One dimensional transmission detector for radiotherapy

Abstract

A sensing device for a radiation therapy apparatus, the apparatus comprising an accelerator and a beam-shaping device the beam shaping device being a multi-leaf collimator (MLC) (2) having a plurality of pairs of leaves, and a rotatable gantry, the sensing device comprising: a transmission electronic detector (1) comprising an array of ionization chambers. The ionization chambers are defined by a bias electrode (11a,11b, 34,42) on the one hand and by a planar array of conductive strips (40) or strip assemblies (30) on the other hand. The strips or strip assemblies are associated to the leaf pairs of the MLC. The strips are the collecting electrodes of the ionization chambers. Each strip assembly or in the case of one particular embodiment, each strip, yields two currents which allow to determine the position of the leaves of a leaf pair associated with the strip or strip assembly, a gantry sensor configured to determine at least one position associated with a gantry angle; and a processor, adapted to determine the position of at least one pair of leaves of the MLC using the currents i.sub.1 and i.sub.2 obtained from the collecting electrode strips.

Claims

1. A sensing device for a radiation therapy apparatus, the apparatus comprising an accelerator, a beam-shaping device, the beam shaping device being a multi-leaf collimator (MLC) having a plurality of pairs of leaves, and a rotatable gantry, the sensing device comprising: a transmission electronic detector comprising one or two bias electrodes and a planar 1D array of electrically conductive strip assemblies, and wherein: each strip assembly comprises two strips which are oriented in the direction of the leaf pairs, and configured so that each leaf pair of the MLC is associated with one or more strip assemblies, when the detector is positioned between the MLC and a target of interest, each strip together with one of the bias electrodes defines an ion chamber, with the strip forming the collecting electrode thereof, the two strips of each assembly are configured to respectively produce two currents i.sub.1 and i.sub.2 when the detector is irradiated by a beam that is shaped by the MLC, and when an electric field is created between the bias electrode(s) and the strips, said currents are used to determine the position of the pair of MLC leaves associated with the strip assembly, for each strip assembly, a pair of electrometers comprising respective read-out amplifiers coupled respectively to the strips of the assembly, for measuring the currents i.sub.1 and i.sub.2, a gantry sensor configured to determine at least one position associated with a gantry angle; and a processor, adapted to determine the position of at least one pair of leaves of the MLC using the currents i.sub.1 and i.sub.2 obtained from the strip assemblies.

2. The sensing device according to claim 1, wherein each strip assembly comprises two collecting electrodes mounted back to back, and wherein the detector comprises two planar bias electrodes which are conductive plates placed on either side of the planar 1D array of strip assemblies, and oriented in such a way that the distance between every strip on one side of the planar 1D array and the corresponding bias electrode increases monotonically along the length of the strip in a given direction along said length, while the distance between every strip on the other side and the corresponding bias electrode decreases monotonically along the length of the strip, in the same given direction.

3. The sensing device according to claim 2, wherein the bias electrodes are parallel to each other.

4. The sensing device according to claim 1, wherein each strip assembly comprises two adjacent collecting electrode strips, and wherein the width of the first electrode strip decreases monotonically along the length of the strip assembly in a given direction along said length, while the width of the second electrode strip increases monotonically in the same given direction, and wherein the detector further comprises one bias electrode in the form of a conductive plate that is parallel to the planar array of strip assemblies.

5. The sensing device according to claim 4, further comprising a large area collecting electrode on the back side of the planar array of strip assemblies, and a second bias electrode parallel to the planar array of strip assemblies and facing said large area collecting electrode.

