A perda dentária tem como consequência a atrofia óssea. Além disso, em regiões posteriores da maxila, a pneumatização do assoalho do seio maxilar (ASM) pode comprometer a instalação de implantes dentários em uma posição protética ideal. O objetivo deste estudo foi avaliar em imagens de tomografia computadorizada de feixe cônico (TCFC) a estimativa da pneumatização (EP) do ASM após perda dentária unitária em região posterior da maxila. 183 imagens de TCFC foram analisadas bilateralmente e divididas em 2 grupos: lado edêntulo (LE) - região edêntula unitária em segundo pré-molar, primeiro ou segundo molares superiores; lado dentado (LD) - região contralateral homóloga à região do LE, com dente presente. Os seguintes parâmetros foram avaliados para comparação LE versus LD: altura do seio (AS), desfecho primário, e área do seio maxilar (SM). A EP (diferença de AS do LE menos LD) e a altura do rebordo (AR - analisada apenas no LE) foram comparadas de acordo com tipo de dente e presença de pneumatização entre as raízes (PR) do LD. Além disso, o rebordo cicatrizado foi categorizado de acordo com a opção terapêutica para maxilar posterior. Os resultados mostraram que o LE apresentou a AS maior que o LD (p<0,05) (6,90 ± 3,15 mm e 6,0 ± 3,01 mm, para LE e LD respectivamente e a EP de 0,9 ± 2,93 mm). Ao separar EP por tipo de dente observou-se que os segundos pré-molares e segundos molares apresentaram valores próximos (1,36 ± 2,43 e 1,21 ± 2,98 mm, respectivamente), enquanto que os primeiros molares apresentaram valores menores (0,55 ± 3,03 mm) (p>0,05). Ao separar os dados de EP de acordo com a presença da pneumatização no LD, observou-se que a maioria dos dentes com esta condição foram os primeiros molares (36 em 43). A EP na presença dessa condição foi de -0,26 ± 2,82 mm, enquanto que na ausência de PR, a EP foi de 1,22 ± 2,99 (p<0,05). Em relação aos parâmetros da crista alveolar, o rebordo cicatrizado apresentou menor altura que às cristas alveolares palatina e vestibular. 54% dos casos de molares que apresentaram pneumatização entre as raízes obtiveram AR < 5mm, sugerindo necessidade de enxerto para uma futura instalação de implante dentário. Portanto, podemos concluir que perdas dentárias em região posterior de maxila favorecem a pneumatização do ASM e que esta varia de acordo com o tipo de dente. Além disso, a identificação de pneumatização entre as raízes parece indicar uma maior tendência à necessidade de cirurgias de levantamento de seio. / The aim of this study was to evaluate the estimation of MSF pneumatization (EP) after single tooth loss in the posterior maxilla region in cone beam computed tomography (CBCT) images. 183 CBCT images were analyzed bilaterally and divided into 2 groups: edentulous side (ES) - edentulous single region in the maxillary second premolar, first or second molars; Tooth side (TS) - contralateral region homologous to the ES region, with tooth present. The following parameters were evaluated for ES vs TS: sinus height (SH), primary outcome, and maxillary sinus area (MSA). The EP (SH difference between ES and TS) and ridge height (RH - analyzed only in ES) were compared according to tooth type and the presence of pneumatization between roots (PR) in TS. In addition, the healed ridge was categorized according to the therapeutic option for posterior maxilla. The results showed that the ES presented a higher SH than the TS (p <0.001) (6.90 ± 3.15 mm and 6.0 ± 3.01 mm for ES and TS respectively, and the EP of 0.9 ± 2.93 mm). When separating EP by tooth type, it was observed that the second premolars and second molars presented close values (1.36 ± 2.43 and 1.21 ± 2.98 mm, respectively), whereas the first molars presented minor values (0.55 ± 3.03 mm) (p>0.001). When separating the PR data according to the presence of pneumatization in the TS, it was observed that the majority of the teeth with this condition were the first molars (36 in 43). PE in the presence of this condition was -0.26 ± 2.82 mm, whereas in the absence of PR, the PE was 1.22 ± 2.99 (p <0.001). Regarding the parameters of the alveolar ridge, the healed ridge presented lower height than the palatal and vestibular alveolar ridges. 54% of the cases of molars that presented pneumatization between the roots obtained RH < 5mm, suggesting the need of grafting for a future installation of dental implant. Therefore, we can conclude that tooth loss in the maxillary posterior region favors the pneumatization of the MSF and it varies according to the type of tooth. In addition, the identification of pneumatization between the roots seems to indicate a greater tendency to the need for sinus lift surgeries.

