DE19608971A1 - Method and device for recording diagnostically usable, three-dimensional ultrasound image data sets - Google Patents

Description

The invention relates to a method according to the preamble of claim 1 and a device for performing this method according to the preamble of claim 9.

Such a method and such a device are known in principle (US 41 00 916). In this known method, individual ultrasound images are taken into account Position and orientation data from an ultrasound sensor system are determined, defined in terms of their position in space. A three-dimensional, dia Gnostically usable data record, which includes a tomographic image capture from one to the other searching volume, is not generated.

The invention has for its object to show a device with the free manual guidance of the ultrasound transducer a tomographic acquisition of an entire, too investigating volume or three-dimensional space through a three-dimensional Da rate is possible.

To solve this problem, a method according to the characterizing part of Pa executed claim 1 and a device according to the characterizing part of Claim 9 trained.

In the invention, the ultrasound transducer is to be examined along a path with free guidance volume for a tomographic image acquisition a three-dimensional data set generated from a large number of individual images in accordance with the tomographic image acquisition that consists of the entire volume to be examined and is itself made up of one Put together a multitude of image values.

The invention is described in more detail below with the aid of an embodiment purifies. Show it:  

Fig. 1 shows a schematic representation of an embodiment of the inventive device for generating diagnostically acceptable three-dimensional data for a tomographic image recording of a volume to be examined;

Figure 2 each stored in the image processing system A zelbilder in a schematic representation, together with graphical representations of the ECG cycle and the Respirati ons-cycle of a patient.

Fig. 3-5 is a flow diagram of the operation of the device according to the invention;

Fig. 6 is a schematic representation of the transformation of the raw data into a usable three-dimensional data set.

In FIG. 1, 1 is an image processing system which essentially consists of the computer 2 and the peripheral devices connected to it, such as keyboard 3, screen 4 and mouse control 5 , etc.

The following are assigned to this image processing system:

• - An ultrasound device 6 with an ultrasound head 7 ;
• - An electromagnetic position or position detection device or an electromagnetic sensor 8 with a transmitter 9 and a receiver 10 formed by magnetic coils;
• - An EKG device 11 with associated electrodes 12 ;
• a device 13 for registering the breathing position (respiratory cycle) with associated probes 14 .

15 denotes a patient whose body or body area forms a volume to be examined. The patient 15 is arranged on a couch 16 .

The ultrasound device 6 with the ultrasound head 7 is a device known to the person skilled in the art and used in medical diagnostics for generating ultrasound images, each of which represents a region of the volume to be examined, which is located in the image plane of the ultrasound head 7 that is released . On the ultrasound head 7 , which is manually released by the examining person (doctor), a holder 17 is preferably provided, on which the receiver 10 of the sensor system 8 is also located.

Due to the magnetic fields emitted by the transmitter 9 , which are detected by the receiver 10 or the coils there, the system delivers sensor data (position and rotation raw data) at its output, with which the position and orientation of the receiver 10 in space are accurate is defined by translation data in the X, Y and Z axes as well as by rotation data about these axes.

Such an electromagnetic sensor system is known to the person skilled in the art, for example as ISOTRAK II from POLHEMUS, One Hercules Drive, P.O.Box 560, Colchester VT 05446.

The EKG device 11 is one which is known to the person skilled in the art from medical practice or diagnosis and which, depending on the EKG cycle, supplies a changing voltage at an output, which voltage is shown in diagram a in FIG. 2 and has the typical positive and negative R peaks, between each of which an R interval is formed. The device 13 is also a device known to the person skilled in the art, which is used in medical practice or diagnostics to record the respiration cycle and, depending on this respiration cycle, supplies a changing voltage at the output, which is shown in diagram b of FIG . 2 is reproduced.

As is also shown in FIG. 2, in the image processing system 1, in addition to the image content, an associated image header is also stored in digital form in addition to the image content, which contains additional information, for example general information about the image, a measured value that reflects the respective phase position of the image determined in the EKG cycle, a measured value which determines the phase position of the respective image in the respiration cycle and the position and orientation data which determine the position of the image plane of the respective individual image in space.

