Radiation Medicine Commons^™

Evaluation Of Variables Predicting Pulmonary Function Test (Pft) Changes For Lung Cancer Patients Treated On A Prospective 4dct-Ventilation Functional Avoidance Clinical Trial, Nader Ghassemi, Richard Castillo, Edward Castillo, Bernard L. Jones, Moyed Miften, Brian Kavanaugh, Maria Werner Werner-Wasik, Ryan Miller, Julie Barta, Inga Grills, Thomas Guerrero

Conference Presentation Abstracts

Purpose: Functional avoidance radiotherapy uses functional imaging to reduce pulmonary toxicity by designing radiotherapy plans that reduce doses to the functional lung. A novel form of lung functional imaging uses 4DCT imaging to calculate 4DCT-based lung ventilation (4DCT-ventilation) maps. A phase-II, multi-center, prospective study of 4DCT-ventilation functional avoidance was completed. Pre and post-treatment pulmonary function tests (PFTs) were acquired in order to quantitatively assess pulmonary function change. The purpose of this study is to evaluate which factors predict for PFT changes for patients treated with 4DCT-ventilation functional avoidance radiotherapy. Methods: 56 patients with locally advanced lung cancer receiving radiotherapy were …

Real-Time Tracking Diaphragm Motion On During-Treatment Kv Cone Beam Projection Images Using Resnet50, J Liang, Q Liu, E Porter, D Yan

Conference Presentation Abstracts

Purpose: Diaphragm manifested on Cone Beam projection image has been used as surrogate to monitor lung tumor position. We propose a deep learning method to detect diaphragm apex position of the ipsilateral lung in real-time on KV projections acquired during Lung SBRT delivery. Methods: 7 hypo-lung SBRT patients with right lower lobe tumor were selected in this retrospective study. The imaging projection angle ranges from about -270 to 112 degrees. The right lung diaphragm apex was manually labeled as ground-truth on KV projections acquired during-treatment sessions with 899±167 projections per session. The backbone of our machine learning model is 50 …

An Evolutionary Algorithm Of The Spot Sparsity Optimization In Proton Arc Therapy, Lewei Zhao, Gang Liu, Xiaoqiang Li, Xuanfeng Ding

Conference Presentation Abstracts

Purpose: Spot-scanning Proton Arc (SPArc) plan normally contains thousands of spot numbers in which is the delivery time is proportional to the number of spots. It is critical to find an optimal SPArc with a fast delivery speed while maintaining a good plan quality. Thus, we developed a novel evolutionary algorithm to directly search the optimal spot sparsity solution in the balance of plan quality and beam delivery time (BDT). Methods: The planning platform included a plan quality objective, a selector and a generator. A selector was designed to filter or add the spot according to the expected spot number, …

Physics-Informed Machine Learning For Estimating Pulmonary Perfusion From Non-Contract 4dct, Y Liu, A Nowacki, R Castillo, Y Vinogradskiy, G Nair, C Stevens, E Castillo

Conference Presentation Abstracts

Purpose: Novel methods have been developed for functional avoidance that proposes to use 4DCT to derive lung ventilation and perfusion images. Previous methods for quantifying lung perfusion on non-contrast 4DCT rely either on HU-based physical models or black-box deep learning models. While deep learning typically achieves higher accuracies in image processing tasks, physical models provide a rationale for model predictions. The purpose of this study is to introduce a biophysics-informed machine learning method for estimating pulmonary perfusion from noncontrast 4DCT. Our approach is designed to combine the predictive power of neural networks with the interpretability of physical modeling, with the …

Per-Fraction Deviation In Logfile Parameters Reported In A Single Vmat Arc, M Snyder, D To, D Drake, J Liang, D Lack