6. A sensing device for a radiation therapy apparatus, the apparatus comprising an accelerator, a beam-shaping device, the beam shaping device being a multi-leaf collimator (MLC) having a plurality of pairs of leaves, and a rotatable gantry, the sensing device comprising: a transmission electronic detector comprising a bias electrode and a planar 1D array of electrically conductive strips oriented parallel to the bias electrode, and wherein: each strip is oriented in the direction of the leaf pairs, and configured so that each leaf pair of the MLC is associated with one or more strips, when the detector is positioned between the MLC and a target of interest, each strip together with one of the bias electrodes defines an ion chamber, with the strip forming the collecting electrode thereof, each strip is configured to respectively produce two currents i.sub.1 and i.sub.2 at the opposite outer ends of the strip, when the detector is irradiated by a beam that is shaped by the MLC, and when an electric field is created between the bias electrode(s) and the strips, said currents are used to determine the position of the pair of MLC leaves associated with the strip assembly, for each strip, a pair of electrometers comprising respective read-out amplifiers coupled respectively to the opposite outer ends of the strip, for measuring the currents i.sub.1 and i.sub.2, a gantry sensor configured to determine at least one position associated with a gantry angle; and a processor, adapted to determine the position of at least one pair of leaves of the MLC using the currents i.sub.1 and i.sub.2 obtained from the strips.

7. The sensing device according to claim 6, further comprising a large area collecting electrode on the back side of the planar array of strips, and a second bias electrode parallel to the planar array of strips and facing said large area collecting electrode.

8. The sensing device according to claim 1 wherein the sensing device further comprises a receiver, configured to receive a patient plan (including patient dose, patient images taken before planning and radiation beam configuration); and the processor is adapted to compare expected (from plan) and measured LINAC behavior (in term of leaves position, flux and gantry angle).

9. The sensing device of claim 8 wherein the processor is further configured to enable real-time signalling of a deviation from an expected behavior and a measured behavior exceeding a user-defined tolerance.

10. The sensing device of claim 8, further comprising: a storage means in communication with the processor, the storage means configured to store a beam model of the accelerator; a fluence-reconstructing engine configured to compute a photon fluence based on the position of pairs of the leaves determined by the processor and on the beam model wherein the photon fluence is determined independently from a presence of a patient or phantom in the beam; and a dose calculation engine configured to compute a delivered 3D dose distribution using fluence and patient images.

11. The sensing device of claim 9 wherein the processor is further adapted to compare a calculated delivered dose to a planned dose.

12. The sensing device of claim 9 wherein: the processor is further configured to extract at least one subset of a 3D dose distribution and a set of quality parameters from the calculated 3D dose distribution.

13. The sensing device of claim 9 further comprising: an image acquisition station configured to acquire updated patient images; an image processor configured to communicate the images to the dose calculation engine whereby the dose calculation engine calculates a dose distribution based on measured fluence and the updated patient images.

14. The sensing device according to claim 9 wherein the processor is further adapted to accumulate a plurality of data measurements relating to a corresponding plurality of dose delivered to the patient during consecutive fractions.

15. The sensing device according to claim 12 wherein the processor is further adapted to accumulate a plurality of data measurements relating to a corresponding plurality of dose delivered to the patient during consecutive fractions.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a 3D view of the main components of a detector in accordance with a first embodiment of the invention, and its position relative to a multi-leaf collimator (MLC).

(2) FIG. 2 is a side view of the detector in accordance with the first embodiment.

(3) FIG. 3 is another side view of the collimator in accordance with the first embodiment, including indications of the geometrical parameters needed to determine the position of two leaves of the MLC.

(4) FIG. 4 is a 3D view of the main components of a detector in accordance with a second embodiment of the invention.

(5) FIG. 5 is a plane view of one collecting electrode strip in a detector according to the second embodiment.

(6) FIG. 6 illustrates a number of geometrical considerations made in relation to a detector according to the second embodiment.

(7) FIG. 7 is a 3D view of the main components of a detector in accordance with a third embodiment of the invention.

(8) FIG. 8 is a side view of one collecting electrode strip of the detector according to the third embodiment, including a number of geometrical parameters needed to calculate the position of two leaves of the MLC.

(9) FIG. 9 illustrates the electrical equivalent of one collecting electrode strip in a detector according to the third embodiment.

