Method and system for determining the SID and the thickness of a patient in a

radiographic system

Technical Field

[001] The invention generally relates to the measurement of the SID (source- image-distance) and the thickness of a patient for setting the exposure parameters in a conventional radiography configuration, wherein the radiography detector is positioned against one side of the patient, but the detector surface is obscured by the patient when seen from the viewpoint of the X-ray source position.

Background Art

[002] It has been a longstanding problem in conventional radiography to be able to accurately determine the optimal exposure settings for a selected exam, given that the age and gender of a patient are known. In order to be able to calculate the optimal exposure settings (being kV, mAs & filtration), the acquisition geometry of the radiography system and the thickness of the patient has to be known. The patient thickness parameter determines the attenuation and scatter of the X-ray beam before hitting the imaging detector, and therefore has a significant impact on the overall image quality of the acquired image. The term "acquisition geometry" has to be understood as the set of parameters which determine the physical characteristics of the X-ray source (such as the collimator settings) and the geometrical relationships between the X-ray source and the patient or imaging detector (such as for instance the SID (source image distance)).

[003] In conventional radiography, the beam of an X-ray source (in most cases an X-ray tube) is modified or shaped to optimally expose the patient tissue subject to examination to render an optimal image quality result, while minimizing the patient exposure to radiation. The part of the X-ray beam exposing the region of interest of a patient is partially attenuated by the tissue encountered by the beam on its path to the imaging detector, and forms a latent image that is accumulated (CR-image) or integrated (DR- image) in a detector sensitive to X-rays. So-called exposure settings fully determine the quality of the X-ray beam and have an important impact on the resulting image, but similarly have an impact on the absorbed dose by the patient. The exposure settings are essentially determined by the type of exam to be carried out, and depend on the following: type of

examination (soft tissue examination requires different settings than the examination of bony structures), age of the patient (pediatric or not), gender of the patient, specific acquisition geometry SID (source image distance), ODD (object detector distance), incidence angle of the beam and thickness of the patient. The exposure settings are defined in terms of mA (electric current passed through the anode of the X-ray tube), s (exposure duration in seconds), and kV (tube voltage in kilo-volts).

In fixed radiography installations, the acquisition geometry is relatively easy to be determined since the degrees of freedom of the movements of the different modality components are defined by the modality design and the movements of the individual components can be easily tracked by measuring their displacements. The exact locations in space of the main determining components for the acquisition geometry (namely the X-ray source, the image detector and table surface) can be relatively easily calculated by tracking or measuring the movements of those components relative to reference positions. Digital readings of the displacements or rotations of a C-arm gantry can unambiguously define the emission point and inclination angle of the X-ray beam, for instance. In a typical fixed radiography installation, also the location of the image detector can be unambiguously located because of the fact that it resides in a so-called detector "bucky" of which the location in the table is predetermined or can be easily measured.

The situation is different for older radiography equipment that does not have digital position readings on all the components, and for mobile radiography devices or for examination types that require the detector to be outside the bucky. Mobile X-ray devices are used as a versatile solution for acquiring radiography images under circumstances where a patient cannot be easily transported to a dedicated X-ray room, or cannot be positioned easily. Mobile X-ray devices are used in emergency rooms, in interventional setups, or in cases where the patient needs an

examination in his hospital bed. It is clear that under these circumstances it is a lot less obvious to accurately determine the acquisition geometry when the relative position information between the different components, such as the X-ray source and the image detector, is not available.

[006] Especially, the determination of the exact position and orientation of an image detector relative to the X-ray source is challenging since the two objects typically do not have a mechanical relationship, and may even not be in a line of sight of each other. Also, the thickness of the patient cannot be directly derived from easy measurements, as it requires at least the determination of the relative position of the image detector and with respect to the incidence point of the X-ray beam into the patient. One has to assume that the object or patient to be imaged obscures the image detector when looking from the source perspective. Therefore, distance- measuring techniques that require a clear line of sight cannot be applied. They however can be applied for determining the distance between the X- ray source and the patient surface.