Conference Presentation Abstracts

Purpose: Measurement-based QA identifies large deviations between planned and delivered dose, but subtle—yet clinically unacceptable—deviations can pass through the conventional QA process unnoticed. Incorporating logfilebased information into pre-treatment QA should result in more robust QA overall, provided such information is stable across the entire treatment course. We present here a case-study of logfile stability to investigate the potential of logfiles as useful adjuncts in the QA process. Methods: We analyzed logfiles from six fractions of a VMAT plan delivered on an Elekta Infinity linac with an Agility MLC. To simplify the analysis, we considered a single 52 control point VMAT …

A Novel Simultaneous Plan Quality And Beam Delivery Time Sparc Optimization Platform Using Primal Dual Active Set With Continuation (Pdasc), Lewei Zhao, J You, Gang Liu, X Lu, Xuanfeng Ding

Conference Presentation Abstracts

Purpose: Proton arc is a new treatment modality that delivers proton beams while continuously rotating the gantry. This study proposed a regularized l0-minimization primal dual active set with continuation (PDASC) algorithm for proton arc spot sparsity optimization to simultaneously optimize the plan quality and the beam delivery time (BDT). Methods: Based on the previously published beam delivery sequence model of IBA ProteusONE®, proton treatment delivery time is actually dominated by spot switching time (SSWT). SSWT is approximately linearly dependent on spot numbers. So we used a non-convex l0-norm to control the sparsity level of the regularized solution. The clinical objective …

Clinical Experience Of Exactrac To Guide Six Degree Of Freedom Fiducial Marker Based Patient Positioning For Hypo-Prostate Fractionation, R Sandhu, C Knill, B Loughery, L Lin, Z Seymour

Conference Presentation Abstracts

Purpose: ExacTrac is clinically used for inter- and intra- fraction shift corrections for hypofractionated prostate patients by registering to implanted fiducial markers. ExacTrac corrections for thirty-three prostate patients were collated and analyzed. Methods: ExacTrac shifts were exported for prostate patients (140 total fractions) who were treated with two 180° partial arcs with beam-on time less than 60seconds/arc. Prior to each treatment fraction, a cone beam computed tomography (CBCT) image was acquired to verify bladder and rectal filling against simulation. Following CBCT, ExacTrac was used for initial prostate localization and shifts were applied using a 6 degree-of-freedom (6D) couch. Following initial …

A Team Learning Initiative In Undergraduate Medical Physics, B Loughery, J Mcbrady, M Joiner

Conference Presentation Abstracts

Purpose: To improve comprehension in traditionally difficult subjects for a senior-level undergraduate medical physics course, we designed and implemented a Team Learning technique. This work evaluates the translation of Team Learning to medical physics.  Methods: At the beginning of the term, students were sorted into 3 even teams (students/year = 14.5).  The class met twice weekly: a traditional lecture was taught first, and each team was assigned a specific 1/3 of the lecture material in which to specialize.  Each team was assigned to create a PowerPoint presentation on their material using outside references and without copying the original figures or …

Quantifying Copd Disease Severity With Ct-Derived Perfusion Imaging, A Nowacki, G Nair, Y Liu, C J. Galban, C Stevens, E Castillo

Conference Presentation Abstracts

Purpose: CT-derived perfusion (CTP) is a novel image processing modality that quantifies pulmonary perfusion from non-contrast inhale/exhale CT image pairs. While existing imaging markers for COPD primarily depend on morphological features or HU thresholding, CTP provides a quantitative marker of blood flow and has been shown to identify disease progression prior to the appearance structural abnormalities. In this study, we assess the potential utility of CTP as a disease severity marker for COPD. Methods: CTP was computed from high-resolution inhale-exhale CT scans for 785 patients with a Global Initiative for Obstructive Lung Disease (GOLD) score ranging from 0 to 4 …

Mri-Based Synthetic Ct Images Generated Using 3d Conditional Gan For Impt Treatment Planning, Shupeng Chen, An Qin, Rohan Deraniyagala, Xuanfeng Ding, Y Peng, Y Liu, C Zhao, X Deng