A respiração predominantemente oral é constantemente citada como um dos fatores associados ao desenvolvimento da deficiência transversal da maxila. A Expansão Rápida da Maxila (ERM) é um excelente método para a correção desta alteração, por meio da abertura da sutura palatina. A tomografia computadorizada por feixe cônico é tem sido descrita como um método preciso de exame de imagens e diante das limitações dos métodos radiográficos convencionais o objetivo deste estudo é avaliar as alterações morfológicas imediatas, decorrentes da ERM, na cavidade nasal e na região da naso e orofaringe, por meio da TCFC. Foram avaliadas 15 pares de imagens tomográfica, correspondentes a 15 pacientes portadores deficiência transversal da maxila, tratados com ERM, que realizaram a TCFC ao início e após o período de contenção de 4 meses. Os resultados encontrados mostram que a cavidade nasal apresenta aumento transversal significativo em seu terço inferior, nas regiões anterior (p=0,045), média (p=0,009) e posterior (p=0,001). Não há alteração significativa do volume (p=0,11), área sagital mediana (p=0,33) e menor área axial (p=0,29) decorrente da ERM na nasofaringe. Há alteração significativa do volume (p=0,05), área sagital mediana (p=0,01) e menor área axial (p=0,04) nos momentos antes e imediatamente após a ERM, na orofaringe. Após análise dos resultados concluímos que a ERM é capaz de aumentar a largura transversal da cavidade nasal, não tendo o mesmo efeito na região da nasofaringe, e que as alterações encontradas na orofaringe podem ser decorrentes de falta de padronização o posicionamento da cabeça e lingual no momento da aquisição da imagem. / The predominantly oral breathing is constantly cited as an etiological factor for the transverse maxillary deficiency. Rapid Maxillary Expansion is an excellent method for the correction of malocclusion, through the opening of the midpalatal sutures. The literature shows that the benefits of this procedure are beyond the dental benefits, and could have repercussions in the upper airways, due to its close relationship with the maxilla. The cone beam computed tomography has been described as is an accurate method of taking pictures and before the limitations of conventional radiographic methods the aim of this study is to evaluate the immediate morphological changes resulting from the ERM, the nasal cavity and the nasal region and oropharynx, through the CBCT. We evaluated 15 patients with maxillary width deficiency were treated with RME, which hosted the CBCT to the beginning and after the retention period of 3 months. The results show that the nasal cavity presents significant increase in cross their lower third, in the anterior (1.08 mm ± 0.15), medium (1.28 mm ± 0.15) and posterior (0.77 mm ± 0.12). No significant change in volume (p=0.11), median sagittal area (p=0.33) and lower axial area (p=0.29) resulting from the RME in nasopharynx. There is significant change in volume (p = 0.05), median sagittal area (p = 0.01) and lower axial area (p = 0.04) before and immediately after the RME in the oropharynx. After analysis and discussion of results in this study, we concluded that RME is able to increase the transverse width of the nasal cavity, not having the same effect in the nasopharynx, and that the changes found in the oropharynx may be due to the lack of positioning standardization of the head and tongue at the time of image acquisition.

