2D large area ionisation planar silicon detector

 Even though clinical dosimetry in radiotherapy is well known matter, the introduction of high conformal therapy set new spectra of problems to be solved. The main problems posed by high conformal radiotherapy modalities (IMRT, stereotactic treatments and proton therapy) are mainly  due to the small radiation fields with high dose gradients, to the variation in space and time of the dose rate and to the variation in space and time of the beam energy spectrum.

In particular, the accurate determination of the 2D absorbed dose distribution requires detectors with high spatial resolution (i.e. small sensor size and short distance between matrix elements); response independent of the dose rate, of the energy and LET radiation, fast and stable in time; good linearity and high dynamic range.

The most spread devic   es for dose distribution verification are here recalled.

Film dosimetry is a well established method for dose distributions verification in phantoms [1], [2], [3]. It permits high spatial resolution, limited only by the reading system. On the other hand, passive planar dosimeters show a rather small dynamic range and non-linear dose response (conventional X-ray films) and are expensive, also in terms of labor-intensive readout (dye films).

On the market, tools are also available that allow a 2D verification with matrices of ionization chambers or silicon diodes. In this case, the response is immediately available in a digital form and can be compared to the TPS predictions. The spatial resolution and granularities are worse than with films.

The most widely spread 2D arrays of Silicon diodes is the Sun Nuclear MapCHECK.
(http://www.sunnuclear.com/products/mapcheck/mapcheck.asp)
This (Model 1175, MPC) consists of 445 n-type silicon diodes arranged in a 22 x 22 cm² matrix. Each detector has an active area of 0.8 x 0.8 mm². It has been found that the used Silicon is radiation hard. The main problem arising from the use of this array is that the centre-to-centre spacing is high (7-14 mm). A dosimetric characterization of the MPC can be found in [4].

The PTW 2D-Array (http://www.ptw.de) consists of 729 equally spaced ionization chambers distributed in an area of 27 x 27 cm². Each detector covers an area of 5 x 5 mm² and the measuring depth is at 5 mm water equivalent. The sensitive volume of each chamber is 0.125 cm³. The center-to-center distance between two chambers is 10 mm. The only problem arising with the use of this array is again the spatial resolution (sensor size and granularity).

A different detector for the 2D verification of photon beams has been developed at the Torino University/INFN and then engineered by Scanditronix-Wellhofer in the framework of MAESTRO project (http://www.scanditronix-wellhofer.com/I_mRT_MatriXX.957.0.html).

The device (MatriXX) is a pixel-segmented ionization chamber. In summary it consists of a 32 x 32 matrix of 1024 cylindrical ionization chambers arranged in a square of 24 x 24 cm² area. Each chamber is 4.5 mm in diameter and 5 mm high. The centre-to-centre spacing is 7.62 mm. The sensitive volume of each single ionization chamber is about 0.07 cm³. A detailed description and results of the dosimetric characterization can be found in [5].

Also the EPID (Electronic Portal Imaging Devices) systems, routinely used for the verification of the geometrical accuracy of the treatment, could be employed for dosimetric applications (pretreatment verification of TPS dose distribution or in vivo portal dosimetry) [6]. The features of the device are promising (very large sensitive area, high spatial resolution), but several points require further investigation before the practical use [7]. The main issues remain the method for correcting the pixel-to-pixel variations and, even more, the effects of ghosting and lag on the measurement. These effects can be accentuated for IMRT fields due to the potential range of intensities in a given field. At the moment no commercial EPID is available with the suitable characteristics for dosimetric applications.

With the aim of developing a dosimetric system adequate for 2D pre-treatment dose verifications, DFC of Florence University designed a modular detector, based on a monolithic silicon segmented sensor, with a n-type implantation on an epitaxial p-type layer. The epitaxial layer is grown on a p-type Magnetic Czochralski silicon wafer.

First, a 6.29 x 6.29 cm² module has been manufactured and assembled with the read-out electronics. Each pixel element is 2 x 2 mm² and the distance center-to-center is 3 mm. The sensor is composed of 21 x 21 pixels, that is 441 channels all together.

The sensor is designed to work without polarization. Through a pitch adapter (b) and a printed circuit board (c), the sensor is connected to the mother board where the read-out electronics is located (d). The circuits are distributed on several boards (e), each containing the electronics necessary to read 16 channels.

Each channel is read by a current integrator based on an operational amplifier with low polarization current (LMC6084). The integrators outputs are routed, through a multiplexer reducing by a factor 8 the number of output channels, to the buffers driving the signals toward an external (f) DAQ device (NI 6071E). The integration capacitance value (Panasonic ECHU(X)) is C=1.00±5% nF. The integrator outputs are read at regular intervals and, after each reading, capacities are discharged in order to prevent circuit saturation. The 441 channels are read in a time T_r< 1 ms at the end of the integration period T_i. The acquisition period T=T_r+T_i can be varied between 0.1 s and 4 s.

The sensor is covered by a Bakelite sheet (g) to shield the light, and by a PMMA plate (h).

Upon the electronics there are a metallic lid and a lead shield (i) in order to prevent the scattered photons to damage the integrated circuits.

Figure 1 : Exploded view of the prototype. The 9 element detector is located on the right.

The detector is now under dosimetric characterization with high energy photon beams from linear accelerator.
The last realization of the electronic equipment is shown in the following photograph.

Figure 2 : Exploded view of the new and more compact prototype. The detector is located on the right.

In the final configuration nine modules will be assembled together in order to cover an area of 19 x 19 cm², while integrated electronics will replace a large part of the circuitry (f), in order to obtain a compact and easy-to-handle device.

[1] Ting J. Y., Davis L. W., Dose verification for patients undergoing IMRT, Med. Dosim. 26 (2), 205-213, 2001.

[2] Xing L., Curran B., Hill R., Holmes T., Ma L., Forster K.M., Boyer A.L., Dosimetric verification of a commercial inverse treatment planning system, Phys. Med. Biol. 44 (2), 463-478, 1999.

[3] Bucciolini M., Banci Buonamici F., Casati M., Verification of IMRT fields by film dosimetry, Med. Phys. 31 (1), 161-168, 2004.

[4] Jursinic P.A., Nelms B.E., A 2-D diode array and analysis software for verification of intensity modulated radiation therapy delivery, Med Phys.,30 (5), 870-879, 2003.

[5] Amerio S., Boriano A., Bourhaleb F., Cirio R., Donetti M., Fidanzio A., Garelli E., Giordanengo S., Madon E., Marchetto F., Nastasi U., Peroni C., Piermattei A., Sanz Freire C.J., Sardo A., Trevisiol E., Dosimetric characterization of a large area pixel-segmented ionization chamber, Med. Phys. 31 (2), 414-420, 2004.

[6] El-Mohri Y., Antonuk L.E., Yorkston J., Jee K.W., Maolinbay M., Lam K.L., Siewerdsen J.H., Relative dosimetry using active matrix flat-panel imager (AMFPI) technology, Med. Phys. 26 (8), 1530-1541,1999.

[7] Moran J.M., Roberts D.A., Nurushev T.S., Antonuk L.E., El-Mohri Y., Fraass B.A., An active matrix flat panel dosimeter (AMFPD) for in-phantom dosimetric measurements, Med. Phys. 32(2), 466-472, 2005.