Conference Presentation Abstracts

Purpose: To develop an advanced deep convolutional neural network architecture to generate synthetic CT (SCT) images from MR images for intensity-modulated proton therapy (IMPT) treatment planning of head and neck cancer patients. Methods: T1-weighted MR images and paired CT (PCT) images were obtained from 189 nasopharyngeal cancer (NPC) patients under radiotherapy immobilization. Deformable image registration was performed between MR and PCT images for each patient to create an MR-CT image pair. Thirteen pairs were randomly chosen as independent test sets and the remaining 176 pairs (14 for validation and 162 for training) were used to build two conditional generative adversarial …

Evaluation Of 4pi Arc Positioning For Cranial Vmat Srs In Elements, B Loughery, C Knill, R Sandhu, L Lin, Z Seymour

Conference Presentation Abstracts

Purpose: Brainlab’s Elements Cranial VMAT (CVMAT) v3.0 introduces a 4Pi optimization that includes automated gantry and couch positioning alongside collimator optimization. In contrast, v1.5 uses a fixed collimator angle and predefined templates for gantry and couch angles that are manually modified by the planner. This study evaluates the 4Pi generated plans for clinical acceptability in terms of plan quality and delivery accuracy on a VersaHD. Methods: We chose a recent sample of 9 clinical plans that represent our catalog of target locations, fractionation schemes, and target sizes. Cases were replanned in CVMAT v3.0 using 4Pi optimization. These plans were evaluated …

Does Lung Function Imaging Modality Have A Dosimetric Impact On Functional Avoidance Treatment Planning: Assessment Using Prospective Clinical Trial Data, J Belardo, S Zhang, E Castillo, R Castillo, C Rusthoven, B Jones, M Miften, T Guerrero, I Grills, Y Chen

Conference Presentation Abstracts

Purpose: Functional avoidance radiotherapy proposes to use functional imaging to design treatment plans that reduce doses to functional regions of lung. Prospective data has shown that reducing doses to functional lung results in reduced pulmonary toxicity. Various lung function imaging modalities have been proposed for functional avoidance; however, there have been no studies evaluating whether using different imaging modalities results in dosimetrically different functional avoidance plans. The purpose of this study was to quantify the differences in treatment plans as a function of lung function imaging modality. Methods: 25 lung cancer patients enrolled on a prospective trial underwent nuclear medicine …

Validation And Implementation Of A Dedicated Spine Stereotactic Radiosurgery Treatment Planning System, C Knill, R Sandhu, B Loughery, L Lin, Z Seymour, P Chinnaiyan, T Quinn, M Almahariq, R Deraniyagala

Conference Presentation Abstracts

Purpose: A dedicated spine stereotactic radiosurgery optimization algorithm was commissioned for 6FFF Versa HD deliveries. Plan comparisons with the existing clinical optimization algorithm along with the dosimetric validation of the deliveries were investigated. Methods: Nine stereotactic spine patients, previously treated with Pinnacle generated plans, were re-optimized using Brainlab’s SRS spine planning Element. Three additional Elements plans were created for each patient: 1) initial optimization using a pencil beam calculation algorithm (PBC), 2) Monte Carlo (MC) recalculation of the PBC plan, and 3) re-optimization of the PBC plan using Monte Carlo. Patient dose from PBC plans was compared to MC recalculated …

Comparison Of Aperture Complexity Between Vmat Plans Using Conventionally Flattened (Ff) And Flattening-Filter-Free (Fff) Beams, Q Liu, J Liang

Conference Presentation Abstracts

Purpose: VMAT plans with high amount of complex apertures could lead to less confidence in dose calculation accuracy due to machine modelling and dose algorithm limitations. It was hypothesized that VMAT plans generated with FFF beams would result in more complex aperture shapes due to the non-flattened beam profile. The purpose of this study was to test this hypothesis by assessing various plan complexity metrics. Methods: Target and organ-at-risk structures were created on an anthropomorphic phantom in various sites: lung, prostate, spine, pancreas, breast, rectum and HN. Hypo-fractioned VMAT plans were generated using Elekta Agility MLC in Pinnacle TPS. Both …

Financial Toxicity Tumor Board: A Multi-Disciplinary Team Activity Required In Low And Middle-Income Countries (Lmic), Ahmed Nadeem Abbasi, Sohail Rasool, Laraib Khan

Department of Radiation Oncology

No abstract provided.

