Browsing by Department "Medical Physics"

Purpose: This project aimed to establish a method for computing 3D cardiac motion given continuous 2D-Cine MRI frames as the inputs. This approach would be useful for continuously monitoring cardiac and respiratory motion during MR-guided cardiac radiation therapy, and thus supporting radiation delivery guidance and gating.Methods: 5D-MRI datasets of seven patients, with each consisting of 3D spatial volumes of the cardiac cycle and respiratory cycle, were used for quantitative evaluation of the heart motion due to respiratory and cardiac movements. This was achieved through deformable image registration (DIR). Subsequently, principal component analysis (PCA) was performed on the computed deformation vector fields (DVF) to extract scores that effectively represent the characteristics of the DVFs. A deep learning model was then trained to predict the cardiac motion PCA scores given the inputs of 2D-Cine MRI. The predicted PCA scores were then transformed into 3D DVFs, which were then used to track 3D target motion. Results: The model’s performance was quantitatively evaluated on ground truth data that were withheld from model training. Across all 7 subjects, the average 3D DVF prediction errors for the heart region consistently remained around 0.3 ± 0.1mm. The predicted target motion, computed from the predicted DVFs, was visually evaluated, and found to be satisfactory. Conclusion: The developed method demonstrated promising potential in accurately computing and tracking real-time 3D cardiac motion given 2D-Cine MRI inputs. This approach presents a viable solution for continuously monitoring the 3D cardiac and respiratory motion of the heart during MR-guided cardiac radiation therapy.

Purpose: To develop an automatic collimator setting optimization algorithm to improve dosimetric quality of pancreas Volumetric Modulated Arc Therapy (VMAT) plans for Stereotactic Body Radiation Therapy (SBRT).

Methods: Fifty-five pancreas SBRT cases were retrospectively studied. Different from the conventional practice of initializing collimator settings manually, the proposed algorithm simultaneously optimizes the collimator angles and jaw positions which are customized to the patient geometry. This algorithm includes two key steps: an iterative optimization algorithm via simulated annealing that generates a set of collimator settings candidates, and a scoring system that choose the final collimator settings based on organs-at-risk (OARs) sparing criteria and dose prescription. The scoring system penalizes 3 factors: 1) jaw opening ratio on Y direction to X direction; 2) unmodulated MLC area within the jaw aperture in a dynamic MLC sequence; 3) OAR shielding capability by MLC with MLC aperture control constraints. For validation, the other 16 pancreas SBRT cases were analyzed. Two dual-arc plans were generated for each validation case, an optimized plan (Planopt) and a conventional plan (Planconv). Each plan was generated by a same set of auxiliary planning structures and dose-volume-histogram (DVH) constraints in inverse optimization. Dosimetric results were analyzed and compared. All results were tested by Wilcoxon signed-rank tests.

Results: Both plan groups had no statistical differences in target dose coverage V95% (p=0.84) and Root Conformity Index (p=0.30). Mean doses of OARs were improved or comparable. In comparison with Planconv, Planopt reduced maximum dose (D0.03cc) to stomach (-49.5cGy, p=0.03), duodenum (-63.5cGy, p<0.01), and bowel (-62.5cGy, p=0.01). Planopt also showed lower modulation complexity score (p=0.02), which implies its higher modulation complexity of the dynamic MLC sequence.

Conclusions: The proposed collimator settings optimization algorithm successfully improved dosimetric performance for dual-arc VMAT plans in pancreas SBRT. The proposed algorithm was demonstrated with great clinical feasibility and readiness.

Purpose: A deep learning framework based on a conditional Generative Adversarial Network (cGAN) was developed to synthesize high-contrast Magnetization Prepared Rapid Gradient Echo (MP-RAGE) images from common spin-echo MR imaging sequences. This framework utilizes combinations of inputs from T1-weighted (T1-w), T2-weighted (T2-w), and Proton Density-weighted (PD-w) images. The primary objective was to augment the diversity of clinical data by capitalizing on the inherent advantages of MP-RAGE imaging, such as superior contrast, while mitigating its susceptibility to metal artifacts.Methods and Materials: A cGAN image synthesis model, incorporating a U-Net-based generator and a Patch GAN discriminator, was developed. The training was conducted across four distinct configurations, employing combinations of T1-w, T2-w, and PD-w images as inputs to synthesize MP-RAGE images, with and without Proton Density (PD) information, designated as PD-0 and PD-1, respectively. For training, data from 51 patients, comprising 8,160 slices, were used, following a training-to-validation ratio of 90:10. For prediction, data from 14 patients, comprising 2,240 slices, were utilized. The efficacy of the synthesized MP-RAGE images was evaluated using a suite of quantitative metrics, including Mean Absolute Error (MAE), Normalized Cross-Correlation (NCC), Percentage Mutual Information (PMI), and Structural Similarity Index (SSIM). Additionally, a Freesurfer brain segmentation task was performed on both synthesized and ground truth brain images, with the fidelity of synthesized images being indirectly assessed by the calculated Dice coefficient. Results: It was observed that the cGAN-synthesized MP-RAGE images exhibited comparable contrast to the ground truth in the axial view. A decrease in input channel numbers resulted in diminished contrast between certain anatomical structures in the synthetic MP-RAGE images, albeit within an acceptable range. The MAE approached (0.02±0.01), the PMI for two Three-in-One-out synthesis approached(0.76±0.07), the NCC was about (0.91±0.05), and the SSIM was about (0.9±0.1). The Freesurfer segmentation results showed desirable Dice coefficients (mostly above 0.8) for different kinds of inputs, except the One-in-One-out T1-w synthesis. Conclusion: The cGAN framework developed in this study has proven to be a robust and versatile tool for synthesizing high-contrast MP-RAGE images, even in scenarios with single-channel input images. The Freesurfer segmentation results demonstrated that the synthesized MP-RAGE images are highly similar to the ground truth in segmentation tasks, underscoring the potential clinical and research value of the proposed image synthesis model.

Brain metastases are a common complication of many types of cancer, including lung, breast, and melanoma. Approximately 30-40% of patients develop brain metastases that originate from primary systemic tumors during the course of cancer treatment. One treatment method is a LINAC-based single-isocenter multiple-target (SIMT) stereotactic radiosurgery (SRS). High plan quality has been one of the important goals in radiotherapy treatment planning. Generation of a high quality SRS treatment plan, particularly a SIMT plan, usually requires planners’ extensive planning experience, multiple runs of planning and trial-and-error, and frequent communication among planners, physicians and other radiation oncology team members. In clinical practice with potentially limited resources, SIMT SRS planning could be time-consuming and may have large variations in plan dosimetric quality. Therefore, an estimation of achievable dosimetric outcome can help reduce plan quality variation and improve planning efficiency. Assuming 20Gy in a single fraction of treatment, the volume of normal brain tissue receiving 10Gy (V50%), 12Gy (V60%), and 13Gy (V66.7%) are known predictors of brain tissue toxicity, or radionecrosis. We developed deep learning networks for the prediction of V50%, V60%, and V66.7% based on each patient’s target delineation. A prediction of achievable V10Gy, V12Gy, and V13Gy (assuming 20Gy x 1fx) can assist physicians in the determination of fractionation schemes (i.e., single fx vs. multiple fx). Such predictions can be used as guidelines for planners to generate a SIMT plan more rapidly with reduced dosimetric variability. A key technical innovation of this work is the spherical projection design: by projecting target distribution on a spherical surface, the target distribution in 3D is collapsed to a polar-azimuthal angular distribution map. This transformation enables a dimensional reduction for deep learning input without losing volumetric information. Our results indicate promising potential but there is a need for further work to improve the accuracy of our predictions.