DETAILED DESCRIPTION OF THE INVENTION

(10) This detailed description sets forth specific details of exemplary embodiments of the present invention to enable a thorough understanding: However, those of ordinary skill in the art understand that the scope of the present invention is not restricted to those embodiments. Other embodiments, whether described herein or omitted, may be practiced without these specific details. In other instances, well-known methods, procedures and components are not described in detail, thus avoiding obscuring the scope of the invention. Accordingly, this description does not limit the invention to the scope of the embodiments described. Although certain features are illustrated and described, many modifications, substitutions, changes, and equivalents will now occur to those of ordinary skill in the art. It is understood that the appended claims are intended to cover such modifications and changes as fall within the true spirit of the invention.

(11) A beam model is a mathematical description of a radiation therapy apparatus in general and contains a number of parameters. These parameters take into account the characteristics of the accelerator (e.g., energy spectrum and lateral beam quality variations), the shapes and positions of the effective radiation sources, and the geometry and material of the beam shaping devices. A fluence is the number of photons/particles passing through a plane perpendicular to the beam per unit time and unit area. A fluence computation algorithm is a set of mathematical rules that computes the fluence according to the beam model and a given parameter set. The representation of the computed fluence (units, coordinate systems) is such that it is compatible with additional computational procedures for computing deposited dose in tissue and/or detector response. Useful descriptions of basic beam modelling techniques are provided, for example, by Wolfgang A. Tome, Beam Modelling for a Convolution/Superposition-Based Treatment Planning System, Medical Dosimetry, Vol. 27, No. 1, pp. 11-19, 2002; or by Nikos Papanikolaou, Investigation of the convolution method for polyenergetic spectra, Med. Phys. 20(5), 1993.

(12) Three technical solutions for the sensor have been identified, which allow the measurement of the aperture of each leaf pair AND the center of gravity of the aperture (it is not simply the measurement of a linear dose integral). These solutions are:

(13) a) A double array of regular strips with inclined electrodes (FIG. 1-3)

(14) b) Strips with variable width (FIG. 4-6)

(15) c) Resistive strips (FIG. 7-9).

(16) Barthelmes et al. (U.S. Pat. No. 4,803,368 issued 1989 Feb. 7) generally describes an ion chamber having inclined electrodes. Further, Islam et al. (U.S. Pat. No. 8,119,978 issued on 2012 Feb. 21) describes the use of strips with variable width in a single ion chamber. However, the present invention extends the teaching of the prior art. For example, the present invention includes for every leaf pair of an MLC, at least one ion chamber and two readout channels.

(17) The sensing device according to the present invention comprises a detector that is built on the basis of a one-dimensional (1D) array of elongate ion chambers. The detector is configured to be attached to the head of a particle accelerator, usually a linear accelerator (LINAC). The accelerator head is equipped with a multi-leaf collimator and coupled to a rotatable gantry. Therefore, the detector rotates integrally to (meaning together with) the MLC, i.e. the detector is stationary relative to the rotating MLC, which itself can rotate together with the gantry or about the beam axis. There is a direct mapping between the ion chambers and the leaf-pairs of the MLC. This means that the number of ion chambers is chosen in accordance with the number of leaf pairs of the MLC. The ion chambers are defined by elongate conductive strips on a planar surface, preferably on a PCB, said strips acting as the collector electrodes of the ion chambers, and by a single bias electrode facing the strips, if the strips are present on one side of the planar surface, and by two bias electrodes, if the strips are present on both sides. According to a first and second embodiment, the strips are arranged in strip assemblies, each strip assembly comprising two strips, yielding two respective output currents. An integer number of strip assemblies is associated with each leaf pair when the detector is attached to the accelerator head. Associated with means that a strip assembly is positioned to receive radiation passing through a particular leaf pair. For instance, the motion of a collimator's two banks of leaves, with 80 leaf-pairs may be tracked with a detector having 80 collector electrode strip assemblies, each strip assembly yielding two output currents. The strips connect to 160 corresponding channels of readout electronics. In contrast to conventional 2D arrays presently commercialized as online dosimeters, the present invention's number of 160 channels is remarkably lower than the number of pixels (from 1600 to 4000) of these 2D detectors.

