OHSU Digital Collections (Oregon Health and Science University)

Tapia-Chávez, César

Gaps in healthcare provider training are placing children at risk of not being identified as human trafficking victims. The lack of formal training hinders providers from recognizing the signs of trafficking and knowledge of the local, state, and national resources created to facilitate safe intervention.  Students participated in a child trafficking (CT) simulation, followed by an educational debrief session. The simulation utilized a standardized patient (SP) to exhibit realistic, high-yield, red-flag characteristics of trafficking to illicit students’ recognition of and response to a victim

Clark, Samantha

Advancing pediatric nurse practitioner training: a child trafficking simulation

Alexander, Nancy J.

Rowland, Levi Lindsay

This dissertation evaluates how the decision-making regarding blood transfusion is done in trauma care and what variables are used. Then the variables discovered in the first part are evaluated statistically and used to build machine learning models to predict patients in need of blood transfusion, then the fusion model using tabular data and chest x-rays was build and evaluated. Lastly, weather data and time series were used to build a predictive model to predict trauma admissions

Kolesnikov, Michael

Feasibility of using machine learning for clinical decision support to optimize transfusion practices in trauma care

A work focused on creating a model which simulates patient treatment shift information. This is useful for the purpose of exploring the effects of motion on patient treatment outcomes as well as exploring how current treatments might be changed given new patient information

Christensen, Chancelor

Wilson, Byron

Title: THE USE OF HIERARCHICAL MODELING IN PROSTATE STEREOTACTIC BODY 
RADIATION THERAPY 
 
Background: Prostate SBRT demands high geometric precision due to large fractional 
doses and steep dose gradients. Conventional PTV margins are typically derived from 
population-level estimates and may not reflect individual patient-specific motion. This 
thesis introduces a Bayesian hierarchical framework to simulate intrafraction motion and 
evaluate its dosimetric consequences, with emphasis on anatomically defined posterior 
margins used in clinical practice at OHSU. 
Methods: A cohort of 205 patients treated with five-fraction prostate SBRT from 2020 to 
2024 was retrospectively analyzed. Mid-treatment shifts were extracted from CBCT records 
and used to represent unintentional motion during treatment. These data trained a 
multivariate Student-T model with partial pooling, capturing population-level trends while 
retaining individual variability. Posterior predictive sampling generated synthetic shifts, 
which were applied to patient dose grids using rigid transformations. Cumulative dose 
distributions were recalculated and evaluated using standard DVH metrics. 
Results: Simulated shifts reproduced the central tendency, spread, and directional 
correlations of recorded patient data. Failure rates under clinical margins closely matched 
those in patient records, and correlation structures between axes were preserved. Case 
studies confirmed directional fidelity and dose displacement, particularly in high-motion 
scenarios. Population-level analysis showed close agreement in DVH deviations across 
prostate, bladder, and rectum. For instance, mean anterior shifts were nearly identical (–
0.151 mm patient vs. –0.119 mm simulated), and anterior–superior correlations were 
similar (r=0.422r = 0.422r=0.422 vs. r=0.357r = 0.357r=0.357, R2=0.178R^2 = 
0.178R2=0.178 vs. 0.127). Most DVH differences were small and statistically insignificant. 
Conclusion: This modeling framework offers a flexible approach to reconstructing patient-
specific motion, even with limited empirical data. By leveraging partial pooling, it balances 
individual detail with population-informed inference. Its ability to replicate real-world 
distributions, failure rates, and dose perturbations underscores its validity as a simulation 
engine. Importantly, this approach enables probabilistic “what-if” testing without requiring 
additional imaging or patient intervention—offering a principled method for stress-testing 
clinical practices. In doing so, it advances the clinical conversation around personalization 
in radiation therapy and highlights the role of statistical modeling in improving treatment 
safety and precision. 
 


Not a member yet

Advancing pediatric nurse practitioner training: a child trafficking simulation

Feasibility of using machine learning for clinical decision support to optimize transfusion practices in trauma care

The use of hierarchical modeling in prostate stereotactic body radiation therapy

3,119

38,149