In order to achieve a defined position or positional relationship of the respective image plane, the electromagnetic sensor system 8 is used, the receiver 10 , which has at least one receiver coil, is provided on the holder 17 of the freely guided sound transducer or ultrasound head 7 , in one predetermined spatial relationship active part or the image plane of this sound transducer.

When generating the position and orientation data (translation data for the X, Y and Z axes and rotation data about these axes), the spatial relationship between the image plane of the ultrasound head 7 and the receiver of the sensor system is then calibrated, ie under Calculated using special calibration parameters. It is then also possible, inter alia, to use the holder 17 with the receiver 10 on different ultrasound heads and to take these into account in each case by different calibration or conversion of the raw position data supplied by the receiver of the sensor system.

The calibration thus takes into account the geometric positional relationship between the receiver and the orientation of the active part (crystal) of the ultrasound head 7 or the plane (image plane) in which the image generated by the ultrasound head 7 lies.

When capturing the images, it is not the actual position of the respective image plane in space that matters, but it is sufficient for the tomographic image acquisition to detect the relative position of the individual image planes to one another. The transmitter 9 of the sensor system 8 can thus be positioned as desired in the area surrounding the device.

The sensor system 8 thus calculates the position and orientation values - defined by the three translation degrees of freedom (X, Y and Z axes) and by the three degrees of rotation - and ascertains this data to the image processing system, which results from this raw position and orientation data by calibration calculates the real position and orientation data that are assigned to the respective image content.

The individual images are delivered by the ultrasound system 6 in a predetermined temporal sequence (sequence) as an analog or digital image or video sequence to the input of the image processing system 1 connected downstream. When this system is switched on or activated, the images, which are, for example, fields or full images, are digitized in the image processing system 1 (in the case of an analog video sequence) or in digital form (in the case of a digital video sequence) and with the associated image header or the header data is stored digitally in a memory of the image processing system 1 (image acquisition).

The image acquisition on the image processing system 1 is triggered by the user of the device (for example by operating a foot switch). Likewise, the image acquisition is ended, for example, by the user or automatically by the image processing system 1 when the three-dimensional data record formed by the large number of individual images, each with an associated header, has reached a predetermined size.

If, for example, this predefined size is 30 megabytes, this means that with a size of each individual image of 256 × 256 pixels and with a gray value depth of 8 bits and with an indentation of 50 fields per second, the storage volume after approximately 8 seconds and after an indentation of 457 single images is filled and the image acquisition is then automatically ended by the image processing device. The use of image masking enables larger amounts of images to be drawn in by reducing the storage space required per individual image.

Since the image planes of the individual images are oriented very differently in space due to the free guidance of the ultrasound head, the individual images are initially stored as raw data as raw data set 18 during the image acquisition, ie in a data acquisition or acquisition phase ( FIG. 6). After completion of the image acquisition, a transformation of this raw data then takes place in the image processing system 1 using a special program on a common reference system, for example a cathesic coordinate system with the axes X ', Y' and Z '( FIG. 6), so as to be used later to obtain usable three-dimensional data set 19 about the volume examined.

This three-dimensional digital data record 19 is basically constructed so that it contains a large number of individual volume elements 20 (voxels), which in this data record 19 have a given position in relation to the common reference system X'-, Y'- and Z'- Have axis. Each pixel 20 'of the images forming the raw data 18 is sorted correctly in this transformation into a corresponding voxel 20 of this three-dimensional data set. If a certain area of the volume to be examined has not been recorded during the image acquisition, then no image value can be assigned to the corresponding voxel 20 during the transformation, ie the three-dimensional data record 19 would then contain gaps, which, however, can be filled out by interpolation.

But there is also the possibility that certain areas of the volume to be checked were randomly scanned several times during the image acquisition and that then certain image values 20 'are overdetermined. In this case, either an average value is formed for overdetermined image values in the image processing system 1, or only one image value is used after a selection routine.

The device according to the invention is also particularly advantageously suitable for tomo graphic image capture of moving organs, especially organs, their movement is directly related to the heartbeat (e.g. cardiovascular system), again with a free hand tion of the ultrasonic transducer.