[007] In the art, different distance measuring techniques relying on a variety of effects have been proposed as a solution. Some of the solutions rely partly or entirely on the integration of kinetic sensors, which can record and calculate displacements based on the acceleration measurements, or angulations with respect to -for instance- the gravity force. These solutions require complex calculations of the recorded movements and are not sufficiently accurate for the application described here as the sensor may drift over time. A solution based on this principle also always requires a calibration step, into which the object to be tracked has to be registered in a reference position. Additionally, the type of kinetic sensors used in this kind of application have certain intrinsic limitations that do not allow certain parameters to be measured. For instance, acceleration sensors are not able to detect a rotation of an object oriented in plane parallel with the ground surface (perpendicular to the force of gravity). As an example, US patent US2014376700 (Samsung Electronics Co. Ltd.) proposes a solution to align at least the orientations of the X-ray tube and the detector using such angle measurement sensors.

[008] In the same disclosure, also magnetic field sensors are proposed to

measure and detect a relative position of an X-ray image detector (to which they are attached) relative to a magnetic field generator which is connected to an X-ray source assembly. The magnetic field generator generates a static magnetic field, of which the intensity is measured at the positions of the magnetic field sensors and brought in relation with the relative distances between the magnetic field sensors and the magnetic field generator (the latter is positioned close to the X-ray source). The magnetic field is approximately inversely proportional to the third power of the respective distance between the magnetic field generator and the individual magnetic sensors. The measured magnetic field intensity values are thus indicative of the distances between the magnetic field generator and the magnetic sensor. The magnetic field measurements are used in combination with the acceleration sensors in order to obtain a relative location and relative orientation of an image detector, which allow, in combination with the determination of a start location, to calculate absolute distances from the X-ray source position. This solution is however not very accurate, and is very susceptible to external influences which may influence the magnetic field such as the presence of metallic objects or the presence of other static or non-static magnetic fields, such as the earth's magnetic field or the presence of electromagnets or coils in the vicinity of the detectors.

[009] US9179886 (Carestream Health, Inc.) discloses an approach whereby a method for alignment of an image detector with the beam of an X-ray source is also based on magnetic principles. In the disclosure, a time- varying magnetic field pattern is generated at a position connected to the X-ray source, so that this signal can be picked up and measured by a number of magnetic sensors (at least 2) which are connected to the imaging detector. The magnetic field pattern is generated at a predetermined frequency chosen so that the signal is made transparent for human tissue (i.e. that the energy is not absorbed by human tissue). The advantage of the alternating magnetic field is that the amplitude

measurements allow to compensate for any present static magnetic fields such as the earth's magnetic field. The time-varying signal is picked up and measured at multiple spatially distributed magnetic sensor elements (or coils) which are arranged at fixed positions of an image detector, and of which the combined read-outs induced by the generated magnetic field are indicative for the location of the detector in relation to the magnetic field generator. The technique is thus an improvement over the previously mentioned solution in that it is less susceptible to external magnetic disturbances, but at the same time, it is clear that the technique does not intend (and does not achieve) to obtain absolute location measurements of the sensors attached to the detector. The disclosure achieves a more reliable estimation of the relative positions of all detectors in relation to the magnetic field generator. By comparing the measured values against a set of reference values, the method is capable of providing an indication of the alignment of the image detector with the X-ray source location.

[0010] An important aspect that contributes to the accuracy of the system is that the different magnetic sensors are differently aligned, although only in the same plane as the image detector. At least two magnetic sensors are required, and are preferably aligned under at least 45° of each other.

Additional magnetic sensors may be added to further improve the accuracy of the method. Additional detectors are preferably arranged in the same plane but under different offset angles.