Recently, there has been a lot of effort placed on monitoring patient dose from medical procedures. The majority of people's concern has been focused on computed tomography because of the higher amounts of patient dose associated with CT exams. Our institution currently has dose monitoring programs for CT, nuclear medicine, and digital projection radiography. However, there is currently no established way to track patient dose for computed radiography. The current method of tracking computed radiography is to track exposure indicators which are not directly meaningful to patient dose. In order to address this issue, I have expanded on the exposure indicator tracking by adding a conversion for estimated patient effective dose in computed radiography.

Purpose: The purpose of this work is to characterize three novel formulations of a radiochromic material Presage and identify optimal imaging procedures for accurate 3D dosimetry. The dosimetric qualities of interest were studied for each formulation of Presage dosimeter in the context of accurate 3D dosimetry. The formulation of Presage showing the most promise is compared to a clinical 3D quality assurance device to investigate the accuracy of a complex state-of-the-art brain IMRT treatment.

Methods and Materials: Three novel formulations of Presage were studied for their temporal stability, sensitivity, linearity of dose response, and feasibility of absolute dose calibration in large volume dosimeters (1 kg) with small volume cuvettes (4g). Large cylindrical dosimeters with 11 cm diameter and 10 cm height were irradiated with 5 2x2 cm fields on the upper flat surface with 3 distinct dose levels (3, 6 and 9.5 Gy, representing low, medium and high). This irradiation pattern is used to determine the dosimetric characteristics mentioned above and was chosen because of its repeatability and it lends to simple measurements of linearity and sensitivity. Measurements were taken at various time points from 0 hours to 24 hours post-irradiation using the high resolution (6.45 m pixels) Duke Medium-Sized Optical-CT Scanner (DMOS) and reconstructed with a Matlab-based reconstruction GUI created in-house. Analysis of the pertinent dosimetric characteristics was performed in the GUI. A comprehensive end-to-end QA test was performed on the optimal formulation using optimal scan timing determined from the formulation studies described above. A 5-field IMRT plan was created for head treatment. The plan was delivered both to a head phantom containing a Presage insert, and to the Delta4 QA device. Comparison of both delivered distributions together with the Eclipse predicted dose distribution enabled investigation of the accuracy of the delivery, and the consistency of independent measurement devices.

Results: The DEA-1 formulation showed up to 10% variation from 0-2 hours post-irradiation, but showed excellent temporal stability (<2% variation) between 3-7 hours post irradiation, and maintained good stability until 24 hours post-irradiation (up to 3% variation). The DEA-2 also showed up to 10% variation from 0-2 hours post-irradiation. The DEA-2 formulation then showed good stability (up to 2.1% variation) from 3-7 hours, but optical density values dropped by up to 11% after 24 hours. The DX formulation did not maintain stability of optical density for any significant time with values decreasing by ~20% by the 24-hour time point and optical density decreasing at different rates for different dose levels. Linearity of dose response was good for all formulations with an R2 value > 0.99. Gamma analysis with criteria of 3%/2mm was performed on two irradiations of the 5-field pattern on DEA-1 formulation. Voxel passing rates were 96.68% and 97.96%. Comparison of the DEA-1 formulation large dosimeter was done with small volume cuvettes of the same formulation and batch. Sensitivity of the large dosimeter was less than half the sensitivity of the cuvettes. For clinical 3D QA comparison, the DEA-1 formulation was used because it had optimal performance showed the most promise for accurate 3D dosimetry. Line dose profiles showed that Presage compared very well with the Eclipse calculation and had a much better 3D gamma passing rate for 3%/3mm criteria than the Delta4 (>99% vs 75%).

Conclusions: The DEA-1 formulation shows the most promise because of its temporal stability and linearity of dose response. The optimal imaging window for this formulation was determined to be 3-24 hours post-irradiation. The DEA-2 and DX formulation also showed potential for accurate dosimetry. The optimal imaging window for the DEA-2 formulation was determined to be 2-6 hours post-irradiation. The optimal scan time for the DX formulation was determined to be immediately post-irradiation. The amount of accuracy loss depending on the scan time is dependent on the formulation and when the dosimeter is scanned. Line dose profiles and gamma analysis results from the comparison of Presage and Eclipse calculation provide strong validation of the accuracy of the IMRT treatment delivery. Comparison of Presage to the Delta4 show the Delta4 to be somewhat lacking in its ability to calculate 3D dose in the phantom/Presage geometry.

Purpose: Current Monte Carlo simulations modeling space radiation exposure typically use simplistic human phantoms with low anatomical detail and minimal variability in physical characteristics. This thesis describes the development of a GEANT4-based simulation framework (EVEREST – Evaluation of Variable-Environment Radiation Exposure during Space Travel) that incorporates highly realistic and diverse 4D extended cardiac-torso (XCAT) digital phantoms, combined with advanced NASA models of planetary atmospheres, spaceflight trajectories, and space radiation spectra, to evaluate radiation exposure in interplanetary missions and on planetary habitats.

Methods: Galactic cosmic radiation spectra as a function of time and radial distance from the Sun were modeled using the Badhwar-O’Neill 2020 model, while the Van Allen belt spectra were modeled using the AE-8/AP-8 models, and solar particle event spectra could be selected from historical data. The magnetic field input to the AE-8/AP-8 model was generated using the 13th generation International Geomagnetic Reference Field. Planetary atmospheres were modeled using NASA Global Reference Atmospheric Models, which provide mean atmospheric data for any altitude, latitude, longitude, and time, and the effect of Earth’s magnetic field was accounted for using a geomagnetic cutoff rigidity algorithm. Planetary orbits, trajectories, and relative positions of objects in the Solar System were determined using the NAIF SPICE observation geometry information system. Finally, highly detailed extended cardiac-torso (XCAT) digital phantoms were integrated into EVEREST in order to accurately model radiation exposure to individual organs. XCAT phantoms model over 100 segmented structures, range in age from neonate to 78 years, and cover various combinations of height, weight, and BMI. The EVEREST framework itself was designed using a novel lookup table method, in which different stages of particle propagation were divided into separate simulations, which are then convolved in post-processing.

