BrachyView: development of an algorithm for real-time automatic LDR brachytherapy seed detection.

BrachyView is a novel in-body imaging system developed to provide real-time intraoperative dosimetry for low dose rate prostate brachytherapy treatments. Seed positions can be reconstructed after in-vivo implantation using a high-resolution pinhole gamma camera inserted into the patient rectum. The obtained data is a set of 2D projections of the seeds on the image plane. The 3D reconstruction algorithm requires the identification of the seed's centre of mass. This work presents the development and techniques adopted to build an algorithm that provides the means for fully automatic seed centre of mass identification and 3D position reconstruction for real-time applications. The algorithm presented uses a local feature detector, speeded up robust features, to perform detection of brachytherapy seed 2D projections from images, allowing for robust seed identification. Initial results have been obtained with datasets of 30, 96 and 98 I-125 brachytherapy seeds implanted into a prostate gel phantom. It can detect 97% of seeds and correctly match 97% of seeds. The average overall computation time of 2.75 s per image and improved reconstruction accuracy of 22.87% for the 98 seed dataset was noted. Elimination processes for initial false positive detection removal have shown to be extremely effective, resulting in a 99.9% reduction of false positives, and when paired with automatic frame alignment and subtraction procedures allows for the effective removal of excess counts generated by previously implanted needles. The proposed algorithm will allow the BrachyView system to be used as a real-time intraoperative dosimetry tool for low dose rate prostate brachytherapy treatments.

BrachyView: Reconstruction of seed positions and volume of an LDR prostate brachytherapy patient plan using a baseline subtraction algorithm.

PURPOSE: BrachyView is a novel in-body imaging system developed with the objective to provide real-time intraoperative dosimetry for low dose rate (LDR) prostate brachytherapy treatments. The BrachyView coordinates combined with conventional transrectal ultrasound (TRUS) imaging, provides the possibility to localise the effective position of the implanted seeds inside the prostate volume, providing a unique tool for intra-operative verification of the quality of the implantation. This research presents the first complete LDR brachytherapy plan reconstructed by the BrachyView system and is used to evaluate the effectiveness of an imaging algorithm with baseline subtraction. METHODS: A plan featuring 98 I-125 brachytherapy seeds, with an average activity of 0.248 mCi, were implanted into a prostate gel phantom under TRUS guidance. Images of implanted seeds were obtained by the BrachyView after the implantation of seeds. The baseline subtraction algorithm is applied as a pixel-to-pixel counts subtraction and is applied to every second projection obtained after the implantation of each needle. Seed positions and effectiveness of the baseline reconstruction in the identification of seeds were verified by a high-resolution post-implant CT scan. RESULTS: A complete brachytherapy plan has been reconstructed with a 100% detection rate. This is possible due to the effectiveness of the baseline subtraction, with its application an overall increase of 11.3% in position accuracy and 8.2% increase in detection rate was noted. CONCLUSION: It has been demonstrated that the BrachyView system shows the potential to be a solution to providing clinics with the means for intraoperative dosimetry for LDR prostate brachytherapy treatments.

BrachyView: initial preclinical results for a real-time in-body HDR PBT source tracking system with simultaneous TRUS image fusion.

A prototype in-body gamma camera system with integrated trans-rectal ultrasound (TRUS) and associated real-time image acquisition and analysis software was developed for intraoperative source tracking in high dose rate (HDR) brachytherapy. The accuracy and temporal resolution of the system was validated experimentally using a deformable tissue-equivalent prostate gel phantom and a full clinical HDR treatment plan. The BrachyView system was able to measure 78% of the 200 source positions with an accuracy of better than 1 mm. A minimum acquisition time of 0.28 s/frame was required to achieve this accuracy, restricting dwell times to a minimum of 0.3 s. Additionally, the performance of the BrachyView-TRUS fusion probe for mapping the spatial location of the tracked source within the prostate volume was evaluated. A global coordinate system was defined by scanning the phantom with the probe in situ using a CT scanner, and was subsequently used for co-registration of the BrachyView and TRUS fields of view (FoVs). TRUS imaging was used to segment the prostate volume and reconstruct it into a three-dimensional (3D) image. Fusion of the estimated source locations with the 3D prostate image was performed using integrated 3D visualisation software. HDR BrachyView is demonstrated to be a valuable tool for intraoperative source tracking in HDR brachytherapy, capable of resolving source dwell locations relative to the prostate anatomy when combined with TRUS.

BrachyView: Combining LDR seed positions with transrectal ultrasound imaging in a prostate gel phantom.

