INFN Sezione di Catania Universit degli Studi di

INFN – Sezione di Catania Università degli Studi di Catania – Dipartimento di Fisica e Astronomia Roma, 1 st February 2017 Research and Development in Hadrontherapy – Innovation in Radio and Particle Therapy Meeting QBe. RT Qualification of particle Beam Real Time Spokesperson: Giuseppe Gallo (Ph. D student) Participants: D. Lo Presti, D. L. Bonanno, D. Bongiovanni, F. Longhitano, N. Randazzo, E. Leonora, V. Sipala, S. Reito, F. Tommasino

2 Description QBERT is an innovative system for the real-time imaging and monitoring of particle beams for hadron therapy. Basic Diagram It is a proton tracking system composed of a Position Sensitive Detector (PSD) and a Residual Range Detector (RRD) based on scintillating optical fibres and on an innovative read-out strategy and reconstruction algorithm. G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017

3 The Position Sensitive Detector Sensitive area 9× 9 cm 2 (red dashed line) 4 layers of square section Sci-Fi BCF-12 (by Saint-Gobain Crystals) Fibres’ nominal section 500 μm Each layer consists of 160 fibres coated with EMA, divided in 4 optically isolated ribbons Measured spatial resolution ~ 150 μm INFN Patented read-out channel reduction system D. Lo Presti, "Detector based on scintillating optical fibers for charged particle tracking with application in the realization of a residual range detector employing a read-out channels reduction and compression method. " INFN, patent n. WO 2013186798 (2013). Maximum measurable spot size of 2 cm (= 1 ribbon) in counting mode/therapy condition up to 109 particles/sec G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017

4 The Residual Range Detector & Sensitive area 9× 9 cm 2 Stack of 60 layers of square section Sci-Fi BCF-12 (by SGC) with optical isolation between them Fibres’ nominal section 500 μm Each layer consists of 180 fibres(without EMA) coupled to 2 WLS fibers of 1 mm square section Maximum measurable range 36 mm in polystyrene/PVC enough to stop 67 Me. V protons (extendable up to higher energy) G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017

5 DAQ chain FRONT – END READ – OUT Photosensor Amplification and filtering A/D conversion Digital data reading – pre analysis Pre analysis – data transmission Visualization Si. PM Array Operational amplifiers Fast comparator array FPGA RT - Target PC National Instruments So. M G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017

6 Front-End electronics Op-Amp DAC Si. PM Housing Optical Coupling Frames PSD (top) and RRD (bottom) 64 channel Hamamatsu MPPC array S 13361 Fast-Comparator Array Photosensor → Amplification and → A/D Conversion filtering Digital data bus * Front-end boards for the last prototype is equipped with Hamamats MPPC power supply module G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017

7 Read-Out electronics SD-CARD POWER LAN So. M USB System on Module (So. M) Custom So. M Board Courtesy of • FPGA sampling frequency up to 300 MHz • Data transfer to So. M processor by means of DMA • Real-time data reconstruction and filtering • Pre-analysed/Raw data transferred to a PC for visualization/further analysis G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017

8 Test beam road map Test site Date Particles CATANA @ LNS (CT) Oct 2014* Feb 2015* Nov 2015 Protons up to 62 Me. V CNAO (PAVIA) Nov 2014* Apr 2016 Protons up to 250 Me. V Carbon Ions up to 400 a. Me. V TIFPA (TRENTO) Jun 2016 ? ? ? 2017 Protons up to 228 Me. V * Previous prototype test. G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017

9 Experimental Results ü PSD Spatial Resolution Imaging condition – up to 106 particles/sec Calibrated brass collimator with 1 mm diameter holes, whose spacing increased from 1. 5 to 1. 9 mm per direction Collimator hole projection (red circles) and reconstructed centre (blue circles) from acquired image Mean distance between reconstructed and projected hole centres ~ 130 μm ≈ 500/√ 12 μm a priori spatial resolution 2 1 3 G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017