In-Phantom Film Measurements Of Two Treatment Planning Systems For Single-Fraction Spine Sbrt, Michael J. Taylor

LSU Master's Theses

Purpose: Treatment planning accuracy for spine stereotactic body radiation therapy (SBRT) varies depending on the dose calculation algorithm utilized in the treatment planning system (TPS). This project compared the end-to-end accuracy between spine SBRT plans calculated in a convolution-superposition based TPS (TPS_CS) and Monte Carlo based TPS (TPS_MC) with radiochromic film measurements. The hypothesis was that TPS_MC would calculate the dose gradient in the critical region between the vertebral body and the spinal cord more accurately than TPS_CS.

Methods: Single-fraction spine SBRT treatments following RTOG 0631 and local institutional guidelines were planned in …

Fluoroscopic Demonstration Of Thoracic Tumor Immobilization With High Frequency Percussive Ventilation, Cristian Solano, Ina M. Sala, Beverly Maurer, Ronald Levitin, Thomas M. Guerrero

Posters

INTRODUCTION
High frequency percussive ventilation (HFPV) is a novel immobilization technique that utilizes high frequency low tidal volume ventilation to produce endotracheal percussion. In a previous departmental study of chest wall motion immobilization, it was found that volunteers were able to tolerate HFPV for varying lengths of time – from a few to tens of minutes. By investigating a novel process to immobilize the chest wall, and thus thoracic tumors, it can allow for more localized radiation delivery and reduction of healthy tissue irradiation.

Development Of A Deep Neural Network For Synthesis Of Non-Contrast Cranial T1-Weighted Magnetic Resonance Imaging, Agueda M. Taylor, Evan Porter, Thomas Guerrero

Posters

INTRODUCTION
Although 122 out of 1000 people in the US have MRI’s done each year, there are over 4 million with contraindications that subsequently forgo the diagnostic benefits. Studies in recent years have implemented artificial intelligence (AI) algorithms such as deep neural networks (DNN) for production of synthetic medical imaging. The goals of this project are to develop a DNN, specifically a Generative Adversarial Network (GAN) that will predict synthetic Cranial T1 Weighted MRI from non-contrast CT, and to evaluate the model quality.

Functional Lung Segmentation From Computed Tomography Images Using Deep Learning, Duyen M. Quach, Evan Porter, Thomas M. Guerrero

Posters

INTRODUCTION
Functional avoidance treatment planning allows radiation oncologists to intentionally minimize radiation dose to higher-functioning lung areas while favoring radiation dose towards lower-functioning regions. This treatment planning method has been shown to reduce pulmonary toxicity for patients receiving radiation therapy. Functional lung information, traditionally ventilation or perfusion SPECT scans, is required to plan for functional avoidance.

Evaluating The Impact Of Software Distortion Correction On Target Doses In Cranial Stereotactic Radiosurgery, Sharjil Shamim, Zachary Seymour

Posters

INTRODUCTION
Brainlab’s Elements suite proposed a novel MR distortion correction method based on CT imaging. This involves registration of sub-volumes from MR imaging to generate single continuous deformation field maps, creating a corrected MR image set. No literature currently exists studying the impact of software correction on target dose for cranial stereotactic radiosurgery. In this study, we aim to evaluate the impact of Brainlab distortion correction on radiation target dose compared to original treatment plans without distortion correction.