PURPOSE: BrachyView is a novel in-body imaging system which aims to provide LDR brachytherapy seeds position reconstruction within the prostate in real-time. The first prototype is presented in this study: the probe consists of a gamma camera featuring three single cone pinhole collimators embedded in a tungsten tube, above three, high resolution pixelated detectors (Timepix). METHODS: The prostate was imaged with a TRUS system using a sagittal crystal with a 2.5mm slice thickness. Eleven needles containing a total of thirty 0.508U125I seeds were implanted under ultrasound guidance. A CT scan was used to localise the seed positions, as well as provide a reference when performing the image co-registration between the BrachyView coordinate system and the TRUS coordinate system. An in-house visualisation software interface was developed to provide a quantitative 3D reconstructed prostate based on the TRUS images and co-registered with the LDR seeds in situ. A rigid body image registration was performed between the BrachyView and TRUS systems, with the BrachyView and CT-derived source locations compared. RESULTS: The reconstructed seed positions determined by the BrachyView probe showed a maximum discrepancy of 1.78mm, with 75% of the seeds reconstructed within 1mm of their nominal location. An accurate co-registration between the BrachyView and TRUS coordinate system was established. CONCLUSIONS: The BrachyView system has shown its ability to reconstruct all implanted LDR seeds within a tissue equivalent prostate gel phantom, providing both anatomical and seed position information in a single interface.

In vivo skin dose measurement using MOSkin detectors in tangential breast radiotherapy.

The purpose of this study is to measure patient skin dose in tangential breast radiotherapy. Treatment planning dose calculation algorithm such as Pencil Beam Convolution (PBC) and in vivo dosimetry techniques such as radiochromic film can be used to accurately monitor radiation doses at tissue depths, but they are inaccurate for skin dose measurement. A MOSFET-based (MOSkin) detector was used to measure skin dose in this study. Tangential breast radiotherapies ("bolus" and "no bolus") were simulated on an anthropomorphic phantom and the skin doses were measured. Skin doses were also measured in 13 patients undergoing each of the techniques. In the patient study, the EBT2 measurements and PBC calculation tended to over-estimate the skin dose compared with the MOSkin detector (p<0.05) in the "no bolus radiotherapy". No significant differences were observed in the "bolus radiotherapy" (p>0.05). The results from patients were similar to that of the phantom study. This shows that the EBT2 measurement and PBC calculation, while able to predict accurate doses at tissue depths, are inaccurate in predicting doses at build-up regions. The clinical application of the MOSkin detectors showed that the average total skin doses received by patients were 1662±129cGy (medial) and 1893±199cGy (lateral) during "no bolus radiotherapy". The average total skin doses were 4030±72cGy (medial) and 4004±91cGy (lateral) for "bolus radiotherapy". In some cases, patient skin doses were shown to exceed the dose toxicity level for skin erythema. Hence, a suitable device for in vivo dosimetry is necessary to accurately determine skin dose.

Monte Carlo validation and optimisation of detector packaging for spectroscopic dosimetry for in vivo urethral dosimetry during low dose rate brachytherapy.

The urethral mini-dosimeter, developed by the Centre for Medical Radiation Physics, University of Wollongong, uses spectroscopic dosimetry to provide real time point dose measurements along the urethra during low dose rate prostate brachytherapy. Spectroscopic dosimetry uses the measured spectrum of the treatment isotope to estimate the dose rate at the point of measurement, however, the silicon mini-detectors employed in the urethral mini-dosimeter require water proof encapsulation which must be capable of providing electromagnetic shielding without greatly increasing the size of the probe. The introduction of non-tissue equivalent materials within the encapsulation can change the spectrum of radiation incident on the detector, which may influence the application of spectroscopic dosimetry within the urethral dosimeter. The Monte Carlo code Geant4 was adopted to study the effect of encapsulation on the operation of the urethral mini-dosimeter, as well as to determine whether an appropriate thickness of aluminium shielding was possible for electromagnetic screening. The depth dose response and angular dependence of the urethral mini-dosimeter with three thicknesses of aluminium shielding (20, 50, 100 µm) was compared with the urethral mini-dosimeter without aluminium shielding. The aluminium shielding had the effect of increasing the depth dose response (up to 3% within 30 mm and up to 5% within 50 mm), slightly reduced the azimuth angular dependence and slightly increased the polar angular dependence. The 100 µm thick shielding provided the least azimuth angular dependence (±2 %) and provided a polar angular dependence of ±1.4 % within the angles of -45° to 45°.

The investigation of prostatic calcifications using µ-PIXE analysis and their dosimetric effect in low dose rate brachytherapy treatments using Geant4.