(18) According to a third embodiment, the strips are arranged as an array of single elongate strips, each strip yielding two output currents at its opposite outer ends. In the third embodiment therefore, an integer number of strips is associated with each leaf pair when the detector is attached to the accelerator head.

(19) Ion chamber technology offers high reliability and performance with moderate manufacturing costs. The embodiments of the present invention may use, or be based upon, collecting electrodes manufactured using standard printed circuit board technology (PCB), which is widely available and economical.

(20) FIGS. 1 to 3 illustrate a first embodiment of the present invention. FIG. 1 shows a 3D view of the main components of a detector 1 according to the first embodiment, as it is positioned relative to a MLC 2. The first embodiment configures, or is adapted, to measure the position of MLC leaves with a 1D array 10 of collector electrode strip assemblies. The detector comprises, or consists of, two sub-arrays 10a and 10b, which mount back to back. In each sub array, segmented collecting electrodes, in the form of printed PCB strips are combined with bias electrodes 11a and 11b in the form of simple, conductive planes placed at non-zero inclination angle relative to the strips. A strip assembly in this embodiment is thus an assembly of two strips mounted back to back. All the strip assemblies are oriented in the direction of the collimator's leaf-movement. The drawing in FIG. 1 is a mere sketch, and does not clearly show the number of collector electrode strip assemblies in the 1D array. This number is preferably equal to the number of leaf pairs of the MLC 2, or it may be an integer multiple thereof. The embodiments described hereafter in more detail are based on the case wherein the number of strip assemblies is equal to the number of leaf pairs, and wherein the width of the strips is configured so that when the detector is attached to the accelerator head, each strip assembly in the array is associated with one corresponding leaf pair in the array of leaf pairs of the MLC (i.e. each strip assembly receives the radiation from one leaf pair). FIG. 2 illustrates in a side view the position of two leaves 20 of a leaf pair, relative to the detector 1. A collimated beam 21 passes through the opening defined by the leaves 20 and generates two currents i.sub.1 and i.sub.2 respectively, in the strips of the strip assembly that faces the leaf pair (i.e. that is positioned to receive the irradiation passing through said leaf pair). The first current i.sub.1 is generated in the upper IC, between the bias electrode 11a and the upper collector electrode 10a. The second current i.sub.2 is generated in the lower IC, between the bias electrode 11b and the lower electrode 10b. The detector comprises electrometers known as such in the art, which comprise at least a pair of amplifiers 22, for reading out the currents i.sub.1 and i.sub.2. According to a preferred embodiment, the bias electrodes 11a, 11 b are mutually parallel.

(21) Assuming that scattering and lateral spread of secondary electrons in the chamber volume can be neglected, the current collected by one side of the n.sup.th strip is given by the following proportionality relation (the symbol a meaning proportional to):
i.sub.n(x,y.sub.n)k(x)ydx(1)
Where is the beam photon fluence, k(x) is the distance between the strip and the corresponding bias electrode, y is the strip width and y.sub.n is the position of the strip center along the array. For the detectors described hereafter, the above equation can be written in a simplified way.

(22) The aim of each IC measurement is the determination of a dose D, applied in the IC volume by the incident radiation. Dose is the energy deposited by radiation in an amount of mass of material:

(23) $\begin{array}{cc}D=\frac{E}{m}=\frac{E}{V.Math.}& \left(2\right)\end{array}$
where D=dose in Gy (1Gy=1 J/kg)

(24) E=energy

(25) m=mass

(26) V=volume in m.sup.3 or mm.sup.3

(27) =density in kg/m.sup.3

(28) What is measured however, is the charge Q produced in the sensitive volume.