Results: EVEREST was validated against personal radiation dosimeter data collected by the lunar module pilot on the Apollo 15 mission and also flux data from the Mars Science Laboratory Radiation Assessment Detector (RAD). Simulation results were found to agree very well with dosimeter readings by the Apollo 15 command module pilot. Comparison of Martian surface particle fluxes simulated by EVEREST to RAD data demonstrated an agreement to within an order of magnitude, with the best agreement seen for protons, He4, Z=6-8, Z=14-24, and Z>24. Finally, as a proof of concept, EVEREST was used to evaluate radiation exposure to a population of eight XCAT phantoms (3 adult and 1 pediatric, male and female) under three different nominal shielding configurations on the surface of Mars (unshielded, 50 cm thick ice, and 50 cm thick Martian regolith) at four different timepoints during the day (12 am, 6 am, 12 pm, and 6 pm). Using the federal yearly occupational dose limit of 50 mSv (effective dose) as a metric, it was found that the phantoms evaluated would reach this limit within 70.9 – 83.8 days unshielded, 139.2 – 161.2 days with 50 cm ice shielding, and 188.1 – 235.7 days with 50 cm Martian regolith shielding, if terrestrial radiation protection standards were to be applied. The results revealed that the brain receives one of the highest organ doses in the body and that unshielded radiation exposure is lowest at midnight when analyzed across all phantoms. Based on these findings, it is recommended that extra care be taken to provide additional radiation shielding in astronauts’ helmets and that extended forays outside of the habitat be planned for late evening to reduce the biological impact of radiation exposure.

Conclusion: EVEREST is a tested and validated framework for accurate estimation of total body and organ dose in space. EVEREST’s geometric versatility makes it ideal for evaluating doses to diverse populations of XCAT phantoms within different types of planetary habitats and spacecraft, enabling optimization of mission planning with respect to radiation exposure in the near future. The model has currently been validated for Lunar and Martian missions, and the framework can be applied to any space travel mission or planetary mission where the atmospheric models for that planet are available.

Purpose:

Our goal was to test a novel concept approximating organ dose measurements using the single mean energy of the two sources in dual-energy (DE) CT environment. Therefore, the purpose of this study was two-fold: (1) To obtain experimental validation of dose equivalency between MOSFET and ion chamber (as gold standard) under a dual-energy environment; (2) To estimate the effective dose (ED) using MOSFET detectors and an anthropomorphic phantom in DE CT scans.

Materials and Methods:

A commercial dual source CT (DSCT) scanner was employed for the study. The scanner was operated at 80kV/140kV (Sn added) using an abdomen/pelvis scanning protocol. A five-phase approach was used. Specific goals for each phase are as follows: (1) Characterize the mean energy from the combined clinical 80kV/Sn140kV beams; (2) Estimate the f-factor for tissues from the mean energy; (3) Calibrate the MOSFET detectors using the mean energy; (4) Validate MOSFET calibration with a CTDI phantom; (5) Measure organ doses for a typical abdomen/pelvis scan using a male anthropomorphic phantom and derive ED using ICRP 103 tissue weighting factors. For validation of dose equivalency, a MOSFET detector and ion chamber measured the dose at the center cavity of a CTDI body phantom. A student t-test was used to determine if the difference between the two was statistically significant.

Results:

The mean energy was calculated to be 67 kVp based on the corresponding spectra for the clinical DE beams. Using the Mean Energy Method, the tissue dose in the center cavity of the CT body phantom was 2.08 ± (2.70%) cGy with an ion chamber and 2.20 ± (4.82%) cGy with MOSFET respectively with a percent difference of 5.91% between the two measurements. The results (p = 0.15) showed no statistically significant difference. ED for DE abdomen/pelvis scan was calculated as 5.01 ± (2.34%) mSv by the MOSFET method and 5.56 mSv by the DLP method respectively.

Conclusion:

There has been no physical method to measure organ doses in DE CT scans. We have developed and validated a novel approach, the Mean Energy Method - for organ dose estimation in DE CT scans. ED from the anthropomorphic phantom compared well (within 11%) between the MOSFET method and DLP method.

Purpose: Radiosurgery has become a widely used procedure in the treatment of both solid tumors and secondary metastases in the brain. In cases with multiple brain lesions, isocenters are typically set up for each target, a process which can take hours and become very uncomfortable for the patient. Recently, multifocal treatments with a single isocenter have emerged as a solution. With the high doses delivered to small regions during radiosurgery, the importance of treatment verification is paramount, especially when delivering high doses to regions off isocenter.

Methods: A 5-arc RapidArc radiosurgery plan with a single isocenter and 5 targets was used to treat a dosimeter placed within a RPC-type head and neck phantom. The treatment was delivered five times at varying prescription doses, depending on the sensitivity of the PRESAGE dosimeter used. The delivered dose distribution was measured using an in-house optical-CT system and compared to the Eclipse-planned dose distribution using dose volume histograms and Gamma analysis.

Results: Reasonable dose agreement was measured between the majority of the dosimeters and the Eclipse plan (80-85% pass rate at 5%/3 mm Gamma critera). The failing voxels were located on the periphery of the dosimeter at regions of extremely high or low dose, suggesting a dose dependent stability of the PRESAGE formulation. The formulation with the best temporal stability had a much higher Gamma pass rate of 98% at 3%/2mm criteria.

Conclusions: The potential of accurate delivery of the complex radiosurgery plan was demonstrated with one of the three formulations of PRESAGE. While agreement was worse in the other formulations, the problem seemed to be an easily-fixable stability issue, resulting in improper scaling of doses. Replication of the most stable formulation would provide an excellent tool for verification of radiosurgery treatment delivery and other complex procedures.

Purpose: Ensuring that tumor motion is within the radiation field for high-dose and high-precision radiosurgery in areas greatly influenced by respiratory motion. Therefore tracking the target or gating the radiation beam by using real-time imaging and surrogate motion monitoring methods are employed. However, these methods cannot be used to depict the effect of respiratory motion on tumor deviation. Therefore, an investigation of parameters for method predicting the tumor motion induced by respiratory motion multiple steps ahead of real time is performed. Currently, algorithms exist to make predictions about future real-time events, however these methods are tedious or unable to predict far enough in advance.

Methods and Materials: The algorithm takes data collected from the Varian RPM$ System, which is a one-dimensional (1D) surrogate signal of amplitude versus time. After the 1D surrogate signal is obtained, the algorithm determines on average what an approximate respiratory cycle is over the entire signal using a rising edge function. The signal is further dividing it into three components: (a) training component is the core portion of the data set which is further divided into subcomponents of length equal to the input component; (b) input component serves as the parameter searched for throughout the training component, (c) analysis component used as a validation against the prediction. The prediction algorithm consists of three major steps: (1) extracting top-ranked subcomponents from training component which best-match the input component; (2) calculating weighting factors from these best-matched subcomponents; (3) collecting the proceeding optimal subcomponent and fusing them with assigned weighting factors to form prediction. The prediction algorithm was examined for several patients, and its performance is assessed based on the correlation and root mean square error (RMSE) between prediction and known output.

Results: Respiratory motion data was simulated for 30 cases and 555 patients and phantoms using the RPM system. Simulations were used to optimize prediction algorithm parameters. The simulation cases were used to determine optimal filters for smoothing and number of top-ranked subcomponents to determine optimal subcomponents for prediction. Summed difference results in a value of 0.4770 for the 15 Point Savitzky-Golay filter.