Experimental Results ü PSD X – Y profiles measurement Counting mode/therapy condition up to 109 particles/sec Maximum measurable spot size of ~2 cm (= 1 ribbon, 40 fibres) due to construction features -This number can be changed and optimized for any custom application Profilometer operation mode Examples of profiles at 202 Me. V and 70 Me. V before and after calibration by means of Gaussian fit of raw data (beam spread out at lower energies) G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017 10

Experimental Results ü PSD X – Y profiles measurement Sigma of the Gaussian best fit of the measured X and Y profiles (green) and comparison with IBA Lynx data (red) In the range from 150 to 228 Me. V the value is comparable according to PSD construction specification (beam spot size < ~2 cm) G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017 11

Experimental Results ü 12 PSD Beam monitoring 1 1) Reconstructed position as the centroid of the beam X and Y profile in the energy range from 70 to 228 Me. V @TIFPA (Beam position fixed, Y coordinate variation due to different beam focusing at different energy) 2) Real-time beam position measurement performed @CNAO compared to the beam set fixed pattern 2 G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017

PSD Preliminary Fluence measurement • PSD fluence measurement @ TIFPA as sum of total count for each PSD fiber channel (June 2016) Energy [Me. V] Fluence measurement @ TIFPA by means of a De. Tec. Tor ionization chamber (September 2016) Energy [Me. V] Abs. Diff. IC - PSD • Total Counts ü Total Counts Experimental Results 13 ► Absolute difference between De. Tec. Tor IC and PSD fluence measurement in X and Y direction Energy [Me. V] G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017

14 Experimental Results ü PSD Considerations on Fluence Measurement VMax VTh Vm 0 T VMax’ Vm Do time d VTh 0 0 time G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017 VTh Vm 0 T VMax’ Vm Do time d VTh 0 0 time missing time

Experimental Results ü PSD Considerations on Fluence Measurement ► Calibration coefficients Q power law fit: with adjusted-R 2 = 0. 998. G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017 15

16 Experimental Results ü RRD Range vs Energy Characterization Goal: measurement of the residual energy of the particle (after crossing a target) Theoretical expected relationship R = a∙Ep with p ≈ 1. 7 ÷ 1. 8 for therapeutic protons 1 1) Comparison between simulated range value using Geant 4 and measured range @CATANA with passive beam energy modulation 2 G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017 2) Range measurement extended up to 228 Me. V @ TIFPA by means of RW 3 sheets before the RRD

Experimental Results PSD & RRD → Treatment Plan Verification ü Using the information previous about the position and the energy, available real-time, we can verify the treatment volume on-line. G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017 17

Conferences 2012 IEEE Nuclear Science Symposium and Medical Imaging Conference (2012 NSS/MIC) - Anaheim, CA, USA, 27 Oct - 03 Nov 2012 14 th Vienna Conference on Instrumentation (VCI 16) – Vienna, 15 – 19 February 2016 15 th International Workshop on Radiation Imaging Detectors – Paris, France, 23 – 27 June 2013 14 th Topical Seminar on Innovative Particle and Radiation Detectors (IPRD 16) – Siena, 03 – 06 October 2016 19 th Real Time Conference – Nara, Japan, 26 – 30 May 2014 Publications D. Lo Presti, et al. "A real-time, large area, high space resolution particle radiography system. " Journal of Instrumentation 9. 06 (2014): C 06012. D. Lo Presti, et al. "OFFSET 3: A Real-Time Particle Tracker Based On Scintillating Optical Fibers. " IEEE Transactions on Nuclear Science 62. 3 (2015): 1135 -1141. F. Longhitano, et al. "Design and characterization of a real time particle radiography system based on scintillating optical fibers. " Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment (2016). D. Lo Presti, et al. "OFFSET: Optical fiber folded scintillating extended tracker. " Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 737 (2014): 195 -202. D. Lo Presti, et al. "Design and characterisation of a real time proton and carbon ion radiography system based on scintillating optical fibres. " Physica Medica 32. 9 (2016): 1124 -1134. INVITED D. Lo Presti, et al. "Development of a real-time, large area, high spatial resolution particle tracker based on scintillating fibers. " Advances in High Energy Physics 2014 (2014). INVITED G. Gallo, et al. "QBe. RT: an innovative instrument for qualification of particle beam in real-time. " Journal of Instrumentation 11. 11 (2016): C 11014. D. Lo Presti, et al. "A real time, large area, high spatial resolution tracker based on square scintillating fibers. " Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC), 2012 IEEE, 2012. D. Lo Presti, et al. "Development of a scintillation-fiber detector for real-time particle tracking. " Journal of Instrumentation 8. 04 (2013): P 04015. D. Lo Presti, et al. "Design and characterization of a real time, large area, high spatial resolution particle tracker based on scintillating fibers. " Biomed. Eng. Res 2. 4 (2013): 159 -174. INVITED Thesis 2015, G. Gallo – “Realizzazione e caratterizzazione di un sistema radiografico in tempo reale per particelle cariche ad alta risoluzione basato su fibre ottiche scintillanti” - Master Degree 2014, C. Pugliatti – “Particle scintillating trackers: Design and read-out of real-time, large area, highly segmented detectors” – Ph. D G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017 2013, P. Barone – “PREDATE – Particle Residual Energy and Tracker Enhancement” – Bachelor Degree 2013, G. Petringa – “Studio per la realizzazione di una radiografia con protoni in tempo reale” – Bachelor Degree 18

