A5014172721- Advanced Radiotherapy Techniques
- Medical Imaging Techniques and Applications
- Radiation Therapy and Dosimetry
- Radiomics and Machine Learning in Medical Imaging
- Lung Cancer Diagnosis and Treatment
- Radiation Dose and Imaging
- Advanced X-ray and CT Imaging
- Advances in Oncology and Radiotherapy
- Head and Neck Cancer Studies
- Prostate Cancer Diagnosis and Treatment
- Medical Imaging and Analysis
- Breast Cancer Treatment Studies
- Radiation Effects and Dosimetry
- Brain Metastases and Treatment
- Artificial Intelligence in Healthcare and Education
- Management of metastatic bone disease
- Radiopharmaceutical Chemistry and Applications
- Hepatocellular Carcinoma Treatment and Prognosis
- MRI in cancer diagnosis
- Advanced Neural Network Applications
- Medical Image Segmentation Techniques
- Digital Radiography and Breast Imaging
- Radiation Detection and Scintillator Technologies
- Renal cell carcinoma treatment
- Catalytic Processes in Materials Science
The University of Texas Southwestern Medical Center
2016-2025
Southwestern Medical Center
2016-2025
Brookhaven National Laboratory
2024
University of Pennsylvania
2022
Harold C. Simmons Comprehensive Cancer Center
2022
Fox Chase Cancer Center
2010-2020
Centre for Biomedical Engineering and Physics
2020
University of Maryland, Baltimore
2014-2016
Taichung Veterans General Hospital
2016
University of Maryland Medical Center
2016
The use of neural networks to directly predict three-dimensional dose distributions for automatic planning is becoming popular. However, the existing methods only patient anatomy as input and assume consistent beam configuration all patients in training database. purpose this work was develop a more general model that considers variable configurations addition achieve comprehensive with potentially easier clinical implementation, without need train specific models different settings.The...
Accurate segmentation of prostate and surrounding organs at risk is important for cancer radiotherapy treatment planning. We present a fully automated workflow male pelvic CT image using deep learning. The architecture consists 2D organ volume localization network followed by 3D volumetric prostate, bladder, rectum, femoral heads. used multi-channel U-Net with encoding arm modified aggregated residual networks, known as ResNeXt. models were trained tested on dataset comprising 136 patients....
A significant subset of patients with stage II/III non-small cell lung cancer (NSCLC) cannot receive standard concurrent chemoradiotherapy owing to the risk toxic effects outweighing potential benefits. Without chemotherapy, however, efficacy conventional radiotherapy is reduced.To determine whether hypofractionated image-guided (IGRT) would improve overall survival in NSCLC who could not and therefore were traditionally relegated receiving only conventionally fractionated (CFRT).This...
Renal cell carcinoma is refractory to conventional radiation therapy but responds higher doses per fraction. However, the dosimetric data and clinical factors affecting local control (LC) are largely unknown. We aimed evaluate safety efficacy of stereotactic ablative (SAbR) for extracranial renal metastases.We reviewed 175 metastatic lesions from 84 patients treated with SAbR between 2005 2015. LC toxicity after were assessed Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1...
The advent of cone beam computed tomography-based online adaptive radiation therapy (oART) has dramatically reduced the barriers adaptation. We present first prospective oART experience data in head and neck cancers (HNC).Patients with HNC receiving definitive standard fractionation (chemo)radiation who underwent at least 1 session were enrolled a registry study. frequency adaptations was discretion treating physician. Physicians given option delivering 2 plans during adaptation: original...
This paper investigates the clinical significance of real-time monitoring intrafractional prostate motion during external beam radiotherapy using a commercial 4D localization system. Intrafractional was tracked 8,660 treatment fractions for 236 patients. The following statistics were analyzed: 1) percentage in which shifted 2-7 mm certain duration; 2) proportion entire tracking time 2-7mm; and 3) each minute shift exceeded mm. ten patients exhibiting maximum intrafractional-motion patterns...
Recently, artificial intelligence technologies and algorithms have become a major focus for advancements in treatment planning radiation therapy. As these are starting to incorporated into the clinical workflow, concern from clinicians is not whether model accurate, but can express human operator when it does know if its answer correct. We propose use Monte Carlo Dropout (MCDO) bootstrap aggregation (bagging) technique on deep learning (DL) models produce uncertainty estimations therapy dose...
Abstract Purpose Varian Ethos utilizes novel intelligent‐optimization‐engine (IOE) designed to automate the planning. However, this introduced a black box approach plan optimization and challenge for planners improve quality. This study aims evaluate machine‐learning‐guided initial reference generation approaches head & neck (H&N) adaptive radiotherapy (ART). Methods Twenty previously treated patients on C‐arm/Ring‐mounted were retroactively re‐planned in planning system using fixed...
Ethos CBCT-based adaptive radiotherapy (ART) system can generate an online plan by re-optimizing the initial reference based on patient anatomy at treatment. The optimization process is fully automated without any room for human intervention. Due to change in anatomy, ART be significantly different from terms of parameters such as aperture shapes and number monitor units (MUs). In this study, we investigated feasibility using calculation-based specific QA plans conjunction with...
Abstract Background In radiation therapy (RT), accelerated partial breast irradiation (APBI) has emerged as an increasingly preferred treatment modality over conventional whole due to its targeted dose delivery and shorter course of treatment. APBI can be delivered through various modalities including Cobalt‐60‐based systems linear accelerators with C‐arm, O‐ring, or robotic arm design. Each possesses distinct features, such beam energy the degrees freedom in planning, which influence their...
TPS615 Background: The optimal frontline management strategy for oligometastatic renal cell carcinoma (omRCC) is uncertain. Systemic therapy (ST) the standard, but Stereotactic Ablative Radiation (SAbR) offers an alternative, and may potentially reduce ST-associated toxicity. However, concerns remain about occult micro-metastases progression with SAbR. Retrospective studies suggest focal therapies improve outcomes in select patients, prospective data limited. This ECOG-ACRIN phase 3...
Importance: Artificial intelligence (AI) foundation models such as Segment Anything Model 2 (SAM 2) offer potential for semi-automated image segmentation with minimal fine-tuning, but their performance in specialized clinical tasks like radiation therapy planning are not well characterized. Objective: To evaluate the of SAM segmenting pre-operative intact prostate and post-operative fossa targets radiotherapy planning. Design, Setting, Participants: Retrospective cohort study deploying...
Abstract Purpose Training deep learning dose prediction models for the latest cutting‐edge radiotherapy techniques, such as AI‐based nodal (AINRT) and Daily Adaptive (DA‐AINRT), is challenging due to limited data. This study aims investigate impact of transfer on predictive performance an existing clinical model its potential enhance emerging approaches head neck cancer patients. Method We evaluated benefits by fine‐tuning a Hierarchically Densely Connected U‐net both AINRT DA‐AINRT patient...
In radiotherapy, a trade-off exists between computational workload/speed and dose calculation accuracy. Calculation methods like pencil-beam convolution can be much faster than Monte-Carlo methods, but less accurate. The difference, mostly caused by inhomogeneities electronic disequilibrium, is highly correlated with the distribution underlying anatomical tissue density. We hypothesize that conversion scheme established to boost low-accuracy doses high-accuracy, using intensity information...
Typically, the current dose prediction models are limited to small amounts of data and require re-training for a specific site, often leading suboptimal performance. We propose site-agnostic, 3D distribution model using deep learning that can leverage from any treatment thus increasing total available train model. Applying our proposed new target site requires only brief fine-tuning involves no modifications input channels or its parameters. Thus, it be efficiently adapted different even...