After determining the proper filter for data preprocessing the number of required top-ranked subcomponents for each method was determine. Equal Weighting has a maximum average correlation, c=0.997 when using 1 Subcomponent, Relative Weighting has a maximum average correlation, c=0.997 when using 2 Subcomponents, Pattern Weighting has a maximum average correlation c=0.915 when using 1 subcomponent, Derivative Equal Weighting has a maximum average correlation c=0.976 when using 2 Subcomponents, and Derivative Relative Weighting has a maximum average correlation of c=0.976 when using 5 Subcomponents.

The correlation coefficient and RMSE of prediction versus analysis component distributions demonstrate an improvement during optimization for simulations. This is true for both the full and half cycle prediction. However, when moving to the clinical data the distribution of prediction data, both correlation coefficient and RMSE, there is not an improvement as the optimization occurs. Therefore, a comparison of the clinical data using the 5 Pt moving filter and arbitrarily chosen number of subcomponents was performed. In the clinical data, average correlation coefficient between prediction and analysis component 0.721+/-0.390, 0.727+/-0.383, 0.535+/-0.454, 0.725+/-0.397, and 0.725+/-0.398 for full respiratory cycle prediction and 0.789+/-0.398, 0.800+/-0.385, 0.426+/-0.562, 0.784+/-0.389, and 0.784+/-0.389 for half respiratory cycle prediction for equal weighting, relative weighting, pattern, derivative equal and derivative relative weighting methods, respectively. Additionally, the clinical data average RMSE between prediction and analysis component 0.196+/-0.174, 0.189+/-0.161, 0.302+/-0.162, 0.200+/-0.169, and 0.202+/-0.181 for full respiratory cycle prediction and 0.155+/-0.171, 0.149+/-0.138, 0.528+/-0.179, 0.174+/-0.150, and 0.173+/-0.149 for half respiratory cycle prediction for equal weighting, relative weighting, pattern, derivative equal and derivative relative weighting methods, respectively. The half cycle prediction displays higher accuracy over the full cycle prediction. Wilcoxon signed-rank test reveals statistically highly significant values (p<0.1%) for 4 out of 5 algorithms favoring the half cycle prediction (Equal, Relative, Derivative Equal, and Derivative Relative Weighting Methods). In this method, the relative weighting method has the most correlations coefficients with values greater than 0.9 and also yields the largest number of highest correlations over other prediction methods.

Conclusions: In conclusion, the number of subcomponents used for prediction may be better determined based on individual breathing pattern. The prediction accuracy using patient data is better using half cycle prediction over full cycle prediction for all algorithms for the majority of methods tested. Finally, relative weighting method performed better than other methods.

Purpose: To develop a radiomics-embedded vision transformer (RE-ViT) model by incorporating radiomics features into its architecture, seeking to improve the model's efficiency in medical image recognition towards enhanced breast ultrasound image diagnostic accuracy.Materials and Methods: Following the classic ViT design, the input image was first resampled into multiple 16×16 grid image patches. For each patch, 56-dimensional habitat radiomics features, including intensity-based, Gray Level Co-Occurrence Matrix (GLCOM)-based, and Gray Level Run-Length Matrix (GLRLM)-based features, were extracted. These features were designed to encode local-regional intensity and texture information comprehensively. The extracted features underwent a linear projection to a higher-dimensional space, integrating them with ViT’s standard image embedding process. This integration involved an element-wise addition of the radiomics embedding with ViT’s projection-based and positional embeddings. The resultant combined embeddings were then processed through a Transformer encoder and a Multilayer Perceptron (MLP) head block, adhering to the original ViT architecture. The proposed RE-ViT model was studied using a public BUSI breast ultrasound dataset of 399 patients with benign, malignant, and normal tissue classification. The comparison study includes: (1) RE-ViT versus classic ViT training from scratch, (2) pre-trained RE-ViT versus pre-trained ViT (based on ImageNet-21k), (3) RE-ViT versus VGG-16 CNN model. The model performance was evaluated based on accuracy, ROC AUC, sensitivity, and specificity with 10-fold Monte-Carlo cross validation. Result: The RE-ViT model significantly outperformed the classic ViT model, demonstrating superior overall performance with accuracy = 0.718±0.043, ROC AUC = 0.848±0.033, sensitivity = 0.718±0.059, and specificity = 0.859±0.048. In contrast, the classic ViT model achieved accuracy = 0.473±0.050, ROC AUC = 0.644±0.062, sensitivity = 0.473±0.101, and specificity = 0.737±0.065. Pre-trained versions of RE-ViT also showed enhanced performance (accuracy = 0.864±0.031, ROC AUC = 0.950±0.021, sensitivity = 0.864±0.074, specificity = 0.932±0.036) compared to pre-trained ViT (accuracy = 0.675±0.111, ROC AUC = 0.872±0.086, sensitivity = 0.675±0.129, specificity = 0.838±0.096). Additionally, RE-ViT surpassed VGG-16 CNN results (accuracy = 0.553±0.079, ROC AUC = 0.748±0.080, sensitivity = 0.553±0.112, specificity = 0.777±0.089). Conclusion: The proposed radiomics-embedded ViT was successfully developed for ultrasound-based breast tissue classification. Current results underscore the potential of our approach to advance other transformer-based medical image diagnosis tasks.

Purpose: The fleet of x-ray computed tomography systems used by large medical institutions is often comprised of scanners from various manufacturers. An inhomogeneous fleet of scanners could lead to inconsistent image quality due to the different features and technologies implemented by each manufacturer. Specifically, image noise could be highly variable across scanners from different manufacturers. To partly address this problem, we have performed two studies to characterize noise magnitude and texture on two scanners: one from GE Healthcare and one from Siemens Healthcare. The purpose of the first study was to evaluate how noise magnitude changes as a function of image quality indicators (e.g., "noise index" and "quality reference mAs") when automatic tube current modulation is used. The purpose of the second study was to compare and match reconstruction kernels from each vendor with respect to noise texture.

Methods: The first study was performed by imaging anthropomorphic phantoms on each scanner using a clinical range of scan settings and image quality indicator values. Noise magnitude was measured at various anatomical locations using an image subtraction technique. Noise was then modeled as a function of image quality indicators and other scan parameters that were found to significantly affect the noise-image quality indicator relationship.

The second study was performed by imaging the American College of Radiology CT accreditation phantom with a comparable acquisition protocol on each scanner. Images were reconstructed using filtered backprojection and a wide selection of reconstruction kernels. We then estimated the noise power spectrum (NPS) of each image set and performed a systematic kernel-by-kernel comparison of spectra using the peak frequency difference (PFD) and the root mean square error (RMSE) as metrics of similarity. Kernels that minimized the PFD and RMSE were paired.