Low dose rate brachytherapy is a widely used modality for the treatment of prostate cancer. Most clinical treatment planning systems currently in use approximate all tissue to water, neglecting the existence of inhomogeneities, such as calcifications. The presence of prostatic calcifications may perturb the dose due to the higher photoelectric effect cross section in comparison to water. This study quantitatively evaluates the effect of prostatic calcifications on the dosimetric outcome of brachytherapy treatments by means of Monte Carlo simulations and its potential clinical consequences.Four pathological calcification samples were characterised with micro-particle induced x-ray emission (µ-PIXE) to determine their heavy elemental composition. Calcium, phosphorus and zinc were found to be the predominant heavy elements in the calcification composition. Four clinical patient brachytherapy treatments were modelled using Geant4 based Monte Carlo simulations, in terms of the distribution of brachytherapy seeds and calcifications in the prostate. Dose reductions were observed to be up to 30% locally to the calcification boundary, calcification size dependent. Single large calcifications and closely placed calculi caused local dose reductions of between 30-60%. Individual calculi smaller than 0.5 mm in diameter showed minimal dosimetric impact, however, the effects of small or diffuse calcifications within the prostatic tissue could not be determined using the methods employed in the study. The simulation study showed a varying reduction on common dosimetric parameters. D90 showed a reduction of 2-5%, regardless of calcification surface area and volume. The parameters V100, V150 and V200 were also reduced by as much as 3% and on average by 1%. These reductions were also found to relate to the surface area and volume of calcifications, which may have a significant dosimetric impact on brachytherapy treatment, however, such impacts depend strongly on specific factors in the patient's individual treatment. These factors include the number, size, composition and spatial distribution of calcifications in the prostate as well as the distribution of brachytherapy seeds.

Characterization of a MOSkin detector for in vivo skin dose measurements during interventional radiology procedures.

PURPOSE: The MOSkin is a MOSFET detector designed especially for skin dose measurements. This detector has been characterized for various factors affecting its response for megavoltage photon beams and has been used for patient dose measurements during radiotherapy procedures. However, the characteristics of this detector in kilovoltage photon beams and low dose ranges have not been studied. The purpose of this study was to characterize the MOSkin detector to determine its suitability for in vivo entrance skin dose measurements during interventional radiology procedures. METHODS: The calibration and reproducibility of the MOSkin detector and its dependency on different radiation beam qualities were carried out using RQR standard radiation qualities in free-in-air geometry. Studies of the other characterization parameters, such as the dose linearity and dependency on exposure angle, field size, frame rate, depth-dose, and source-to-surface distance (SSD), were carried out using a solid water phantom under a clinical x-ray unit. RESULTS: The MOSkin detector showed good reproducibility (94%) and dose linearity (99%) for the dose range of 2 to 213 cGy. The sensitivity did not significantly change with the variation of SSD (± 1%), field size (± 1%), frame rate (± 3%), or beam energy (± 5%). The detector angular dependence was within ± 5% over 360° and the dose recorded by the MOSkin detector in different depths of a solid water phantom was in good agreement with the Markus parallel plate ionization chamber to within ± 3%. CONCLUSIONS: The MOSkin detector proved to be reliable when exposed to different field sizes, SSDs, depths in solid water, dose rates, frame rates, and radiation incident angles within a clinical x-ray beam. The MOSkin detector with water equivalent depth equal to 0.07 mm is a suitable detector for in vivo skin dosimetry during interventional radiology procedures.

Radiation dose enhancement at tissue-tungsten interfaces in HDR brachytherapy.

HDR BrachyView is a novel in-body dosimetric imaging system for real-time monitoring and verification of the source position in high dose rate (HDR) prostate brachytherapy treatment. It is based on a high-resolution pixelated detector array with a semi-cylindrical multi-pinhole tungsten collimator and is designed to fit inside a compact rectal probe, and is able to resolve the 3D position of the source with a maximum error of 1.5 mm. This paper presents an evaluation of the additional dose that will be delivered to the patient as a result of backscatter radiation from the collimator. Monte Carlo simulations of planar and cylindrical collimators embedded in a tissue-equivalent phantom were performed using Geant4, with an (192)Ir source placed at two different source-collimator distances. The planar configuration was replicated experimentally to validate the simulations, with a MOSkin dosimetry probe used to measure dose at three distances from the collimator. For the cylindrical collimator simulation, backscatter dose enhancement was calculated as a function of axial and azimuthal displacement, and dose distribution maps were generated at three distances from the collimator surface. Although significant backscatter dose enhancement was observed for both geometries immediately adjacent to the collimator, simulations and experiments indicate that backscatter dose is negligible at distances beyond 1 mm from the collimator. Since HDR BrachyView is enclosed within a 1 mm thick tissue-equivalent plastic shell, all backscatter radiation resulting from its use will therefore be absorbed before reaching the rectal wall or other tissues. dosimetry, brachytherapy, HDR.