<p>Contemporary algorithms employed for reconstruction of 3D volumes from helical cone beam projections are so called non-exact algorithms. This means that the reconstructed volumes will contain artifacts irrespective of the detector resolution and number of projections angles employed in the process.</p><p>It has been proposed that these artifacts can be suppressed using an iterative scheme which comprises computation of projections from the already reconstructed volume as well as the non-exact reconstruction itself.</p><p>The purpose of the present work is to examine if the iterative scheme can be applied to the non-exact reconstruction method PI-original in order to improve the reconstruction result. An important part in this implementation is a careful design of the projection operator, as a poorly designed projection operator may result in aliasing and/or other artifacts in the reconstruction result. Since the projection data is truncated, special care must be taken along the boundaries of the detector. Three different ways of handling this interpolation problem is proposed and examined.</p><p>The results show that artifacts caused by the PI-original method can indeed be reduced by the iterative scheme. However, each iteration requires at least three times more processing time than the initial reconstruction, which may call for certain compromises, smartness and/or parallelization in the innermost loops. Furthermore, at higher cone angles certain types of artifacts seem to grow by each iteration instead of being suppressed.</p>

Contemporary algorithms employed for reconstruction of 3D volumes from helical cone beam projections are so called non-exact algorithms. This means that the reconstructed volumes will contain artifacts irrespective of the detector resolution and number of projections angles employed in the process. It has been proposed that these artifacts can be suppressed using an iterative scheme which comprises computation of projections from the already reconstructed volume as well as the non-exact reconstruction itself. The purpose of the present work is to examine if the iterative scheme can be applied to the non-exact reconstruction method PI-original in order to improve the reconstruction result. An important part in this implementation is a careful design of the projection operator, as a poorly designed projection operator may result in aliasing and/or other artifacts in the reconstruction result. Since the projection data is truncated, special care must be taken along the boundaries of the detector. Three different ways of handling this interpolation problem is proposed and examined. The results show that artifacts caused by the PI-original method can indeed be reduced by the iterative scheme. However, each iteration requires at least three times more processing time than the initial reconstruction, which may call for certain compromises, smartness and/or parallelization in the innermost loops. Furthermore, at higher cone angles certain types of artifacts seem to grow by each iteration instead of being suppressed.

<p>Digital tomosynthesis (DTS) is a quasi-three-dimensional (3D) imaging technique which reconstructs images from a limited angle of cone-beam projections with shorter acquisition time, lower imaging dose, and less mechanical constraint than full cone-beam CT (CBCT). However, DTS images reconstructed by the conventional filtered back projection method have low plane-to-plane resolution, and they do not provide full volumetric information for target localization due to the limited angle of the DTS acquisition. </p><p>This dissertation presents the optimization and clinical implementation of image guided radiation therapy using limited-angle projections.</p><p>A hybrid multiresolution rigid-body registration technique was developed to automatically register reference DTS images with on-board DTS images to guide patient positioning in radiation therapy. This hybrid registration technique uses a faster but less accurate static method to achieve an initial registration, followed by a slower but more accurate adaptive method to fine tune the registration. A multiresolution scheme is employed in the registration to further improve the registration accuracy, robustness and efficiency. Normalized mutual information is selected as the criterion for the similarity measure, and the downhill simplex method is used as the search engine. This technique was tested using image data both from an anthropomorphic chest phantom and from head-and-neck cancer patients. The effects of the scan angle and the region-of-interest size on the registration accuracy and robustness were investigated. The average capture ranges in single-axis simulations with a 44° scan angle and a large ROI covering the entire DTS volume were between -31 and +34 deg for rotations and between -89 and +78 mm for translations in the phantom study, and between -38 and +38 deg for rotations and between -58 and +65 mm for translations in the patient study.</p><p>Additionally, a novel limited-angle CBCT estimation method using a deformation field map was developed to optimally estimate volumetric information of organ deformation for soft tissue alignment in image guided radiation therapy. The deformation field map is solved by using prior information, a deformation model, and new projection data. Patients' previous CBCT data are used as the prior information, and the new patient volume to be estimated is considered as a deformation of the prior patient volume. The deformation field is solved by minimizing bending energy and maintaining new projection data fidelity using a nonlinear conjugate gradient method. The new patient CBCT volume is then obtained by deforming the prior patient CBCT volume according to the solution to the deformation field. The method was tested for different scan angles in 2D and 3D cases using simulated and real projections of a Shepp-Logan phantom, liver, prostate and head-and-neck patient data. Hardware acceleration and multiresolution scheme are used to accelerate the 3D estimation process. The accuracy of the estimation was evaluated by comparing organ volume, similarity and pixel value differences between limited-angle CBCT and full-rotation CBCT images. Results showed that the respiratory motion in the liver patient, rectum volume change in the prostate patient, and the weight loss and airway volume change in the head-and-neck patient were accurately estimated in the 60° CBCT images. This new estimation method is able to optimally estimate the volumetric information using 60-degree projection images. It is both technically and clinically feasible for image-guidance in radiation therapy.</p> / Dissertation