[0011] In another disclosure (US7581885) the idea of using a set of 3 GPS

sensors built into the corners of an encasement of an imaging detector is used to perform the absolute GPS-localization of the 3 sensors

(measurements which rely on trilateration techniques for determining the individual position of the GPS sensors), using the satellites of the GPS. In the document, the idea is raised that once the exact locations of the 3 sensors in the corners of the imaging detector are known, also the relative position of the imaging detector with respect to the X-ray source can be calculated. Today, it is however generally known that the accuracy of the (civil) GPS is insufficient for this intended application of locating an object on the sub-centimeter level, as the standard horizontal accuracy of a civil GPS receiver under optimal signal receiving conditions is 3.5 meters. The most optimized GPS location enhancement techniques (such as RKP used in the mining industry) result in an accuracy of 4 cm. Especially when it is generally known that many external factors can degrade the GPS positioning accuracy (such as signal blockage in buildings, or signal reflections against walls), it is clear that the accuracy of the proposed system will especially suffer under the operating conditions of such an X- ray imaging detector alignment system; namely inside a hospital building.

[0012] In summary, a number of solutions have been described in the art which at best resolve one of the partial problems encountered when looking out for a full and accurate solution to determine the position and orientation of an image sensor with respect to the position of an X-ray source. Many good and accurate solutions exist in case that direct line-of-sight distance measurement techniques can be used, but only partial solutions exist when the visual path between the source and the detector is blocked. The most promising techniques evading the direct-line-of-sight limitation are based on magnetic field measurement techniques, but so far, only estimations or relative measurements with limited accuracy and reliability have been achieved. It is clear that the problem of measuring the thickness of a patient requires an accurate measurement of the position of the image detector and an accurate method of measuring the location of the incidence point of the X-ray beam on the patient (assuming that the detector is positioned against the back of the patient). The measurement of the incidence point can be performed using methods known in the art that require clear line of sight for their measurement (such as used in optical rangefinders, laser distance measuring, tape measurements or alike).

Summary of invention

[0013] The invention provides for a method for determining the source image distance (SID) in a radiographic system, the method comprising the steps of sequentially generating a signal in at least three spatially distributed signal generator arrays which are spatially associated with an emission point of an X-ray source of the radiographic system, simultaneously measuring said successively generated signals in at least three signal sensor arrays, said signal sensors arrays being spatially associated with a portable image detector assembly, and said measured signals being indicative of an absolute distance between said generator arrays and detector arrays respectively, obtain coordinate data for each sensor array position by performing trilateration on said derived absolute distances between said generator arrays and sensor arrays respectively, calculating the coordinate data for said center of the surface of said portable image detector assembly based on the known positions of said center and said sensor array positions, and calculating said SID by subtracting the coordinate data of said center of the surface of said portable image detector assembly from the coordinate data of said emission point of said X-ray source. It further provides for a method for determining the thickness of a patient through measurement and subsequent subtraction of the SID and SSD (source-skin-distance).

[0014] It is an object of this invention to resolve the above-mentioned problems, and provide a system that allows to accurately and reliably determine the thickness of a patient who is positioned against the surface of an imaging detector in a radiography system.

[0015] In order to resolve part of this problem, it is necessary to identify in a

reliable way the location and orientation of an imaging detector with respect to the position and orientation of an X-ray beam. At the same time, the distance between the patient's skin surface and the X-ray source has to be known. It is necessary to be able to find a technique that allows to determine all 6 degrees of freedom defining said position and orientation. While other techniques, referred to above, are limited to 4 or 5 degrees of freedom (because they rely on partially relative measurement techniques such as angle measurements or alike), our invention resolves this limitation by determining the actual spatial coordinates (x, y, z) of three known points in the imaging detector. These three known points refer to the physical locations of three sensors of which the exact location is determined by the method of this invention. These three points mathematically define a plane in which the image detector is oriented. But also they define the rotation angle of the image detector about all axes, and the distances from the reference point located on the X-ray source assembly. Knowing the three coordinates for the three sensors determines the 6 degrees of freedom of the position of the imaging detector in space relative to a reference coordinate system determined by the location of three signal generators attached to the X-ray source assembly.