(29) $\begin{array}{cc}Q=\frac{E}{W_{\mathrm{air}}}.Math.e=\frac{D.Math.V.Math..Math.e}{W_{\mathrm{air}}}& \left(3\right)\end{array}$
where W.sub.air is the ionization constant for air (the energy which is required to produce one pair of charges in air).
The measured current i is the time derivative of the charge Q collected:

(30) $\begin{array}{cc}i=\frac{\stackrel{.}{D}.Math.V.Math..Math.e}{W_{\mathrm{air}}}=K.Math.V\mathrm{with}K=\frac{\stackrel{.}{D}.Math..Math.e}{W_{\mathrm{air}}}& \left(4\right)\end{array}$
where {dot over (D)} is the dose rate (in Gy/min or Gy/s). {dot over (D)} depends on the treatment machine emitting the ionizing radiation and its calibration. Air density is approximated with 1.2 kg/m.sup.3. W.sub.air and e are physical constants with the values W.sub.air=33.97 eV respectively e=1.602.Math.10.sup.19 C. Here and in the following formulae, the proportionality constant K is calculated based on the system architecture or, alternatively, experimentally determined with a calibration procedure.

(31) With the assumption that the fluence is uniform within the collimator aperture, the aperture x of a leaf pair and its center of gravity x.sub.0, are determined from the currents i.sub.1 and i.sub.2 by the following proportionality relationships
X(i.sub.1+i.sub.2)
x.sub.0(i.sub.1i.sub.2)/(i.sub.1+i.sub.2)

(32) With reference to FIG. 3, the above relationships can be derived for the detector according to the first embodiment:
i.sub.1=K.Math.V.sub.1=K.Math.p.sub.s.Math.x.Math.(y.sub.min+sin().Math.x.sub.0)(5)
i.sub.2=K.Math.V.sub.2=K.Math.p.sub.s.Math.x.Math.(y.sub.min+sin().Math.(Lx.sub.0))(6)
Herein, p.sub.s is the pitch of the array of collector electrodes (i.e. the pitch of each sub-array). This is a constant value for a given detector geometry. p.sub.s is equal to the strip width y plus the distance between two adjacent strips.

(33) x and x.sub.0 can be obtained from building the sum i.sub.1+i.sub.2 and the difference i.sub.1i.sub.2 and result in:

(34) $\begin{array}{cc}x=\frac{i_1+i_2}{K.Math.p_s.Math.\left(2y_{\min }+\sin \left(\right).Math.L\right)}=\frac{i_1+i_2}{K^{}}& \left(7\right)\\ x_0=\frac{L}{2}+\frac{i_1-i_2}{2K.Math.p_s.Math.x.Math.\sin \left(\right)}& \left(8\right)\end{array}$

(35) Additionally, the radiation flux is proportional to the sum of the current in all the strips. Knowing the position of the leaves and flux enables one to prove delivery of the prescribed plan (expressed, e.g., in terms of MU/sampling interval, MU/segment or MU/control point), wherein MU stands for Monitor Units.

(36) Those having ordinary skill in the art will appreciate that other solutions, alternative to the above-described inclined electrode ion chambers, can easily be figured out, leading to similar relationship between leave positions and measured current. The key point is that the sensitivity of each strip has to be modulated along its length to give sensitivity to beam position x.sub.0.

(37) A second embodiment of the present invention includes the 1D array 10 of collector electrode strips shown in FIGS. 4-6. Here each strip assembly 30 consists of a set of adjacent electrodes 31 and 32 printed on the same side of a PCB 33. single bias electrode 34 is provided. The bias electrode 34 is a conductive surface parallel to the collecting electrodes, i.e. at a constant distance h from the array of collecting electrode strips 30. The width of the collecting electrodes 31 and 32 changes monotonically in the x-direction but at an opposite rate. In other words, the first electrode 31 decreases monotonically from high width to low width in the direction +x, while the second electrode 32 increases monotonically from low width to high width in the same direction +x. According to the preferred embodiment, for a given position x, the sum of the width of both strips at this position is constant. The width of the electrodes thereby modulates sensitivity along the length. The detector is again positioned relative to an MLC (not shown) so that an integer number of strip assemblies 30 is associated with each leaf pair. In detail, when each strip assembly 30 is associated with only one leaf pair, the area below the leaf-pair is shared by two collecting electrodes 31 and 32, which are read-out by independent electrometers (comprising at least two amplifiers 22 coupled respectively to the first and second electrodes 31 and 32), and which produce i.sub.1 and i.sub.2 for each IC strip 30. The geometry for a single collecting electrode strip is shown in detail in FIG. 5, from which the formulas for x and x.sub.0 can be derived.