<p>Based on the compressed sensing theorem, we present the integrated software and hardware platform for developing a total-variation based image restoration algorithm by applying prior image information and free-form deformation fields for image guided therapy. The core algorithm we developed solves the image restoration problem for handling missing structures in one image set with prior information, and it enhances the quality of the image and the anatomical information of the volume of the on-board computed tomographic (CT) with limited-angle projections. Through the use of the algorithm, prior anatomical CT scans were used to provide additional information to help reduce radiation doses associated with the improved quality of the image volume produced by on-board Cone-Beam CT, thus reducing the total radiation doses that patients receive and removing distortion artifacts in 3D Digital Tomosynthesis (DTS) and 4D-DTS. The proposed restoration algorithm enables the enhanced resolution of temporal image and provides more anatomical information than conventional reconstructed images.</p><p>The performance of the algorithm was determined and evaluated by two built-in parameters in the algorithm, i.e., B-spline resolution and the regularization factor. These parameters can be adjusted to meet different requirements in different imaging applications. Adjustments also can determine the flexibility and accuracy during the restoration of images. Preliminary results have been generated to evaluate the image similarity and deformation effect for phantoms and real patient's case using shifting deformation window. We incorporated a graphics processing unit (GPU) and visualization interface into the calculation platform, as the acceleration tools for medical image processing and analysis. By combining the imaging algorithm with a GPU implementation, we can make the restoration calculation within a reasonable time to enable real-time on-board visualization, and the platform potentially can be applied to solve complicated, clinical-imaging algorithms.</p> / Dissertation

<p>A bench-top dual cone-beam computed tomography (CBCT) system was developed consisting of two orthogonally placed 40x30 cm<super>2</super> flat-panel detectors and two conventional X-ray tubes with two individual high-voltage generators sharing the same rotational axis. The X-ray source to detector distance is 150 cm and X-ray source to rotational axis distance is 100 cm for both subsystems. The objects are scanned through 200° of rotation. The dual CBCT (DCBCT) system utilized 110° of projection data from one detector and 90° from the other while the two individual single CBCTs utilized 200° data from each detector. The system performance was characterized in terms of uniformity, contrast, spatial resolution, noise power spectrum and CT number linearity. The uniformity, within the axial slice and along the longitudinal direction, and noise power spectrum were assessed by scanning a water bucket; the contrast and CT number linearity were measured using the Catphan phantom; and the spatial resolution was evaluated using a tungsten wire phantom. A skull phantom and a ham were also scanned to provide qualitative evaluation of high- and low-contrast resolution. Each measurement was compared between dual and single CBCT systems.</p><p>Compared with single CBCT, the DCBCT presented: 1) a decrease in uniformity by 1.9% in axial view and 1.1% in the longitudinal view, as averaged for four energies (80, 100, 125 and 150 kVp); 2) comparable or slightly better contrast to noise ratio (CNR) for low-contrast objects and comparable contrast for high-contrast objects; 3) comparable spatial resolution; 4) comparable CT number linearity with R<super>2</super> &#8805; 0.99 for all four tested energies; 5) lower noise power spectrum in magnitude. DCBCT images of the skull phantom and the ham demonstrated both high-contrast resolution and good soft-tissue contrast.</p><p>One of the major challenges for clinical implementation of four-dimensional (4D) CBCT is the long scan time. To investigate the 4D imaging capabilities of the DCBCT system, motion phantom studies were conducted to validate the efficiency by comparing 4D images generated from 4D-DCBCT and 4D-CBCT. First, a simple sinusoidal profile was used to confirm the scan time reduction. Next, both irregular sinusoidal and patient-derived profiles were used to investigate the advantage of temporally correlated orthogonal projections due to a reduced scan time. Normalized mutual information (NMI) between 4D-DCBCT and 4D-CBCT was used for quantitative evaluation.</p><p>For the simple sinusoidal profile, the average NMI for ten phases between two single 4D-CBCTs was 0.336, indicating the maximum NMI that can be achieved for this study. The average NMIs between 4D-DCBCT and each single 4D-CBCT were 0.331 and 0.320. For both irregular sinusoidal and patient-derived profiles, 4D-DCBCT generated phase images with less motion blurring when compared with single 4D-CBCT.</p><p>For dual kV energy imaging, we acquired 80kVp projections and 150 kVp projections, with an additional 0.8 mm tin filtration. The virtual monochromatic (VM) technique was implemented, by first decomposing these projections into acrylic and aluminum basis material projections to synthesize VM projections, which were then used to reconstruct VM CBCTs. The effect of the VM CBCT on metal artifact reduction was evaluated with an in-house titanium-BB phantom. The optimal VM energy to maximize CNR for iodine contrast and minimize beam hardening in VM CBCT was determined using a water phantom containing two iodine concentrations. The linearly-mixed (LM) technique was implemented by linearly combining the low- (80kVp) and high-energy (150kVp) CBCTs. The dose partitioning between low- and high-energy CBCTs was varied (20%, 40%, 60% and 80% for low-energy) while keeping total dose approximately equal to single-energy CBCTs, measured using an ion chamber. Noise levels and CNRs for four tissue types were investigated for dual-energy LM CBCTs in comparison with single-energy CBCTs at 80, 100, 125 and 150kVp.