Results: From the fist study, on the GE scanner, noise magnitude increased linearly with noise index. The slope of that line was affected by changing the anatomy of interest, kVp, reconstruction algorithm, and convolution kernel. The noise-noise index relationship was independent of phantom size, slice thickness, pitch, field of view, and beam width. On the Siemens scanner, noise magnitude decreased non-linearly with increasing quality reference effective mAs, slice thickness, and peak tube voltage. The noise-quality reference effective mAs relationship also depended on anatomy of interest, phantom size, age selection, and reconstruction algorithm but was independent of pitch, field of view, and detector configuration. From the second study, the RMSE between the NPS of GE and Siemens kernels varied from 0.02 to 0.74 mm. The GE kernels "Soft", "Standard", "Chest", and "Lung" closely matched the Siemens kernels "B35f", "B43f", "B46f", and "B80f" (RMSE<0.07 mm, PFD<0.02 mm-1). The GE "Bone", "Bone+", and "Edge" kernels all matched most closely to Siemens "B75f" kernel but with sizeable RMSE and PFD values up to 0.62 mm and 0.5 mm-1 respectively. These sizeable RMSE and PFD values corresponded to visually perceivable differences in the noise texture of the images.

Conclusions: From the first study, we established how noise changes with changing image quality indicators across a clinically relevant range of imaging parameters. This will allow us target equal noise levels across manufacturers. From the second study, we concluded that it is possible to use the NPS to quantitatively compare noise texture across CT systems. We found that many commonly used GE and Siemens kernels have similar texture. The degree to which similar texture across scanners could be achieved varies and is limited by the kernels available on each scanner. This result will aid in choosing appropriate corresponding kernels for each scanner when writing protocols. Taken together, the results from these two studies will allow us to write protocols that result in images with more consistent noise properties.

The main objective of this dissertation research is to develop measurement and data-analysis tools for improving the quantitative accuracy of radiotracer studies of small plants, e.g., grasses in their early growth stages and tree seedlings. Improved accuracy is needed due to the thin nature of plant organs, e.g., leafs and stem. In addition, the methods developed in this thesis are applied to study the plant-environment interface of barley. Some of the approaches explored have potential to increase the statistical accuracy of counting data using PET imaging techniques. Improving the statistical precision of radionuclide tracking data will add to the analysis options. Another important goal is to measure the absolute photosynthetic rate. The standard approach in plant radiotracer experiments is to perform measurements of the relative distribution of radioactivity in various parts of the plant being studied. A limitation of this approach is that it does not take into account differences in the amount of radioisotope assimilated that are available for transport and allocation to the various sinks, that is, absolute CO2 uptake and photosynthetic rates are important factors in understanding the holistic physiological responses of plants to external conditions. For example, monitoring the movement of carbon-11 (11C) tagged carbohydrates in a plant requires an estimate of the average photosynthetic rate to determine the actual quantity of carbohydrates in each plant region (e.g. leaf, shoot, and root).

Radiotracing provides a method for real-time measurements of substance absorption, allocation and metabolic consumption and production in living organisms. Application of radioactive labelling in plants enables measurements associated with core physiological processes, e.g., photosynthesis, water uptake and nitrogen absorption and utilization. Plant uptake of radiotracers allows for tracking spatial and temporal distribution of substances, which enables studies of the plant-environment interface and the mechanisms involved in the allocation of resources (e.g., sugars, nutrients, and water). As such, these techniques are increasingly becoming an important tool for investigating the processes involved in the physiological responses of plants to changes in their local environmental conditions.

This dissertation has two major components: (1) development of experiment techniques for absolute photosynthetic rate measurements in plants using radio-isotope labeling, and (2) application of radioisotope tracing techniques to study the plant-environment interface in barley. The first component is covered in chapters one through three. The second component is presented in chapter four. An introduction into radio-tracing techniques is provided in chapter one. Chapter two describes radio-isotope production, radio-labelled compound preparation and delivery of labels to plant measurements. Chapter three outlines methods that can be employed to measure the absolute photosynthetic rate (µmol/m2/s) for a closed-loop system with [CO2] monitoring capabilities. Chapter four describes the background and results of our study on changing environmental conditions on a model system, barley seedlings. Chapter 5 will introduce the use of Monte-Carlo modeling for scaling the collected data to adjust the detected coincidence counts for losses due to positron escape from plant tissue. Chapter 6 describes the development of a novel imaging technique using direct positron detection that takes advantage of the high fraction of positrons escaping thin plant tissue.

In this dissertation, we have performed the most extensive measurements of carbohydrate allocation and translocation in a plant species using radio-isotope tracing techniques. A major practical limitation of studies based on radio-isotope labeling is the number of samples that can be measured in a single project. Our study on barley (Hordeum distichum) includes measurements on more than 30 plants. The short-lived radionuclide, 11C, was used to determine the real-time response to metabolite transport in barley. Sugars are photosynthesized and tagged with a positron-emitting radioisotope by flowing carbon dioxide (11CO2) tagged air over an active leaf. Data analysis of measurements taken in this dissertation indicates that the fraction of carbohydrates allocated to below ground sinks decreased, by 31% ± 9% in ambient [CO2] and by 37% ± 14% in elevated [CO2], when the nutrient conditions were rapidly changed from high to low nutrient.

In external beam radiation therapy (EBRT) for cancer patients, it is highly desirable to completely eradicate the cancerous cells for the purpose of improving the patient’s quality of life and increasing the patient’s likelihood of survival. However, there can be significant side effects when large regions of healthy cells are irradiated during EBRT, particularly for organs-at-risk (OARs). Due to the juxtaposition of the cancerous and non-cancerous tissue, trade-offs need to be made between target coverage and OAR sparing during treatment planning. For this reason, the treatment planning process can be posed as a multi-criterial optimization (MCO) problem, which has previously been studied extensively with several exact solutions existing specifically for radiation therapy. Typical MCO implementations for EBRT involve creating, optimizing, and calculating many treatment plans to infer the set of feasible best radiation doses, or the Pareto surface. However, each optimization and calculation can take 10-30 minutes per plan. As a result, generating enough plans to attain an accurate representation of the Pareto surface can be very time-consuming, particularly in higher-dimensions with many possible trade-offs.

The purpose of this study is to streamline the MCO workflow by using a machine-learning model to quickly predict the Pareto surface plan doses, rather than exactly computing them. The primary focus of this study focuses on the development and analysis of the dose prediction model. The secondary focus of this study is to develop new metrics for analyzing the similarity between different Pareto surface interpolations. The tertiary focus of this study is to investigate the feasibility of deliberately irradiating the epidural space in spine stereotactic radiosurgery (SRS), as well as estimate its potential effect on preventing tumor recurrence.

For the primary focus of this study, the model’s architecture proceeds as follows. The model begins by creating an initial dose distribution via an inverse fit of inter-slice and intra-slice PTV distance maps on a voxel-wise basis. The model proceeds by extracting three sets of transverse patches from all structure maps and the initialized dose map at each voxel. The model then uses the patch vectors as inputs for a neural network which updates and refines the dose initialization to achieve a final dose prediction. The primary motivation behind our model is to use our understanding of the general shape of dose distributions to remove much of the nonlinearity of the dose prediction problem, decreasing the difficulty of subsequent network predictions. Our model is able to take the optimization priorities into account during dose prediction and infer feasible dose distributions across a range of optimization priority combinations, allowing for indirect Pareto surface inference.