(38) In case of the triangular strips design shown in FIG. 5, the active area of the IC varies while the height h is constant (see FIG. 4). Therefore, the current i in the IC is proportional to the area A. The areas A.sub.1 and A.sub.2 (see FIG. 5) can be expressed using geometric parameters of the strip as well as the parameters x and x.sub.0 as follows:

(39) $\begin{array}{cc}{i}_1=K.Math.A_1=K.Math.x.Math.z=K.Math.x.Math.\frac{z}{L}.Math.x_0& \left(9\right)\\ i_2=K.Math.A_2=K.Math.x.Math.\left(Z-z\right)=K.Math.x.Math.Z\left(1-\frac{x_0}{L}\right)& \left(10\right)\end{array}$
To obtain these equations, some geometric considerations shown in FIG. 6 were made. On the left hand side, it is illustrated that the respective area A.sub.i can be calculated like a square area, since A=A. Additionally, the following relationship between z, Z, L, and x.sub.0 (illustrated on the right hand side) is applied:

(40) $\begin{array}{cc}\frac{Z}{L}=\frac{z}{x_0}Z=\frac{Z}{L}.Math.x_0.& \left(11\right)\end{array}$
From the two currents i.sub.1 and i.sub.2, x and x.sub.0 can be determined as follows

(41) $\begin{array}{cc}x=\frac{i_1+i_2}{K.Math.Z}& \left(12\right)\\ x_0=\frac{L}{2}.Math.\frac{i_1-i_2}{i_1+i_2}& \left(13\right)\end{array}$

(42) FIGS. 7 to 9 illustrate the third embodiment of the present invention. Here, each strip 40 of the array 10 is a single collecting electrode on a common PCB 41. The strips are adjacent and isolated from each other, and arranged at a pitch p.sub.s. A single bias electrode 42 is placed at a constant height h above the strips 40 (i.e. parallel to the strip array). The strips 40 are read-out at both ends by separate electrometers comprising at least amplifiers 22. This solution works as long as the resistivity of the strip is high enough to limit the parasitic current generated by the voltage offset of the readout amplifiers, and/or to obtain signals with a sufficiently high signal to noise ratio 22. FIG. 8 shows the geometry of one strip 40 relative to the bias electrode 42. The resistance R of the strip 40 depends on the length L (or rather x in the general case as indicated in FIG. 8 for the length coordinate) and resistance per unit length A of the resistive material:

(43) $\begin{array}{cc}=\frac{R}{L}\mathrm{and}R=L.Math.& \left(14\right)\end{array}$

(44) When a pencil beam creates free charges in the region above the resistive strip at location x, these currents are collected on the strip and flow towards the electrometer amplifiers through resistors R.sub.1 and R.sub.2 formed by the resistive material. The current causes a voltage drop over those resistors on both sides of the intersection of the pencil beam with the strip. As the inputs to the electrometer amplifiers are at ground potential, this can be described as a parallel circuit with two branches in which a division of currents into partial currents i.sub.1 and i.sub.2 takes place. This principle is visualized in FIG. 9. The sum of i.sub.1 and i.sub.2 always equals the total current.