</p><p>The VM technique showed a substantial reduction of metal artifacts at 100 keV with a 40% reduction in the background standard deviation compared with a 125 kVp single-energy scan of equal dose. The VM energy to maximize CNR for both iodine concentrations and minimize beam hardening in the metal-free object was 50 keV and 60 keV, respectively. The difference in average noise levels measured in the phantom background was 1.2% for dual-energy LM CBCTs and equivalent-dose single-energy CBCTs. CNR values in the LM CBCTs of any dose partitioning were better than those of 150 kVp single-energy CBCTs. The average CNRs for four tissue types with 80% dose fraction at low-energy showed 9.0% and 4.1% improvement relative to 100 kVp and 125 kVp single-energy CBCTs, respectively. CNRs for low contrast objects improved as dose partitioning was more heavily weighted towards low-energy (80kVp) for LM CBCTs.</p><p>For application of the dual-energy technique in the kilovoltage (kV) and megavoltage (MV) range, we acquired both MV projections (from gantry angle of 0° to 100°) and kV projections (90° to 200°) with the current orthogonal kV/MV imaging hardware equipped in modern linear accelerators, as gantry rotated a total of 110°. A selected range of overlap projections between 90° to 100° were then decomposed into two material projections using experimentally determined parameters from orthogonally stacked aluminum and acrylic step-wedges. Given attenuation coefficients of aluminum and acrylic at a predetermined energy, one set of VM projections could be synthesized from two corresponding sets of decomposed projections. Two linear functions were generated using projection information at overlap angles to convert kV and MV projections at non-overlap angles to approximate VM projections for CBCT reconstruction. The CNRs were calculated for different inserts in VM CBCTs of a CatPhan phantom with various selected energies and compared with those in kV and MV CBCTs. The effect of overlap projection number on CNR was evaluated. Additionally, the effect of beam orientation was studied by scanning the CatPhan sandwiched with two 5 cm solid-water phantoms on both lateral sides and an electronic density phantom with two metal bolt inserts.</p><p>Proper selection of VM energy (30keV and 40keV for low-density polyethylene (LDPE), polymethylpentene (PMP), 2MeV for Delrin) provided comparable or even better CNR results as compared with kV or MV CBCT. An increased number of overlap between kV and MV projections demonstrated only marginal improvements of CNR for different inserts (with the exception of LDPE) and therefore one projection overlap was found to be sufficient for the CatPhan study. It was also evident that the optimal CBCT image quality was achieved when MV beams penetrated through the heavy attenuation direction of the object. </p><p>In conclusion, the performance of a bench-top DCBCT imaging system has been characterized and is comparable to that of a single CBCT. The 4D-DCBCT provides an efficient 4D imaging technique for motion management. The scan time is reduced by approximately a factor of two. The temporally correlated orthogonal projections improved the image blur across 4D phase images. Dual-energy CBCT imaging techniques were implemented to synthesize VM CBCT and LM CBCTs. VM CBCT was effective at achieving metal artifact reduction. Depending on the dose-partitioning scheme, LM CBCT demonstrated the potential to improve CNR for low contrast objects compared with single-energy CBCT acquired with equivalent dose. A novel technique was developed to generate VM CBCTs from kV/MV projections. This technique has the potential to improve CNR at selected VM energies and to suppress artifacts at appropriate beam orientations.</p> / Dissertation

X-ray scatter in cone-beam computed tomography (CBCT) is known to reduce image quality by introducing image artifacts, reducing contrast, and limiting computed tomography (CT) number accuracy. The extent of the effect of x-ray scatter on CBCT image quality is determined by the shape and magnitude of the scatter distribution in the projections. A method to allay the effects of scatter is imperative to enable application of CBCT to solve a wider domain of clinical problems. The work contained herein proposes such a method. A characterization of the scatter distribution through the use of a validated Monte Carlo (MC) model is carried out. The effects of imaging parameters and compensators on the scatter distribution are investigated. The spectral frequency components of the scatter distribution in CBCT projection sets are analyzed using Fourier analysis and found to reside predominately in the low frequency domain. The exact frequency extents of the scatter distribution are explored for different imaging configurations and patient geometries. Based on the Fourier analysis it is hypothesized the scatter distribution can be represented by a finite sum of sine and cosine functions. The fitting of MC scatter distribution estimates enables the reduction of the MC computation time by diminishing the number of photon tracks required by over three orders of magnitude. The fitting method is incorporated into a novel scatter correction method using an algorithm that simultaneously combines multiple MC scatter simulations. Running concurrent MC simulations while simultaneously fitting the results allows for the physical accuracy and flexibility of MC methods to be maintained while enhancing the overall efficiency. CBCT projection set scatter estimates, using the algorithm, are computed on the order of 1-2 minutes instead of hours or days. Resulting scatter corrected reconstructions show a reduction in artifacts and improvement in tissue contrast and voxel value accuracy.