The model’s performance was analyzed on conventional prostate volumetric modulated arc therapy (VMAT), pancreas stereotactic body radiation therapy (SBRT), and spine stereotactic radiosurgery (SRS) with epidural space irradiation. For each of these treatment paradigms, the Pareto surfaces of many patients were thoroughly sampled to train and test the model. On all of these cases, our model achieved good performance in terms of speed and accuracy. Overfitting was shown to be minimal in all cases, and dose distribution slices and dose-volume histograms (DVHs) were shown for comparison, confirming the proficiency of our model. This model is relatively fast (0.05-0.20 seconds per plan), and it is capable of sampling the entire Pareto surface much faster than commercial dose optimization and calculation engines.

While these results were generally promising, the model achieved lower error on the prostate VMAT treatment plans compared to the pancreas SBRT and spine SRS treatment plans. This is likely due to the existence of heavier beam streaks in the stereotactic treatment plans which are generated by a sharper control of the delivered dose distribution. However, the Pareto surface errors were similar across all three cases, so these dose distribution errors did not propagate to the Pareto objective space.

The secondary focus of this study is the development and analysis of Pareto surface similarity metrics. The dose prediction model can be used to rapidly estimate many Pareto-optimal plans for quick Pareto surface inference. This could allow for a potentially significant increase in the speed at which Pareto surfaces are inferenced to provide treatment planning assistance and acceleration. However, previous investigations into Pareto surface analysis typically do not compare a ground truth Pareto surface with a Pareto surface prediction. Therefore, there is a need to develop a Pareto surface metric in order to evaluate the ability of the model to generate accurate Pareto surfaces in addition to accurate dose distributions.

To address these needs, we developed four Pareto surface similarity metrics, emphasizing the ability to represent distances between the interpolations rather than the sampled points. The most straightforward metric is the root-mean-square error (RMSE) evaluated between matched, sampled points on the Pareto surfaces, augmented by intra-simplex upsampling of the barycentric dimensions of each simplex. The second metric is the Hausdorff distance, which evaluates the maximum closest distance between the sets of sampled points. The third metric is the average projected distance (APD), which evaluates the displacements between the sampled points and evaluates their projections along the mean displacement. The fourth metric is the average nearest-point distance (ANPD), which numerically integrates point-to-simplex distances over the upsampled simplices of the Pareto surfaces. These metrics are compared by their convergence rates as a function of intra-simplex upsampling, the calculation times required to achieve convergence, and their qualitative meaningfulness in representing the underlying interpolated surfaces. For testing, several simplex pairs were constructed abstractly, and Pareto surfaces were constructed using inverse optimization and our dose prediction model applied to conventional prostate VMAT, pancreas SBRT, and spine SRS with epidural irradiation.

For the abstract simplex pairs, convergence within 1% was typically achieved at approximately 50 and 100 samples per barycentric dimension for the ANPD and the RMSE, respectively. The RMSE and the ANPD required approximately 50 milliseconds and 3 seconds to calculate to these sampling rates, respectively, while the APD and HD required much less than 1 millisecond. Additionally, the APD values closely resembled the ANPD limits, while the RMSE limits and HD tended to be more different. The ANPD is likely more meaningful than the RMSE and APD, as the ANPD’s point-to-simplex distance functions more closely represent the dissimilarity between the underlying interpolated surfaces rather than the sampling points on the surfaces. However, in situations requiring high-speed evaluations, the APD may be more desirable due to its speed, lack of subjective specification of intra-simplex upsampling rates, and similarity to the ANPD limits.

The tertiary focus of this study is the analysis of the feasibility of epidural space irradiation in spine SRS. The epidural space is a frequent site of cancer recurrence after spine SRS. This may be due to microscopic disease in the epidural space which is under-dosed to obey strict spinal cord dose constraints. We hypothesized that the epidural space could be purposefully irradiated to prescription dose levels, potentially reducing the risk of recurrence in the epidural space without increasing toxicity. To address this, we sought to analyze the feasibility of irradiating the epidural space in spine SRS. Analyzing the data associated with this study is synergistic to our MCO acceleration study, since the range of trade-offs between epidural space irradiation and spinal cord sparing represents an MCO problem which our dose prediction model may quickly solve.

Spine SRS clinical treatment plans with associated spinal PTV (PTVspine) and spinal cord contours, and prior delivered dose distributions were identified retrospectively. An epidural space PTV (PTVepidural) was contoured to avoid the spinal cord and focus on regions near the PTVspine. Clinical plan constraints included PTVspine constraints (D95% = 1800 cGy, D5% < 1950 cGy) and spinal cord constraints (Dmax < 1300 cGy, D10% < 1000 cGy). Prior clinical plan doses were mapped onto the new PTVepidural contour for analysis. Plans were copied and revised to additionally target the PTVepidural, optimizing PTVepidural D95% after meeting clinical plan constraints. Tumor control probabilities (TCPs) were estimated for the PTVepidural using a radiobiological linear-quadratic model of cell survival for both clinical and revised plans. Clinical and revised plans were compared according to their PTVepidural DVH distributions, D95% distributions, and TCPs.

Seventeen SSRS plans were identified and included in this study. Revised plan DVHs demonstrated higher doses to the epidural low-dose regions, with D95% improving from 10.96 Gy ± 1.76 Gy to 16.84 Gy ± 0.87 Gy (p < 10-5). Our TCP modeling set the clinical plan TCP average to 85%, while revised plan TCPs were all greater than 99.99%. Therefore, irradiating the epidural space in spine SRS is likely feasible, and purposefully targeting the epidural space in SSRS should increase control in the epidural space without significantly increasing the risk of spinal cord toxicity.

There always exists a demand to accelerate the time-consuming MRI acquisition process. Many methods have been proposed to achieve this goal, including deep learning method which appears to be a robust tool compared to conventional methods. While many works have been done to evaluate the performance of neural networks on standard anatomical MR images, few attentions have been paid to accelerating other less conventional MR image acquisitions. This work aims to evaluate the feasibility of neural networks on accelerating Brain DTI and Gynecological Brachytherapy MRI. Three neural networks including U-net, Cascade-net and PD-net were evaluated. Brain DTI data was acquired from public database RIDER NEURO MRI while cervix gynecological MRI data was acquired from Duke University Hospital clinic data. A 25% Cartesian undersampling strategy was applied to all the training and test data. Diffusion weighted images and quantitative functional maps in Brain DTI, T1-spgr and T2 images in GYN studies were reconstructed. The performance of the neural networks was evaluated by quantitatively calculating the similarity between the reconstructed images and the reference images, using the metric Total Relative Error (TRE). Results showed that with the architectures and parameters set in this work, all three neural networks could accelerate Brain DTI and GYN T2 MR imaging. Generally, PD-net slightly outperformed Cascade-net, and they both outperformed U-net with respect to image reconstruction performance. While this was also true for reconstruction of quantitative functional diffusion weighted maps and GYN T1-spgr images, the overall performance of the three neural networks on these two tasks needed further improvement. To be concluded, PD-net is very promising on accelerating T2-weighted-based MR imaging. Future work can focus on adjusting the parameters and architectures of the neural networks to improve the performance on accelerating GYN T1-spgr MR imaging and adopting more robust undersampling strategy such as radial undersampling strategy to further improve the overall acceleration performance.

