A method and apparatus for recording a three-dimensional image of a body part

The invention relates to a method and an apparatus for recording a three dimensional image of a part of a patients body according to the preamble of the independent claims.

Various methods and devices are known for recording three-dimensional images of a part of a patients body. Three dimensional images of this type are usually generated from a plurality of planar scans and information describing the spatial relation between these scans and/or between the patient and the scans. Known methods are e.g. based on the principles of computer tomography, NMR tomography, ultransonic analysis, positron emission tomography or X-ray techniques. Especially while recording extended areas it is important that the part of the body is fixed in place because movement may affect the quality of the image. However, immobilization of the position of a body part, such as a patients head, is usually difficult and/or uncomfortable for the patient due to invasive attachments of head rings or frames.

<patcit id="pcit0001" dnum="DE3447365A"><text>DE-A-3447365</text></patcit> teaches a method according to the preamble of claim 1, and an apparatus according to the preamble of claim 8.

It is therefore an object of the present invention to provide a device and a method of the type mentioned above that obviates these disadvantages at least partially.

This object is met by the independent claims.

By measuring movements, i.e. positional changes, of the body part during image acquisition, it becomes possible to correct the acquired image data. If the patient e.g. moves (especially turns) during acquisition, the image data are corrected such that all scanned images are transformed to the same coordinate system and can be combined with high accuracy.

Preferably, the movements or positional changes are detected optically, e.g. by affixing reference markers to the body part, which are observed by a camera system. However, it is also possible to detect the position of the body part directly by means of suited image processing techniques, i.e. without using reference markers. In any case, an optical position detection is preferred for its simplicity and high immunity to noise. But it is also possible to detect movements and positional changes by other means, e.g. ultrasonic triangulation.

Preferably, a reference device with suited reference marks is mounted to the body part to be measured.

If the measurement is performed on a patient's head, the reference device should be attached to at least one tooth of the patient's upper jaw. Further connections between the head and the reference device are not necessary. The reference device can be affixed to upper jaw by conventional dental impression material or by other means of fixation, such as clamps or dental prostheses for toothless patients.

Preferably, the reference device comprises a mouthpiece, which is non-invasively connected to the upper jaw in the manner described above, as well as a reference member attachable to the mouth piece for being detected by the positional detector. The reference member can be tilted in respect to the mouthpiece, which allows to establish an optimum orientation for each recording geometry.

In a preferred embodiment the relation of the reference markers (i.e. the coordinate system of the positional detector) in respect to the body part (i.e. the coordinate system of the three-dimensional image or the coordinate system of the imaging system) are stored together with the image. This makes it possible to position the patient later by means of the reference markers.

Further advantages, applications and preferred embodiments are described in the dependent claims and the following disclosure of an embodiment with reference to the figures. The figures show: <ul id="ul0001" list-style="none" compact="compact"><li>Fig. 1 a view of a topographic apparatus according to the present invention,</li><li>Fig. 2 a block diagram of the apparatus of Fig. 1,</li><li>Fig. 3 a reference device in the mouth of a patient,</li><li>Fig. 4 the reference device of Fig. 3, and</li><li>Fig. 5 a view of the device of Fig. 4 perpendicular to the mouth piece.</li></ul>

Figure 1 shows an application of the invention in a computer tomograph 1. Computer tomograph 1 comprises a measuring zone 2 for receiving the head of the patient 3. As will be described in more detail, the patient 3 carries a reference device 4 in his/her mouth, the three-dimensional position of which is monitored during the whole image acquisition process by means of three cameras 5.

As can be seen from Fig. 2, the three cameras 5 are connected to a 3-D detection unit 6. Tomograph 1 and 3-D detection unit 6 are connected to a processing unit 1, which assembles the scans from the tomograph into a three dimensional image.

As shown in Figs. 3-5, reference device 4 consists of a mouth piece 10 and a reference member 11. The mouth piece 10 comprises a U-shaped bite-down plate 12 connected to an L-shaped arm 13. A attachment bracket 14 is connected to arm 13 by means of two screws 15. Bracket 14 carries a rod 16 of reference member 11. Rod 16 is rotatably held in bracket 14 and can be fixed against rotation by means of two screws 17. This provides the possibility to use any scanning plane of the imaging system, such as coronal or sagittal scanning. Two lateral bodies 18 are arranged at the ends of rod 16, each carrying two fingers 19.

Fourteen infrared diodes are mounted to the lateral bodies 18 and the fingers 19. They are used as reference markers and their position can be measured by the cameras 5 and the 3-D detection unit 6. For this purpose, the diodes 20 are connected to 3-D detection unit 6 by means of cables 21 and are driven in multiplexed operation.

As can be seen from Fig. 5, the lateral bodies 18 are slightly inclined in respect to rod 16 such that not all of the diodes are arranged in the same plane. This increases the accuracy of the measured position of reference device 4.

Reference points 23 consisting of a material that can be detected by tomograph 1 are attached to the lateral bodies 18. Different positions are used to arrange the reference points 23 and the diodes 20 on the lateral bodies. The function of the reference points 23 is described below.

The apparatus works as follows:

Before image data acquisition, bite-down plate 12 is embedded in rapidly settling dental impression material and introduced into the patient's mouth. The patient bites bite-down plate 12 until the material is hardened. This provides a substantially rigid connection between reference device 4 and the teeth of the patient's upper jaw and thereby between reference device 4 and the patient's skull. A further connection between the reference device and the patient's head is not required.

