysis of the images generated by the camera 25 can be implemented. Anatomical landmarks can be automatically set, or sharp edge-features in the optical region of interest and their displacement, can be tracked or followed using a calculation of the optical flow based on the iteratively Lucas-Kanade method in pyramids, as described by Lucas et al., “An Iterative Image Registration Technique with an Application in Stereo Vision,” Proceedings of the International Joint Conference on Artificial Intelligence (1981) pgs. 674-679. Implementation of the method in pyramids is described in Bouguet, “Pyramidal Implementation of the Lucas-Kanade Feature Tracker,” OpenCV Documentation, Intel Corp., Microprocessor Research Labs (1999). The optical data from the camera 25 can be processed at 30 fps online, with a 33 ms sampling time.
The output of the motion detecting algorithm can be a respiration curve that can be used to trigger a conventional DAC interface to the HIFU beam former substantially in real time, in order to dynamically adapt the HIFU beam steering. Alternatively, such a respiration curve can be supplied to the system computer 20 or the sequence control 18 of the magnetic resonance apparatus in order to trigger the acquisition of magnetic resonance data, such as with MR thermometry or MR acoustic radiation force imaging (ARFI), the latter being schematically shown in
As shown in
Optionally, multiple cameras can be used to acquire the 3D shape of a body region, such as the abdomen, using stereoscopic reconstruction, as described in Schaerer et al., Multi-Dimensional Respiratory Motion Tracking from Markerless Optical Surface Imaging Based on Deformable Mesh Registration,” Phys. Med. Biol., vol. 57 (2012) pgs. 357-373. Correlation of respiratory motion between the external patient surface, determined from the optical data obtained with the digital camera 25 and internal anatomical landmarks, obtained from fast dynamic MRI data, may be used for prospective motion compensation during MR guided HIFU treatment. Alternatively, the external patient surface can be reconstructed using a single optical device by fringe projection profilometry, as described by Price et al., “Real-Time Optical Measurement of the Dynamic Body Surface for Use in Guided Radiotherapy,” Phys. Med. Biol, Vol. 57 (2012) pgs. 415-436.
An advantage of the invention is that the image data acquisition is contact-free, and does not place any external obstacle in the HIFU beam entry window to the treatment site. In contrast to conventional mechanical sensors, such as an abdominal belt or a pressure cushion, that are operated in a user-dependent manner and may thus complicate the abdominal interventional procedures, the inventive approach is flexible and user-independent and enables a large field of view for the motion determination.
The available frame rate and resolution of commercial digital cameras is significantly higher compared to analog standards. For example, cameras with up to 500 fps with a 1,280×1,024 pixel CMOS image sensor are available. These features are advantageous for real time motion monitoring and correction. Moreover, digital devices are inherently more robust to EM noise and perturbation, even if the aforementioned RF shielding might in some instances be sub-optimal.
By contrast to belt or cushion-type respiratory sensors, which only provide a temporal curve, in accordance with the invention 2D or 3D dynamic images can be obtained so that mapping of the surface motion is feasible, for example. Moreover, some patients may exhibit primarily thoracic breathing while others may exhibit predominantly abdominal breathing, thereby requiring appropriate and accurate location of a mechanical sensor, which is not a factor in accordance with the present invention.
Moreover, the motion detection in accordance with the present invention does not modify the patient's respiration patterns in any manner as may occur with a mechanical sensor.
The method and apparatus in accordance with the present invention can provide direct measurement of distances without a need for an indirect conversion from other parameters, such as pressure/volume or force/displacement, as is necessary with abdominal belts and pressure cushions. The calibration and response are essentially linear in accordance with the invention. The correction for any geometric distortion of the image is easily achieved with a calibration board.
The inventive method and apparatus can enable establishment of a correlation between respiratory motion and the external patient surface, and internal anatomical landmarks, as described in Fayad et al., “Technical Note: Correlation of Respiratory Motion Between External Patient Surface and Internal Anatomical Landmarks,” Med. Phys., Vol. 38, No. 6 (2011), pgs. 3157-3164.
Since the practical implementation of the invention was substantiated using a commercial-grade camera, the method and apparatus will be economically implemented in clinical practice.
Although modifications and changes may be suggested by those skilled in the art, it is the intention of the inventors to embody within the patent warranted hereon all changes and modifications as reasonably and properly come within the scope of their contribution to the art.