[0016] The accurate determination of the image detector with respect to the

orientation of the X-ray beam, allows then different practical applications in radiography, such as an automated and accurate determination of the SID (source image distance) which is an important parameter to select the exposure settings for any type of radiography study. Orientation and position data allow automated or guided adjustment of the alignment of the detector (and patient) with the X-ray beam. Another application can be found in the automatic determination of the acquisition geometry of a radiography system in a tomosynthesis context, wherein an accurate knowledge of the acquisition geometry is necessary for accurate image reconstruction.

[0017] The actual determination of the spatial coordinates (x, y, z) of the three signal sensors in the imaging detector is based on a known calculation technique used in the field of GPS; namely trilateration. The method relies on reliable measurements of the distances between at least three signal generators (the satellites in GPS context) and a signal sensor. In the assumption that the exact location of the three signal generators is known, then also the exact location of the signal sensor registering the at least three signals from said three signal generators can be derived through this trilateration calculation. The key to the calculation are the reliable measurements of said distances.

[0018] In the context of this invention, the measured signals are indicative of the absolute distance between a signal generator and a signal sensor, which means that every measured signal is directly related to an absolute distance (in a linear or non-linear fashion). Different measurement techniques or physical principles may be envisaged allowing direct derivation of an absolute distance from a measured signal. Examples may be for instance, the measurement of an acoustic echo-signal, a radar signal, a laser distance measurement, magnetic field strength, or alike...

[0019] In the context of this invention, a signal generator consisting of a triplet of orthogonally arranged coils is called a generator array. Similarly, a signal sensor consisting of a triplet of orthogonally arranged coils is called a sensor array.

Brief description of drawings

[0020] Fig.1 shows a schematic representation of a patient in supine position positioned under an X-ray tube with an emission point [1 12]. The distance between the emission point [1 12] and the center of an image detector [120] is called the source-image-distance or SID [130]. The distance between the emission point [1 12] and the point of incidence of the X-ray beam on the patient is called the source-skin-distance or SSD [131]. The thickness of the patient is therefore the difference between the SID and the SSD.

[0021] Fig.2 shows a schematic representation of a mobile X-ray device [100] to which a set of 3 generator arrays [1 11] is attached by means of a structure that is fixed to the X-ray source assembly or collimator [1 10]. The 3 generator arrays are fixed in a frame-like structure which keeps the generator arrays at a fixed distance from each other, and of which the dimensions are known. The image detector assembly [120] is shown in conjunction with the X-ray device. A signal [122] is sent from one of the generator arrays, and is detected by all 3 sensor arrays which are integrated into or attached to the image detector assembly [120].

[0022] Fig. 3 shows details of the X-ray source assembly to which the set of 3 generator arrays [1 1 1] is attached. The detector arrays are spaced from each other. The arrow [130] represents the SID (source to image distance) and can be obtained through measurements performed by the method of the invention. The SID is the distance between the center of the image detector assembly [120] and the emission point [1 12] of the X-ray source.

[0023] Fig. 4 shows the concept of trilateration which is also applied in GPS

technology, which allows to accurately calculate the coordinates of a GPS- receiver (represented by one of the cars in the picture) under the minimum condition 3 satellites send a signal that allows the GPS-receiver to determine its distance to all of the 3 satellites. The minimum requirement to apply this technique is to have at least 3 satellites providing the distance information. In our invention, the GPS-satellites are replaced with generator arrays that each provide a consecutive signal to each of the sensor arrays (in analogy with the cars in the figure) that can be

interpreted by the sensors as their respective distance to the generator array. In the invention, all sensor arrays (cars) calculate their distances to the 3 generator arrays (satellites), after which 3 trilateration calculations are executed to obtain the coordinates from the sensor arrays (cars): (x1 ,y1 ,z1 ), (x2,y2,z2) and (x3,y3,z3).