Basically, there is the possibility here of generating a static three-dimensional data set from such a tissue and thereby avoiding geometric distortions which result from the movement, for example cardiac activity. In this case, the images are taken in by the image processing system 1 synchronously with the cardiac cycle or with the positive or negative R peak of the cardiac cycle of the person to be examined. The cardiac cycle or the R peak are determined by means of an EKG device 11 , to which the patient 15 is connected via probes 12 . The triggering causes each image to be taken in a precisely defined time window within the cardiac cycle, ie each image is taken in at a defined time interval (phase position) after the occurrence of the respective R peak used for triggering in the ECG cycle. With this, all the individual images stored in the image processing system 1 and forming the raw data are phase-synchronized with the heart cycle, so that these individual images can be converted into a three-dimensional, static data record 19 , the individual voxels 20 of which have no geometric distortions.

By changing the phase position of the time window for image acquisition with respect to the triggering R tip, three-dimensional tomographic data records 19 can be created from the examined volume, which reproduce this volume at different times in the EKG cycle.

With the device according to the invention, dynamic, three-dimensional data records 19 of the volume examined, namely of moving organs, etc., can also be produced. In this case, the images are not drawn into the image processing system synchronously with the cardiac cycle, but independently of it. Here, too, the measurement value which determines the phase position in the EKG cycle is also stored in the image header of each individual image 21 .

This three-dimensional, dynamic data set 19 , which in turn is obtained by transforming the raw data 18 onto a common reference system, consists, for example, of a large number of partial data sets 19 ', each of which captures the entire volume to be examined and in turn consists of a large number of volume elements 20 is composed, with the partial data sets 19 'representing the examined volume but at different times in the EKG cycle.

The creation of the dynamic, three-dimensional data record 19 takes place, for example, for a complete EKG cycle or even for only part of such a cycle, 19 images or image values from different ECG cycles being used for the generation of this data record, in such a way that after or during the transformation of the raw data 18, all existing image values from different EKG cycles are each correctly positioned in the voxels 20 of the partial data record 19 'corresponding to their phase position.

If there are no 19 'image values in the generation of the partial data sets because areas of the volume examined were not recorded during the image acquisition, such gaps can in turn be filled by interpolation by the image processing system. If there are several image values per voxel 20 , this over-determination is in turn eliminated by forming an average value or by rejecting certain image values or by combining both methods.

Three-dimensional data records 19 can also be created as static or dynamic data records of organs or parts of the body which are subject to passive movement, for example by breathing. For this purpose, the device 13 is coupled to the image processing system 1 and detects the breathing position or the respiratory cycle of the patient. This respiration cycle, which is shown in FIG. 2 in diagram b, consists of the expiration phase (exhalation phase) and the inspiration phase (inspiration phase). With this re spiraling cycle, for example with the tips A there, the image processing system 1 can be triggered or synchronized again in order to create a static three-dimensional data record 19 undistorted by the passive movement, in such a way that the image is only drawn in during a predetermined window in each case Breathing cycle takes place. In principle, the control can also be carried out in such a way that the image is only drawn in when the breathing position is not activated, ie there is no respiration signal.

In principle, a dynamic, three-dimensional data set can also be generated taking into account the breathing position, the respiration cycle then being used in the manner described above for the EKG cycle when the dynamic data set 19 was created.

Figs. 3-5 show a flow chart illustrating in more detail the operation of the apparatus to the invention OF INVENTION.

The Fig. 3 refers to the initial setting of the image processing system 1 in particular prior to the acquisition of the images (data acquisition). First, a decision is made as to whether the reference system, ie the spatial association between the image plane of the ultrasound head 7 and the receiver 10 , must be recalibrated, for example because of the use of a new ultrasound head 7 which has not yet been used in the system.

If a new reference system is necessary, an appropriate calibration is carried out. Is the use of a known reference system possible, this or this calibration tion can be used. This is followed by a definition of possible out election criteria. A dynamic data acquisition should take into account the cardiac activity a threshold value is set with regard to the size of the RR interval. These Setting is not necessary if dynamic data acquisition is not desired. It the decision is then made as to whether the breathing position should be taken into account. Is that the case, the threshold for the breathing position is defined, for example by on setting a maximum and minimum threshold value for the duration of a respiration cycle. There is no definition of the threshold value if the breathing position is not to be taken into account.