Then rod 16 is positioned in bracket 14 in a position where the diodes 20 can be detected by the cameras 5 during the whole tomographic measurement. At least three diodes should be visible. Then rod 16 is fastened by means of the screws 15.

Now the patient is introduced into tomograph 1 and a plurality of conventionally scanned sectional images are acquired. For each scanned image, the position of the diodes and thereby the three dimensional orientation of reference device 4 relative to the camera system is measured.

The positions of both lateral bodies 18 can be measured independently for detecting a relative movement of the lateral bodies 18 and thereby a deformation of the geometry of reference device 4.

Furthermore, the movement of the table carrying the patient can be determined by infrared diodes 8 (Fig. 2).

In addition to this, at least one image is acquired showing at least three or more of the reference points 23.

This provides the following data: <ol id="ol0001" compact="compact" ol-style=""><li>1. All scanned images that show the reference points 23 allow the determination of the position of the scanning plane of the imaging system in respect to the camera system, i.e. the relation of the coordinate system of tomograph 1 in respect to the coordinate system of 3-D detection unit 6. (Alternatively, the position of the scanning plane of the imaging system in respect ot the camera system may also be known from an earlier calibration, in which case the reference points 23 are not required).</li><li>2. As mentioned above, the 3-D position of reference device 4 is recorded for each sectional image scanned by the tomograph. This makes it possible to calculate all translatory and rotatory movements of the patient in respect to the coordinate system of the tomograph, i.e. the relative position of the patient's head in respect to the tomograph is known. The corresponding information is stored for each scanned image.</li></ol>

The information stored under point 2 allows 3-D processing unit 7 to assemble the sectional images into a high quality and precise three dimensional image taking account of the patient's head movements.

The three dimensional image is stored together with information describing the relative position of reference device 4 in respect to the three dimensional image (i.e. indicating the relative position of reference device 4 in respect to the patient's head).

After the measurement, reference device 4 is removed from the patient's teeth. However, as long as the hardened impression material remains on the bite-down plate, the reference device can always be attached in the same position on the teeth of the patient's upper jaw. Since information indicating the relative position of reference device 4 in respect to the coordinate system of the 3-D image has been stored together with the 3-D image, the position of the patient's head can immediately be determined when the reference device is reattached. This can e.g. be used when the patient has to undergo surgical treatment several days after image data acquisition. The reference device is again inserted into the patient's mouth and the surgeon's 3-D detection system can determine its position. The same 3-D detection system can e.g. also be used for measuring the position of surgical instruments and for overlaying them with the 3-D image of the patient's head.

Preferably, mouth piece 10 and reference member 11 can be separated from each other such that after establishing the tomographic image reference member 11 can be used for other patients. The mouth piece adapted to the patient is retained and can be reconnected to the reference member during later surgical treatment. For this purpose, the parts are designed such that the relative position between reference member 11 and mouth piece 10 can be re-established accurately.

The concept described here allows the generation of three-dimensional images having high accuracy while it obviates the need to immobilize the position of the head (or any other body part) during image acquisition and surgical treatment. By means of the continuous measurement of the head position during acquisition, movements of the patient can be compensated for.

In addition to the infrared diodes mentioned above, a further set of diodes 9 (Fig. 2) can be attached to the tomograph for marking the scanning plane and for determination of its position. They can be used for measuring the inclination of the tomograph and/or for determining the relative position between cameras and the tomograph (i.e. the relation between the coordinate systems of the 3-D detection unit and the tomograph).

This basic concept can also be embodied using means different from those described so far. For example, it is possible to vary the design of reference device 4. Depending on desired accuracy of the 3-D detection unit, reference device 4 can be more compact and e.g. have the shape of a baby's comforter, which is attached to one or more teeth of the upper jaw.

Instead of using dental impression material for attaching the reference device, suitable clamps or other means, such as dental prostheses, can be used.

In the present embodiment, the position of reference device 4 is determined from the position of the diodes 20. It is conceivable, however, that no active components are arranged on reference device 4 and its position is determined by processing the images from the cameras 5. The position of the head can also be measured directly, without using reference device 4, by recording it using several cameras and calculating its position by image processing.

Determining the position by optical means is preferred because optical signals do not interfere with the operation of the tomograph. Furthermore, it is fast and e.g. allows to detect a movement during acquiring a single sectional image in a short time interval - if necessary, the scanned image can be corrected in its position and/or be re-scanned.

It is, however, possible to perform a position measurement based on non-optical methods, such as ultrasound triangulation or radio-triangulation.

In the present embodiment the invention is applied for a computer tomograph, but it can also be used in other 3-D imaging techniques of body parts, e.g. in NMR tomography, ultrasound scanning techniques, angiography, positron emission tomography, imaging nuclear medical methods, planar radiography etc.

In all these cases, the tomograph (the image acquisition device) forms a first detector, which is acquiring single scans during an extended time interval, from which scans the complete three dimensional images can be assembled. At the same time, as second detector, independent of the first detector, is used for continuously measuring and recording the position of the head (or any other body part to be viewed). The positional data retrieved in this way are used for correcting (if necessary) the data from the first detector and for generating the three dimensional image.