[0024] Fig. 5 shows an embodiment of an X-ray detector assembly [120] which on 3 of its corners is foreseen with a sensor array [121]. In this configuration, it is obvious that the internal distances between the sensor arrays [120a], [120b] and [120c] are constant and known, and therefore can be used to verify the accuracy of the obtained sensor array coordinates via the method of this invention.

[0025] Fig. 6 shows an embodiment of a collimator [1 10] aligned with a set of generator arrays [1 1 1] which are associating their configuration in space with the alignment of the X-ray beam (represented by the Z-axis indicated in the figure).

Description of embodiments

[0026] A first embodiment of this invention is based on the use of a set of at least three signal generators, which are physically connected to an X-ray source assembly of a radiography system. The at least three signal generators are mechanically fixed to the X-ray source assembly in order to fix a reference coordinate system in space which is selected as the point from where the X-ray radiation originates and that determines the orientation of the X-ray beam. This virtual point is called the emission point of the X-ray source. At least three signal generators are required to determine the location and orientation of the reference coordinate system. Adding more signal generator elements may contribute to a better accuracy of the system, but is strictly speaking not required. The at least three signal generators are so arranged that they are spatially separated from each other (for instance located at positions around the X-ray source assembly) so to ensure that their generated signals appear to be different for detectors at different positions.

[0027] In a preferred embodiment, three signal generators sequentially or

successively generate their signal, meaning that the generators are never operated simultaneously, but one after the other. The sequential or successive signal sequence is then repeated until the detector position measurements are completed.

[0028] The particular aspect or characteristic of the generated signal that

determines the absolute distance between the signal sensor location and the generator location is then picked up by signal sensors within detection range of the signal generators. A prerequisite for a signal generation and sensing technology to be used in this application is that at least one measureable parameter of the signal (such as the intensity, the amplitude or signal strength) should be dependent on the physical distance between the signal generator and the signal sensor itself. It is conceivable and acceptable that additional mathematical operations and calibrations are required to convert the signal data into a distance value. Preferably, the measurable parameter of the signal should not be easily influenced or degraded by environmental factors, and should be transparent for human tissue.

[0029] Further to our preferred embodiment above, there are three signal sensors built into the image detector assembly, of which the position is being detected. At least three of the above mentioned signal sensors are required for the invention to work, and they are required to be as far as possible spatially separated from each other to obtain the most accurate results. An obvious design choice for the location of the three signal sensors in an image detector assembly would be in three of four corners of the assembly frame. Since the determination of the exact positions of the three signal sensors is the purpose of the method of this invention, it is clear that these three positions can uniquely identify the location and orientation of the image detector assembly when the dimensions of the image sensor assembly and the locations of the signal sensors within the assembly are known. In geometry, two points identify a line in space, three points define a plane. The invention therefore requires at least three signal sensors, which completely define the position and orientation of the image sensor assembly, while every additional sensor may only contribute to an increased accuracy of the calculation. Moreover in the case that the internal distances between the signal sensors (or the dimensions of the image detector assembly) are known, the accuracy of the measured signal sensor locations of the three signal sensors can be verified by comparing the known internal distances of the sensors against the measured internal distances resulting from the measured locations of the signal sensors.

[0030] The three signal generators produce a continuously cycling time-based generator signal sequence so that all possible distance measurements (consisting of three simultaneous measurements between the active signal generator and the three signal sensors) can be performed one after another. As explained above, the location of the signal sensor determines the measured signal value which is picked up by each signal sensor for every generator signal. This results in our particular embodiment in a total of nine signal measurements. These measured values are then converted into distance measurements, and fed into a trilateration algorithm which calculates the 3D-coordinates of the positions of the signal sensors relative to the position of the (virtual) emission point.