19 Conclusions PSD / Profilometer and Residual Range Detector are ready to use in image and treatment conditions Modularity of each detector permits to customize the system for any specific application Measurement campaign and data analysis almost complete An alternative solution for direct fluence measurement is under study and will be tested next time @ TIFPA Work in progress for technology transfer (suggestions for future uses will be very welcome) G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017

THANK YOU for your ATTENTION! For more information, please contact: Prof. Domenico Lo Presti domenico. lopresti@ct. infn. it skype: domenico. lo. presti mobile: +39 -3474975203 Office: +39 -0953785413 20

Fundings IRPT (Innovation in Radio- and Particle- Therapy) Progetto Premiale INFN Aknowledgements M. Ciocca, M. Pullia CNAO Foundation (Pavia) (Centro Nazionale di Adroterapia Oncologica) M. Durante, F. Tommasino TIFPA (Trento) (Trento Institute for Fundamental Physics and Applications) P. Cirrone, F. Romano CATANA (LNS, Catania) (Centro di Adro. Terapia e Applicazioni Nucleari Avanzate) For more information, please contact: Prof. Domenico Lo Presti domenico. lopresti@ct. infn. it skype: domenico. lo. presti mobile: +39 -3474975203 Office: +39 -0953785413 21

Additional material G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017 22

23 Experimental Results 23 mm 10 mm 15 mm PSD & RRD → Particle Radiography (up to 106 particles/sec) 46 mm ü Range measurement of the single particle 5 mm Beam pipe Target There was a PMMA range shifter between the beam pipe and the target (10, 68 mm) 15 mm PSD Position of each particle crossing the target (PSD) Range to Energy conversion with calibration curve RRD 10 mm Hit area 5 mm + 10. 68 mm range shifter G. Gallo – RDH / IRPT Meeting – Roma, 1 February 2017 160 x 160 matrix with particle mean energy for the corresponding pixel Numerical integration of the energy loss to obtain the thickness of crossed target

Radiography Data analysis

25 Hamamatsu Si. PM array & BCF-12, WLS by SGC 64 channel Hamamatsu MPPC array S 13361 Sci-Fi emission spectrum WLS absorption and emission spectrum Si. PM Photon detection efficiency

Hamamatsu S 13361 Datasheet 26

Resolution q PSD: Differences between measured and real position have a Gaussian Distribution q RRD: 27

Read-out channel reduction system (INFN patent) For classical reading of the PSD will be necessary 160 × 4 = 640 ch. For each direction, X o Y: - 4 ribbon ch (16 mini-ribbon ch) - 40 fiber ch. 112 ch instead of 640, with coincidence reading and without loss of information! 28