After this presetting, data acquisition or acquisition takes place according to FIG. 4. First, the ultrasound images supplied by the ultrasound head 7 are digitized. Then, in the case of dynamic data acquisition, the phase position from the EKG cycle and / or the phase position on the respiration cycle is determined, this determination being omitted if dynamic data acquisition or data acquisition taking into account the respiratory position is not desired.

In any case, the position is recognized each time the picture is taken. The captured image data are then stored together with the other data as raw data set 18 , the further data (measured value for the phase position in the EKG cycle, measured value for the phase position in the respiratory cycle and position data) being stored in the image header.

The data is then checked using the selection criteria made in the method step in FIG. 3. Data that does not meet these selection criteria will be discarded and removed.

The data acquisition is finished when the tomographic volume to be checked is complete is what happens either automatically through the vision system according to a time criterion determined or confirmed by the user.

If the tomographic volume is not complete, the method steps of FIG. 4 are carried out again.

FIG. 5 shows the process steps for post-processing or transformation of the raw data 18 in the data record 19th The transfer or transformation (correct sorting) of the individual image data 20 ′ follows first into the new, common reference system, which is, for example, a Cartesian coordinate system. Gaps are also filled by interpolation. These process steps are carried out using the calibration data.

If a dynamic data acquisition or acquisition is carried out, a Interpolation of temporal gaps, especially using image data different ECG cycles. If no dynamic acquisition took place, this Ver will not apply driving step.

The decision is then made as to whether the breathing position should be taken into account or not. If the latter is the case, then such data are eliminated and discarded for which breathing activity exceeds a predetermined threshold.

The data obtained are finally stored so that they are for a two-dimensional and / or three-dimensional representation and also for a two-dimensional and / or three-dimensional Dimensional quantification can be used.

Claims (16)