[0031] The trilateration algorithm is a known mathematical method that

essentially identifies the point in space that intersects with three

(imaginary) spheres (in the case of our embodiment spheres around the three signal generators) defined by the positions of the signal generators as the center of said spheres and the above mentioned calculated distances as their respective sphere radii. In theory, the results of this equation are 2 points in space, of which one can be easily ruled out since the only valid one should be found in the direction of the X-ray beam, while the other is mirrored through the emission point towards the rear of the X- ray source assembly (and thus away from the X-ray beam direction). [0032] As explained above, this means that in case that any additional signal generators are added to the system (so that there are more than three generators in total), the resulting distance measurements to the different signal sensors will contribute to the accuracy of the calculation of the coordinates by adding a supplementary sphere whose surface should also mathematically intersect with the coordinate point of the signal sensor. In case that the additional sphere does not exactly intersect with the location of the signal sensor calculated from the first 3 spheres, then this data can be used to interpolate between the different possible solutions (or other optimization algorithms know from GPS technology can be used).

[0033] A further important aspect of the invention is the appropriate choice of distance detection principle that can fulfill the above-mentioned

requirements of accuracy and transparency to human tissue. A promising starting point appears to be magnetism, as it is known that magnetic fields pass through non-metallic objects. Nevertheless, other pervasive waves, beams or radiating energies may be considered as alternatives. It is known in the art that the human body is transparent for alternating magnetic fields, on condition that the frequency is below 100 kHz. When the frequency raises above 5 MHz, human tissue starts to absorb a considerable fraction of the emitted energy and the signal becomes more sensitive for electromagnetic disturbances caused by electric circuits in the vicinity. These elements make frequencies above 5 MHz less usable for the purpose of distance measurement. The reason for using alternating magnetic fields as opposed to static magnetic fields is that this approach compensates for the presence of any static magnetic field, such as the earth's magnetic field in case that the amplitude of the received signal is taken as the parameter for distance measurement.

[0034] Therefore it is contemplated to use resonant magnetic signal generator- and sensor-pairs to optimize the magnet signal transfer to perform the distance measurements. In a preferred embodiment, the signal generator is a coil through which an alternating electric current is sent by an electronic generator circuit. The alternating electric current induces an alternating magnetic field of the same frequency which can be detected at a distance from the generator coil. A similar or smaller coil is used as the signal sensor and picks up the alternating magnetic field passing though it; inversely the alternating magnetic field induces an alternating current in the sensor coil that can be picked up and measured. The detector coil may be smaller in size and have a different winding number in comparison with the generator coil. The circuit that reads out the alternating electric current in the sensor coil is "tuned" to the frequency of the generator (hence the name "resonant") by means of the selection of the correct capacitor and resistor network, in order to optimize the sensitivity to the generator signal frequency.

[0035] While in the above preferred embodiment, the magnetic generator and resonant magnetic sensors are coils, different types of magnetic sensors may be used as alternatives, such as MEMS, Hall effect based sensors, magneto-resistive sensors, or alike...

[0036] The amplitude of the measured alternating current in the resonant

magnetic sensor has a relation with the distance between the signal generator and signal sensor. The amplitude is approximately inversely proportional to the cubic distance from the signal generated, when measured in radial direction from the generator. When the sensor coil is kept at the same distance to the generator but its orientation and/or the relative orientation is changed the received signal ranges from 100% to 0%.

[0038] The mutual inductance L _m depends on three vector products; a is a vector describing the orientation of the generator; b describes the orientation of the sensor and r the relative orientation between sensor and generator.

[0039] The maximum amplitude can be measured when the vector determining the magnetic field in a certain position aligns with the sensor coil axis. In all other cases only the component of the magnetic field having the same direction as the sensor coil axis will contribute to the measured value in that particular sensor coil. Relying on the amplitude measurement for distance determination would only be a viable solution in the case that no angular variation could be expected (and this is not the case in our application).

[0040] Another aspect that influences distance measurement values based on the amplitude read-out from a single coil sensor are disturbances of the magnetic field, such as the presence of metallic objects in the vicinity of the sensor coil. Such disturbances have not so much an impact on the amplitude component of the magnetic field vector, but rather on the direction of the magnetic field vector. This means that such disturbances nevertheless can have a significant impact on the measured amplitude by the sensor coil in a certain position. This is another reason why relying on the amplitude measurement of a single magnetic sensor alone is not sufficient to determine the distance in a reliable way between sensor and generator.