X-ray scatter in cone-beam computed tomography (CBCT) is known to reduce image quality by introducing image artifacts, reducing contrast, and limiting computed tomography (CT) number accuracy. The extent of the effect of x-ray scatter on CBCT image quality is determined by the shape and magnitude of the scatter distribution in the projections. A method to allay the effects of scatter is imperative to enable application of CBCT to solve a wider domain of clinical problems. The work contained herein proposes such a method. A characterization of the scatter distribution through the use of a validated Monte Carlo (MC) model is carried out. The effects of imaging parameters and compensators on the scatter distribution are investigated. The spectral frequency components of the scatter distribution in CBCT projection sets are analyzed using Fourier analysis and found to reside predominately in the low frequency domain. The exact frequency extents of the scatter distribution are explored for different imaging configurations and patient geometries. Based on the Fourier analysis it is hypothesized the scatter distribution can be represented by a finite sum of sine and cosine functions. The fitting of MC scatter distribution estimates enables the reduction of the MC computation time by diminishing the number of photon tracks required by over three orders of magnitude. The fitting method is incorporated into a novel scatter correction method using an algorithm that simultaneously combines multiple MC scatter simulations. Running concurrent MC simulations while simultaneously fitting the results allows for the physical accuracy and flexibility of MC methods to be maintained while enhancing the overall efficiency. CBCT projection set scatter estimates, using the algorithm, are computed on the order of 1-2 minutes instead of hours or days. Resulting scatter corrected reconstructions show a reduction in artifacts and improvement in tissue contrast and voxel value accuracy.

The objective of this work is to develop an on-line planning and delivery technique for palliative radiotherapy of spinal metastases using a linear accelerator capable of cone-beam CT (CBCT) imaging. This technique integrates all preparation and delivery steps into a single session equivalent to an initial treatment session. The key technical challenges pertaining to the development and implementation of this novel treatment technique are related to CBCT image performance, efficient system integration, development of on-line planning tools and design of novel quality assurance (QA) phantoms and processes. Hardware and software image corrections were first implemented to make CBCT images suitable for target definition and planning. These corrections reduced CBCT non-uniformity and improved CBCT-number accuracy. The on-line treatment technique workflow and the integration of all the subsystems involved in the process were assessed on a customized spine phantom constructed for the study. The challenges related to the routine QA of the highly integrated on-line treatment technique were addressed with the construction and validation of an integral test phantom. This phantom, which contains point detectors (diodes) allows for real-time QA of the entire image guidance, planning and treatment process in terms of dose delivery accuracy. The integral test phantom was also effective for the QA of high-dose, high-precision spinal radiosurgery. Simulation of the on-line treatment technique on patient data showed that the planning step was the one of the most time consuming tasks due predominantly to manual target definition. A semi-automatic method for detection and identification of vertebrae on CBCT images was developed and validated to streamline vertebra segmentation and improve the on-line treatment efficiency. With a single patient setup at the treatment unit, patient motion during the on-line process represents the main source of geometric uncertainty for dose delivery. Spine intra-fraction motion was assessed on CBCT for a group of 49 patients treated with a palliative intent. The use of surface marker tracking as a surrogate for spine motion was also evaluated. Finally, the complete on-line planning and delivery technique was implemented in a research ethics board (REB) approved clinical study at the Princess Margaret Hospital and 7 patients have been successfully treated at the time of this report with this novel treatment approach.