Purpose: (1) To benchmark the accuracy of effective dose equivalent (EDE) of the single- and double-badge methods (NRC 2002-06) using the commercially available radiation monitors in clinical settings, (2) to study the transmission properties of various shielding materials, (3) to evaluate the accuracy of film badge readings compared to a calibrated ion chamber, (4) to benchmark the accuracy of effective dose (ED) of the single- and double-badge methods (NRC 2002-06) using the MOSFET method, and (5) to investigate the organ dose and image quality in a thoracic MDCT scan under the following conditions: (a) tube current modulation (TCM) without a Bismuth shield, (b) TCM with a Bismuth shield, and (c) manually reduced tube current (RTC) with no Bismuth shield.

Methods and Materials: (Project 1): Radiation workers in interventional radiology and cardiac catheterization laboratory were provided with two monitors and asked to place one at the collar and the other underneath the lead apron. Two commercial radiation monitor vendors were used for the study; both vendors were accredited by the National Voluntary Laboratory Accreditation Program (NVLAP). Effective dose equivalent (EDE) was computed by single-badge and double-badge methods based on the NRC Publication 2002-06. Data were plotted EDE1 (single badge) vs. collar reading and EDE2 (double badge) vs. collar reading. Data on EDE2 vs. collar reading were fitted by linear regression and a new equation for the EDE2collar was derived for routine clinical EDE estimation.

(Project 2): The transmission properties of lead aprons and a thyroid shield were measured using a 6-cc ion chamber and electrometer. These measurements were taken on a GE VCT (64 slice) scanner at 80, 100, and 120 kVp. The different types of lead aprons studied included lead free, lightweight lead, and fully leaded.

(Project 3): The accuracy of film badges was evaluated by comparing the reported deep dose equivalent of the film badge readings to the ion chamber readings measured during the same exposure. The measurements were made on a Philips Standard Radiography unit (Duke North, Room H1) at 80, 100, and 120 kVp. Two badges were exposed with the ion chamber per energy.

(Project 4): An adult male anthropomorphic phantom was loaded with 20 diagnostic MOSFET detectors and scanned without lead aprons using a whole body computed tomography (CT) protocol. All measurements were taken on a GE VCT (64 slice) scanner at 80, 100, 120 kVp. Two commercial film badges were placed on the phantom at the collar location and waist location. Individual organ doses in the phantom were corrected for lead apron attenuation factor and ED was computed using ICRP 103 tissue weighting factors. The single badge conversion coefficient (CC) was determined for each energy by taking the ratio of the ED to collar badge reading. The reported deep dose equivalent for the collar badge was plotted against the MOSFET effective dose and a new equation for EDcollar was derived.

(Project 5): Organ dose was measured with MOSFETs using an adult female anthropomorphic phantom; the phantom was scanned with pulmonary embolus protocol. All measurements were performed with a 64-slice scanner at 120 kVp. The reference exposure and reduced exposure (with 4-ply Bi shield) was measured with an ion chamber located at the level of the breast. The tube current was reduced by normalizing the reference tube current to the ratio of the reduced exposure to the reference exposure. Image quality was measured using a high contrast insert placed in the lung. Regions of interest (ROIs) were drawn in the breast, lung, and heart to measure HU change and noise. ROIs were drawn in the lung and high contrast insert to measure signal-to-noise ratio (SNR) and percent contrast (%Contrast).

Results: (Project 1): From the data, it can be seen that EDE1 read about a factor of six greater than EDE2. The new equation for EDE2collar yielded a slope of 0.06992, a y-intercept of -1.682, and a r2 value of 0.9081.

(Project 2): The transmission for the fully leaded, lightweight lead aprons, and lead free apron were 3.19%, 3.71%, and 7.06% at 80 kVp; 6.58%, 8.07%, and 13.04% at 100 kVp; 7.61%, 12.05%, and 17.84% at 120 kVp, respectively. The attenuation for the thyroid shield was 3.02%, 6.35%, and 7.74% at 80, 100, and 120 kVp, respectively.

(Project 3): The average badge reading was 3.49 ± 1.01% mSv at 80 kVp; 4.80 ± 7.37% mSv at 100 kVp; 4.90 ± 14.1% mSv at 120 kVp. The dose to soft tissue measured by the ion chamber was 4.53 mSv at 80 kVp; 5.71 mSv at 100 kVp; 6.35 mSv at 120 kVp. The film badge reading differed from the ion chamber measurement by -22.8%, -15.9%, and -22.9% at 80, 100, and 120 kVp, respectively.

(Project 4): The ED and % difference between the single-badge method (NRC 2002-06) and the MOSFET method were as follows: 11.65 mSv vs. 0.50 mSv (2331%) for 80 kVp; 27.85 mSv vs. 2.14mSv (1301%) for 100 kVp; 38.59 mSv vs. 4.98 mSv (775%) for 120 kVp, respectively. The ED and % difference between the double-badge method (NRC 2002-06) and the MOSFET method were as follows: 4.07 mSv vs. 0.50 mSv (808%) for 80 kVp; 16.9 mSv vs. 2.14 mSv (791%) for 100 kVp; 25.4 mSv vs. 4.98 mSv (510%) for 120 kVp, respectively. The single badge conversion factors were 0.01 ± 14.8% (80 kVp), 0.02 ± 9.5% (100 kVp), and 0.04 ± 15.7% (120 kVp). The plot of collar badge reading vs. MOSFET effective dose yielded an equation with a slope of 0.0483, a y-intercept of -1.6517, and a R2 value of 0.92929.

(Project 5): Organ doses (mGy) for the three scans (TCM, TCM with Bi, and RTC with no Bi) were 45.8, 27.1, and 27.8 to the breast; 51.6, 47.0, and 29.1 to the lung; and 42.1, 35.0 and 24.9 to the heart, respectively. HU increase was greatest in the TCM with Bi scan. The SNRs were 77.1, 63.7, and 59.2 and the %Contrast values were 369.5, 347.1, and 362.7 with TCM, TCM with Bi, and RTC, respectively.

Conclusions: (Project 1): A new EDE estimation method has been developed based on the results of two-badge system. The method would enable us to compute new EDE values knowing only the collar badge reading. Since EDE2 reads a factor of six less than EDE1, this provides a realistic advantage in regulatory compliance for interventional and cardiac catheterization personnel. Further, new EDE conversion coefficients should be developed for better assessment of EDE.

(Project 2): The fully leaded shielding materials had the lowest percent transmission. It should be noted that radiation workers are generally exposed to only scattered radiation of lower energy. Although this study did not measure attenuation properties at lower energies, it is expected that the percentage of attenuation will only increase with lower energies.

(Project 3): The reported deep dose equivalent (DDE) underestimated the dose to soft tissue compared to the calibrated ion chamber readings. This may be due to the fact that DDE is the dose equivalent at a depth of 10 mm.