[0041 ] Replacing the above described single coil (or other type of magnetic

generator and sensor) structures in both the signal generator and signal sensor with a triplet of orthogonally arranged coils increases the distance measurement accuracy drastically, but only on condition that they are operated in a particular way which is disclosed in this invention. In the context of this invention, a signal generator consisting of a triplet of orthogonally arranged coils is called a generator array, and a signal sensor consisting of a triplet of orthogonally arranged coils is called a sensor array. While a single coil generator read out by a single coil sensor results in a single amplitude measurement, reading out all combinations of a generator triplet and a sensor coil triplet leads to 9 amplitude signal readouts, being 9 induced currents and resulting voltages Vij.

[0042] To simplify the calculation of distance and coordinates of the sensor array it is of advantage to square the voltage matrix elements and to define a signal matrix and the total signal Stot as well as the signal components Sx, Sy &. Sz ^' -

2 2

•Ί ΐ ^V U ^V 13

2 2

su = v ₂₂ ^V 23 =∑∑ S,j ; S _X =∑ S _u ; S _y =∑ S _;,2 ; S _z =∑ 5,3 ;

.2 2 2 1=1 y=l i=l =l

^V 32 ^V 33

[0043] The distance between the generator array and the sensor array ? can be calculated from Sfo/ and the axis intercepts or coordinates x; y & z from S _x ,

Sy & Sz-

R ⁴

x =

const. V 18

R ⁴ s _x -s,

y =

const. 18

R ⁴

const. V 18

[0044] It follows from the above formulas that it is possible to directly calculate the exact position (expressed in the coordinates x, y& z) of a sensor array. However, this method is not desirable for direct determination of the locations of the sensor array. The formulas above immediately illustrate that the determination of each coordinate is very sensitive to changes in the respective angle between the generator and detector coils.

[0045] In contrary to the sensitivity of the determination of the sensor coil

coordinates through measurements of the magnetic signals as described above, the determination of the absolute distance R \s not sensitive to the orientation of the sensor arrays with respect to the generator array; the distance /? only depends on the summation of the signals generated in the coil triplet and picked up in the 3 sensor coils. It is upon this aspect that a reliable calculation of the absolute distance is based for further application in the trilateration method described above. [0046] The distance measurement using the above mentioned set of coil triplets relies upon the sum of the squares of all voltage matrix elements. This measurement is surprisingly stable against external fields or disturbances induced by metal nearby as it uses all measured values (generated by all signal generators, and picked up by all signal detectors). The robustness can be explained by the fact that the distance is calculated from the complete magnetic field of all three emitter coils and not only from field components. Introducing a distance dependent gain matrix G,- can compensate deviations from the simplified model used here which assumes that the distance R is many times larger than the size of the coils themselves, so that the coils may be assumed to behave like dipoles. The gain matrix can also correct for different coupling efficiencies between different receiver and generator coils. The measurement of the coordinate components x; y & z is less stable as only one receiver coil is used for each coordinate axis. Additionally there are different positions which produce the same voltage matrix - all locations that are point-symmetric to the emitter (which results in an uncertainty between 2 different possible solutions giving the same voltage matrix). But this problem is not important as for a radiography setup only half of the sphere is used - we always work in the same hemisphere, namely in the direction of the X-ray beam.