(45) For the circuit as sketched in FIG. 9, the following general equations can be defined:

(46) $\begin{array}{cc}i=i_1+i_2& \left(15\right)\\ u_1=u_2\mathrm{or}{R}_1.Math.i_1=R_2.Math.i_2& \left(16\right)\\ \mathrm{with}{R}_1=.Math.\left(x-\frac{L}{2}\right)\mathrm{and}{R}_2=.Math.\left(\frac{L}{2}-x\right)& \left(17\right)\end{array}$

(47) In order to obtain i.sub.1 and i.sub.2, Equation 15, 16 and 17 can be transformed:

(48) By applying i.sub.1=ii.sub.2 resp. i.sub.2=ii.sub.1

(49) and
R=R.sub.1+R.sub.2 (total resistance=constant)
the currents i.sub.1 and i.sub.2 can be described as:

(50) 0$\begin{array}{cc}{i}_1=i\frac{R_2}{R}\mathrm{and}{i}_2=i\frac{R_1}{R}.& \left(18\right)\end{array}$

(51) As explained above, the current in an IC is proportional to the volume or area of the chamber (depending on the IC design). Therefore, the derivative must be proportional to the derivative of the volume or area, too:
di=K.Math.dA

(52) where K is a constant containing the sensitivity of the chamber and A is the area of the collecting electrode. In case of the resistive strip design, two dimensions of the chamber volume remain constant: the strip pitch p.sub.s as well as the chamber height h (distance between the strip array and the bias electrode, see FIG. 8). Therefore, only one dimension of the volume V varies, namely, its length, x. di can be determined as
di=K.Math.dx.(19)
The broad beam may be described as the sum of an (infinite) number of pencil beams, each of width dx. The current di (comprised of di.sub.1 and di.sub.2) is created by the charges formed in the area which is hit by the pencil beam of width dx. Therefore, applying Equations 18 and 19, the currents di.sub.1 resp. di.sub.2 induced in the strip can be written as

(53) ${\mathrm{di}}_1=\frac{R_2}{R}.Math.\mathrm{di}\mathrm{and}{\mathrm{di}}_2=\frac{R_1}{R}.Math.\mathrm{di}.$
With use of Equation 17, it can be concluded that

(54) ${\mathrm{di}}_1=\frac{R_2}{R}\mathrm{di}=\frac{.Math.\left(\frac{L}{2}-x\right)}{R}.Math.K\mathrm{dx}$
and analogously

(55) ${\mathrm{di}}_2=\frac{R_1}{R}\mathrm{di}=\frac{.Math.\left(x-\frac{L}{2}\right)}{R}.Math.K\mathrm{dx}.$

(56) The currents i.sub.1 and i.sub.2 created by a broad beam, are obtained as the continuous sum of all the partial currents di.sub.1 respectively di.sub.2. Exemplarily, calculations for i.sub.1 are presented here; calculations for i.sub.2 were performed analogously.

(57) $\begin{array}{cc}{i}_1=\underset{x_0-\frac{x}{2}}{\overset{x_0+\frac{x}{2}}{}}\frac{.Math.\left(\frac{L}{2}-x\right)}{R}.Math.K\mathrm{dx}=& \frac{K}{2R}\left(\frac{L}{2}-x\right){.Math.}_{x_0-\frac{x}{2}}^{2^{x_0+\frac{x}{2}}}=\frac{K}{2R}\left[{\left(\frac{L}{2}-x_0+\frac{x}{2}\right)}^2-{\left(\frac{L}{2}-x_0-\frac{x}{2}\right)}^2\right]\\ =& \frac{K}{2R}.Math.2x\left(\frac{L}{2}-x_0\right)=\frac{K}{L}x\left(\frac{L}{2}-x_0\right)\end{array}$
After integration, i.sub.1 and i.sub.2 can be written as:

(58) $\begin{array}{cc}{i}_1=\frac{K}{L}x\left(\frac{L}{2}-x_0\right)\mathrm{and}& \left(20\right)\\ i_2=\frac{K}{L}x\left(\frac{L}{2}+x_0\right)& \left(21\right)\end{array}$
From these two equations x and x.sub.0 can be determined:

(59) $\begin{array}{cc}x=\frac{i_1+i_2}{K}& \left(22\right)\\ x_0=\frac{L}{2}\frac{i_1-i_2}{i_1+i_2}& \left(23\right)\end{array}$

(60) In any of the embodiments described above, as in any ionization chamber measurement, a measurement is required of the dose rate {dot over (D)} in order to calculate the value of K (see equation (4)). The dose rate is typically obtained from ionization chamber detectors included in the standard equipment of most irradiation accelerator types. In a LINAC the dose rate is usually measured by a redundant system in the form of two independent ionization chambers. When the sensing device of the invention is mounted on such a LINAC, the dose rate measurement obtained from this redundant system may be used to determine K. The integration over time of {dot over (D)} yields the dose D.

(61) In special forms of the second and third embodiment, a large area ionization chamber may be added on the back side of the array of strips. The large area chamber may be realized by providing a large collecting electrode on the backside of the PCB carrying the strip assemblies 30 (in the case of the second embodiment), or the strips 40 (in the case of the third embodiment). The surface of the large collecting electrode is large enough to receive a beam shaped by any configuration of the MLC. The surface preferably covers an area corresponding to the combined area of the strip assemblies 30 or of the strips 40. A second bias electrode is then provided, on the opposite side of the first bias electrode 34 or 42, preferably at the same distance h from the large collecting electrode, but this distance may also be different from h. An electrometer is further provided for measuring a current output from the large area collector electrode when an electric field is created between the large area collector electrode and the second bias electrode, and when a beam passes through the large area chamber. The large area ionization chamber thus created on the backside of the strips allows to obtain a measurement of the dose rate {dot over (D)}.

(62) A sensing device according to the invention comprises a detector equipped with an array of ion chambers according to any of the above-described embodiments or equivalents thereof. In addition, the sensing device comprises a processor configured to determine the position of at least one pair of leaves of the MLC using the currents i.sub.1 and i.sub.2 obtained from the collecting electrode strips. In other words, the processor is configured to calculate x and x.sub.0 on the basis of predefined formulae which depend on the treatment setup and the detector type, as described above for three particular embodiments. This functionality of the processor is clear as such. The design and practical implementation of the processor may be executed according to known technology. The sensing device further comprises a gantry sensor configured to determine at least one position associated with a first gantry angle. In other words, the gantry sensor detects the angular position of the gantry, allowing the determined detected parameters of the MLC leaves (x and x.sub.0) to be compared to prescribed parameters included in a treatment program for specific gantry angles. An angle sensor is known as such and therefore not described in detail.

(63) The procedures outlined above represent a first approximation to obtain the position of the leaves from measured current. This approach is acceptable if scattering and lateral spread of electrons are negligible. If this is not the case, a more refined calculation, using the signal from neighbor strips, is performed. In this case x and x.sub.0 are calculated using the signal from neighbor strips too.

(64) Leaf aperture and flux are not sufficient alone to determine fluence (x,y). If, however, this data is supplemented by independent measurements, e.g. beam profiles measured during LINAC commissioning or periodic QA, then (x,y) can be determined. Fluence data is fed into a dose engine in conjunction with patient images (e.g. CT scan) to determine the actual dose delivered in the patient's anatomy. Lastly, this dose distribution is compared to the planned one.

(65) Many LINACs provide a listing of operational parameters during or after the delivery of a treatment (often called log files). These parameters include the evolution of the collimator configuration, the number of delivered MU and the gantry angle. In principle, these parameters should match the requirements of the plan, although deviations are possible due to malfunctioning, or constraints of real operation that were not accounted in the TPS. The machine determines these parameters with a high resolution (e.g. 0.1 mm in leaf position), probably higher than what is achievable by the proposed 1D system: However, log files are not always suitable for QA purposes because the device under exam also generates the data, and a redundant QA process is not possible.

(66) One additional contemplated use of the present invention's proposed system is to verify that measurements are consistent with log files (within experimental uncertainty), and to use the log files to determine the fluence and the dose into the patient.