1. A method for generating a diagnostically usable three-dimensional image data set ( 19 ) using an ultrasound device ( 6 ) with a freely guided ultrasound head ( 7 ) for generating a sequence of a plurality of ultrasound images of the volume to be examined, below Use of an image processing system ( 1 ) connected to the ultrasound device ( 6 ), to which the sequence of the generated ultrasound images are supplied, using a position sensor system ( 8 ) which determines the position and orientation of the ultrasound head ( 7 ) and thus the spatial position the image plane of the respectively generated ultrasound image is determined, with reference to the three degrees of translational and rotational degrees of freedom, these position and orientation data of the sensor system also being transmitted to the image processing system ( 1 ) which uses image data from the ultrasound images and the position and orientation data the three-dimensional len, the volume examined by tomographic data record ( 19 ) is generated, characterized in that an electromagnetic system is used for the three-dimensional tomographic image acquisition of a volume to be examined as a sensor system ( 8 ), whose receiver ( 10 ) on the ultrasound head ( 7 ) or on a holder ( 17 ) attached to this head ( 7 ) is provided that the ultrasound images in a phase forming a data acquisition are first stored in the image processing system ( 1 ) as raw data each time together with an image header which contains at least the position and orientation data, and that the image processing system ( 1 ) transforms the raw data or the raw data set ( 18 ) into the three-dimensional data set ( 19 ), in which all the image data or image values ( 20 ) are based on a common reference system, preferably on a Cartesian coordinate system. 2. The method according to claim 1, characterized in that the receiver or the receiver gerspulen this sensor system on the ultrasound head ( 7 ) or on the ultrasound head be fastened holder ( 17 ) are provided, and that the image acquisition system ( 1 ) provided by the sensor system Position and orientation data calibrated using parameters that take into account the spatial relationship between the receiver ( 11 ) of the sensor system and its orientation in space and the position of the image plane of the ultrasound head ( 7 ). 3. The method according to claim 2, characterized in that the calibration of the position and Orientation data in post-processing after the data acquisition phase phase. 4. The method according to claim 2, characterized in that the calibration of the position and Orientation data takes place during the data acquisition phase and that preferred the calibrated position and orientation data are stored in the respective image header will. 5. The method according to any one of the preceding claims, characterized by an ECG device connected to the image processing system ( 1 ) for generating a signal defining an EKG cycle, in particular a signal representing the R peaks of the EKG cycle, wherein the image processing system ( 1 ) Means has a value in the image hea which represents the phase position between a defined point in time of the EKG cycle, for example the R peak, and the insertion of an ultrasound single image on the image processing system. 6. The method according to any one of the preceding claims, characterized in that for generating a three-dimensional, static data set, the drawing in of the ultrasound images on the image processing system ( 1 ) takes place synchronously with the ECG cycle in each case in a given time window of this cycle. 7. The method according to any one of the preceding claims, characterized in that the generation of a three-dimensional dynamic data record ( 19 , 19 '), the retraction of the ultrasound single images on the image processing system ( 1 ) takes place continuously, and that the image processing system has means to the image values in each case according to the phase position in the heart cycle in the three-dimensional data set ( 19 , 19 '). 8. The method according to any one of the preceding claims, characterized in that a device ( 13 ) is used for detecting the respiration cycle, and that a measured value which determines the phase position in the respiration cycle is stored in the image header. 9. Device for generating a diagnostically usable three-dimensional image data set ( 19 ) with an ultrasound device ( 6 ) with a freely guided ultrasound head ( 7 ) for generating a sequence of a plurality of ultrasound images of the volume to be examined, with one the ultrasound device ( 6 ) connected image processing system ( 1 ), to which the sequence of the generated ultrasound images are fed, with a position sensor system ( 8 ) which determines the position and orientation of the ultrasound head ( 7 ) and thus the spatial position of the image plane of the respectively generated Ultrasound image determined, namely with respect to the three translation degrees of freedom and the three degrees of rotation, these position and orientation data of the sensor system are also transmitted to the image processing system ( 1 ), which from the image data of the ultrasound images and the position and orientation data the three-dimensional, that examined Vo Lumen to mographically recording data record ( 19 ), characterized in that for three-dimensional tomographic image acquisition of a volume to be examined, the sensor system ( 8 ) is an electromagnetic system that the receiver ( 10 ) of the sensor system on the ultrasound head ( 7 ) or on a holder ( 17 ) on this head ( 7 ) it is provided that the ultrasound images in a phase forming a data acquisition are first stored in the image processing system ( 1 ) as raw data in each case together with an image header which contains at least the position and orientation data, and that the image processing system ( 1 ) has means to transform the raw data or the raw data set ( 18 ) into the three-dimensional data set ( 19 ), in which all image data or image values ( 20 ) are based on a common reference system, preferably on a Cartesian Ko ordinate system are related. 10. The device according to claim 9, characterized in that the receiver or the Emp catcher coils of this sensor system on the ultrasound head ( 7 ) or on the ultrasound head-carrying holder ( 17 ) are provided, and that the image acquisition system ( 1 ) has means for the to calibrate the position and orientation data supplied by the sensor system using parameters which take into account the spatial relationship between the receiver ( 11 ) of the sensor system and its orientation in space and the position of the image plane of the ultrasound head ( 7 ). 11. The device according to claim 10, characterized in that the calibration of the position and Orientation data in post-processing after the data acquisition phase phase. 12. The apparatus according to claim 10, characterized in that the calibration of the position and Orientation data takes place during the phase of data acquisition, and that preferred the calibrated position and orientation data are stored in the respective image header will. 13. Device according to one of the preceding claims, characterized by an EKG device connected to the image processing system ( 1 ) for generating a signal defining an EKG cycle, in particular a signal representing the R peaks of the EKG cycle, wherein the image processing system ( 1 ) has means in the image header a value that represents the phase position between a defined point in time of the EKG cycle, for example the R peak, and the insertion of an ultrasound single image on the image processing system. 14. Device according to one of the preceding claims, characterized in that for generating a three-dimensional, static data set, the drawing in of the ultrasound images on the image processing system ( 1 ) takes place synchronously with the EKG cycle in each case in a given time window of this cycle. 15. Device according to one of the preceding claims, characterized in that for generating a three-dimensional dynamic data set ( 19 , 19 ') the retraction of the ultrasound individual images on the image processing system ( 1 ) takes place continuously, and that the image processing system has means for the image values in each case according to the phase position in the heart cycle in the three-dimensional data set ( 19 , 19 '). 16. Device according to one of the preceding claims, characterized in that a device ( 13 ) is provided for detecting the respiration cycle, and in that a measured value which determines the phase position in the respiration cycle is stored in the image header.