[0047] Electrical conductors such as metal plates or objects close to emitter or receiver produce interference and thus distort the measurement. This is caused by so-called "eddy currents" induced in the conductor by the alternating field of the emitter coils. The eddy currents in a conductor produce a magnetic field opposed to the inducing field. The magnetic field thus gets weaker (the eddy currents drain out energy from the magnetic field). On the other hand the eddy current also induce a current in the sensor coils. In case of resonate circuit design there is no phase difference between the directly induced currents and the eddy current induced currents. As a result, the measured (and thus observed) distance can get larger or smaller dependent on the location and orientation of the metal object. The impact on the measured distance depends on the distance of the conductor to the signal generator array or sensor array respectively. The influence on the x, y & z coordinates and the orientation of the sensor array is much larger than the distortion of the distance measurement when using the coil triplet configuration. The magnetic field induced in the conductor depends on the orientation of the conductor to the generator. Thus, the magnetic field is not reduced symmetrically but mainly in the direction perpendicular to the conductor so that the

measured field strength is mainly changed for the sensor coil which is oriented almost perpendicular to the plate. In conclusion, eddy currents influence the distance measurement moderately (a slightly larger or smaller distance is measured) while coordinate and orientation

measurements are influenced strongly.

[0048] Another potential source of distortion of the measurements are external alternating magnetic fields. Almost any electrical device produces alternating electro-magnetic fields in a wide frequency range. The influence of the fields strongly depends on their frequency and on the design of the read-out for the sensor coil. If the frequency of an external field fits the design frequency of any sensor coil the influence on the measurement accuracy can be very large. External magnetic fields may thus influence the measurement accuracy but the influence is less pronounced compared to eddy currents as the external field typically has no fixed phase relation or do not fit to the design frequency.

[0049] So, in conclusion, measuring absolute distances using magnetic field

strength measurements can be made robust when using coil triplet configurations as alternating magnetic field generators and sensors for the above-mentioned reasons. In a preferred embodiment at least three of such alternating magnetic field signal generators (each consisting of an orthogonally arranged coil triplet) are used in combination with at least three such alternating magnetic field signal sensors (each also consisting of an orthogonally arranged coil triplet). As explained above, a reliable distance calculation can be made when for each generator array;

alternating magnetic fields are generated successively in each of the 3 coils making up the triplet. These successive signals may be then read out by all sensor coils simultaneously of each signal sensor triplet or sensor array. Reading out the data from all respective sensor coil triplets results in the distance calculations between the generator coil triplet or generator array in question and the respective sensor coil triplets or sensor array. When the same process is repeated for all generator arrays (at least 3), an accurate calculation can be made based on the trilateration technique of the position of the image detector assembly in which the different sensor arrays are integrated.

In order to determine the thickness of a patient, which is the purpose of this invention, two positions are measured, out of which a distance is calculated. It is assumed that the detector surface is positioned against the patient at the side that is opposite to the position of the X-ray source. Therefore, the thickness of the patient ("c ") is determined as the mathematical difference (or the distance) between the coordinates of the center of the image detector assembly on which the central axis of the X- ray beam coincides, and the coordinates of the incidence point of the X- ray beam on the surface (or skin) of the patient. The distance between these 2 points with coordinates (ai ,a2,a3) and (bi ,b2,b3) is calculated by applying Pythagoras' theorem twice:

[0051] Based on the same information as obtained above -namely the accurate coordinates or position information of the detector arrays- the center of the image detector surface can be easily calculated when the geometric relationship between the (at least) 3 locations of the detector arrays are known in relation to this center of the image detector surface. In case that the detector arrays are physically located on 3 of the corners of the image detector assembly, the center of the image detector surface will be halfway between the detector arrays which are positioned diagonally on opposing corners of the detector.

[0052] The measurement of the coordinates of the incidence point of the X-ray beam on the surface of the patient can be achieved using any known and suitable distance measurement techniques for measuring the distance between the emission point of the X-ray source and said incidence point. These measurement techniques can be, but are not limited to, optical range finding techniques, laser distance measuring, tape measurements or alike. Measuring this latter distance is not restricted the same limitations of not having a clear line of sight, and thus is much easier to achieve. The calculations involved in the execution of the method as explained in this invention may be carried out by means of standard computer equipment or a standard computer configuration, and may be embodied as a computer program, or alternatively may be embodied in a dedicated programmed circuit allowing to perform these calculations.