(Project 4): Current regulatory ED conversion coefficient (CC) with single collar badge is 0.3; for double-badge system, they are 0.04 and 1.5 for the collar and under the apron respectively. Based on our findings we recommend the current collar CC be dropped due to the overestimation of ED. Since occupational workers are exposed mainly to scattered x-rays of lower energy, a collar CC of 0.01 (80 kVp data) may be a more viable option. The double badge system seems to provide a better coefficient for the collar as 0.04; however, exposure readings under the apron are usually negligible to zero with lead aprons.

(Project 5): For thoracic CT using RTC will result in similar global reduction in organ dose; the use of Bismuth with TCM will lead to an overall decrease in organ dose and more marked dose reduction for the breast. There was a significant difference in SNR (p = 0.0003) and %Contrast (p < 0.0001) in the TCM with Bismuth scan compared to the reference scan (TCM). The RTC scan also demonstrated a significant decrease in SNR and %Contrast with p < 0.0001 for both. While the TCM scan demonstrated superior image quality, the trade-off is in the increased dose to the breast.

With the advent of new, more efficient, rotational therapy techniques such as volumetric modulated arc therapy (VMAT), radiation therapy treatment precision requires evolving quality assurance. Two dimensional (2D) detector arrays have shown angular dependence that must be compensated for by the creation of angular correction factor tables. Currently available correction factor tables have several underlying assumptions that leave room for improvement: first, these correction factors assume that the response of all ion chambers is identical for each angle; second, that the ion chamber array response from gantry angles 0°-180° are equivalent to the response from 180°-360° and, third, that the response is independent of the direction of rotation.

Measurements were acquired using a 2D ion chamber array (MatriXX®, IBA Dosimetry) for static open fields delivered every 5° around the MatriXX while dose was calculated using Eclipse v8.6 (analytic anisotropic algorithm, Varian Medical Systems). Customized correction factors were created by dividing the calculated dose by the measured dose for each ion chamber. Two measurement positions were used in the creation of the custom correction factors: a coronal position in which the couch was included, and two sagittal orientations in which the couch was not included.

The correction factors were verified using open field arcs and VMAT patient plans, where measured doses were compared to calculated doses using gamma analysis (3%, 3 mm). Narrow fields were also delivered clockwise and counterclockwise in order to investigate the effect of the internal structure of the ion chamber array.

The angular response of the individual ion chambers appears to vary significantly (1 &sigma &le 4.6%). The response from 0°-180° vs. 180°-360° is significantly different (paired t-test yields p<0.0001). Custom correction factors do enhance the agreement between measured and calculated doses for open field arcs and VMAT patient plans compared to the default correction factors. The direction of rotation appears to affect the dose to the penumbra region of narrow fields, which could affect VMAT patient specific quality assurance.

The custom correction factor tables, using measurements for individual ion chambers over a full 0°-360° range, allows for improved accuracy in measurements by the 2D ion chamber array. However, even the corrected measurements still showed discrepancies with the calculated doses for VMAT plans.

Computerized breast phantoms have been popular for low-cost alternatives to collecting clinical data by combing them with highly realistic simulation tools. Image segmentation of three-dimensional breast computed tomography (bCT) data is one method to create such phantoms, but requires multiple image processing steps to accurately classify the tissues within the breast. One key step in our segmentation routine is the use of a bilateral filter to smooth homogeneous regions, preserve edges and thin structures, and reduce the sensitivity of the voxel classification to noise corruption. In previous work, the well-known process of bilateral filtering was completed on the entire bCT volume with the primary goal of reducing the noise in the entire volume. In order to improve on this method, knowledge of the varying bCT noise in each slice was used to adaptively increase or decrease the filtering effect as a function of distance to the chest wall. Not only does this adaptive bilateral filter yield thinner structures in the segmentation result but is adaptive on a case-by-case basis, allowing for easy implementation with future virtual phantom generations.

As complex radiotherapy techniques become more readily-practiced, comprehensive 3D dosimetry is a growing necessity for advanced quality assurance. However, clinical implementation has been impeded by a wide variety of factors, including the expense of dedicated optical dosimeter readout tools, high operational costs, and the overall difficulty of use. To address these issues, a novel dry-tank optical CT scanner was designed for PRESAGE 3D dosimeter readout, relying on 3D printed components and omitting costly parts from preceding optical scanners. This work details the design, prototyping, and basic commissioning of the Duke Integrated-lens Optical Scanner (DIOS).

The convex scanning geometry was designed in ScanSim, an in-house Monte Carlo optical ray-tracing simulation. ScanSim parameters were used to build a 3D rendering of a convex ‘solid tank’ for optical-CT, which is capable of collimating a point light source into telecentric geometry without significant quantities of refractive-index matched fluid. The model was 3D printed, processed, and converted into a negative mold via rubber casting to produce a transparent polyurethane scanning tank. The DIOS was assembled with the solid tank, a 3W red LED light source, a computer-controlled rotation stage, and a 12-bit CCD camera. Initial optical phantom studies show negligible spatial inaccuracies in 2D projection images and 3D tomographic reconstructions. A PRESAGE 3D dose measurement for a 4-field box treatment plan from Eclipse shows 95% of voxels passing gamma analysis at 3%/3mm criteria. Gamma analysis between tomographic images of the same dosimeter in the DIOS and DLOS systems show 93.1% agreement at 5%/1mm criteria. From this initial study, the DIOS has demonstrated promise as an economically-viable optical-CT scanner. However, further improvements will be necessary to fully develop this system into an accurate and reliable tool for advanced QA.

Pre-clinical animal studies are used as a conventional means of translational research, as a midpoint between in-vitro cell studies and clinical implementation. However, modern small animal radiotherapy platforms are primitive in comparison with conventional linear accelerators. This work also investigates a series of 3D printed tools to expand the treatment capabilities of the X-RAD 225Cx orthovoltage irradiator, and applies them to a feasibility study of hippocampal avoidance in rodent whole-brain radiotherapy.

As an alternative material to lead, a novel 3D-printable tungsten-composite ABS plastic, GMASS, was tested to create precisely-shaped blocks. Film studies show virtually all primary radiation at 225 kVp can be attenuated by GMASS blocks of 0.5cm thickness. A state-of-the-art software, BlockGen, was used to create custom hippocampus-shaped blocks from medical image data, for any possible axial treatment field arrangement. A custom 3D printed bite block was developed to immobilize and position a supine rat for optimal hippocampal conformity. An immobilized rat CT with digitally-inserted blocks was imported into the SmART-Plan Monte-Carlo simulation software to determine the optimal beam arrangement. Protocols with 4 and 7 equally-spaced fields were considered as viable treatment options, featuring improved hippocampal conformity and whole-brain coverage when compared to prior lateral-opposed protocols. Custom rodent-morphic PRESAGE dosimeters were developed to accurately reflect these treatment scenarios, and a 3D dosimetry study was performed to confirm the SmART-Plan simulations. Measured doses indicate significant hippocampal sparing and moderate whole-brain coverage.