1. Current IMAG Meeting
  2. 2023 Agenda
  3. Oral Presentation 1 (IMAG-AND Futures)

Oral Presentation 1 (IMAG-AND Futures)

1:20-1:40 pm               “Integrating multiple scales and imaging modalities to predict tumor response for individual patients and generate personalized therapy regimens”

Angela Jarrett, U. Texas

Angela Jarrett (photo)BIO: Thanks to the Peter O’Donnell, Jr. Postdoctoral Fellowship, Angela came to the Oden Institute for Computational Engineering and Sciences at the University of Texas at Austin in the summer of 2016. Since then she has been working in Dr. Thomas Yankeelov's Oncological Modeling Group in the Center for Computational Oncology and the Livestrong Cancer Institutes. She received bachelor's and master's degrees in Mathematics at Florida State University, and she completed her Ph.D., specializing in Biomedical Mathematics, in 2016 at FSU. Her graduate work consisted of developing uncertainty and sensitivity analysis methods in conjunction with mathematical modeling of persistent and resistant infections with respect to host responses and treatment, specifically for MRSA and HIV. Her research at UT focuses on developing a clinically-relevant tumor growth model for breast cancer that can capture tumor response to neoadjuvant therapies using patient-specific quantitative imaging data. Additionally, she has developed mathematical models at the in vitro and in vivo scales for HER2+ breast cancer that include the effects of cytotoxic and anti-HER2 targeted treatment as well as immune response dynamics. The aim of these projects is to leverage tumor cell response data to individual drugs to not only improve the overall predictive ability of the clinical model but also to explore alternative therapeutic strategies. 

Angela Jarrett.pdf91.89 KB

Comment

Comment

Very nice model and integration with imaging! The effect of stress on diffusion is a key aspect that you have put into your model to enable patient specific modeling. Could you say a few words about why one can neglect dilatational effects on the diffusion coefficient (that is, why is it appropriate to use von Mises equivalent stress)? Thanks! Guy

Submitted by Guy Genin, Was… (not verified) on Tue, 03/17/2020 - 13:32

Comment

Please feel free to email me if you would like to ask a follow-up question!

ajarrett@utexas.edu

Submitted by Angela (not verified) on Tue, 03/17/2020 - 14:00

Comment

Do you have any thoughts on how to include chemoRx side effects so as to identify regimens that would minimize these?

Submitted by Bill Lytton (not verified) on Tue, 03/17/2020 - 13:35

Comment

Great talk! The different therapies (smaller dose over time) give very different results for each patient. Is there a reason for this response that can be learned from the data? Is there an underlying feature in the MR images that can explain which therapy will do better? 

Submitted by abuganza on Tue, 03/17/2020 - 13:35

Comment

Great talk! Very interesting work! Have you shown your model to oncologists/radiologists and, if so, what is their feedback?

Submitted by Shayn Peirce-Cottler on Tue, 03/17/2020 - 13:36

Comment

Great job! Former Genentech imaging scientist new to NIBIB as of early March! We like to think 'more is better', what is the thought process for daily dosing being less effective for patients 6 and 7 (sorry not sure what slide # but this was the plot with colorful dots and labeled Tissue Cellularity)? Is daily dosing ... lower doses on a daily basis to keep TOTAL dose the same? So, maybe missing some critical threshold with higher doses, less frequently?

Submitted by Joan (not verified) on Tue, 03/17/2020 - 13:36

Comment

Nice results!  How do you processing the multiple modes of images, which tools, and how do you register across scans and modalities?

 

Thanks,

Rob

Submitted by rsmacleod on Tue, 03/17/2020 - 13:37

Comment

Excellent presentation! How does the handling of protein vs. small-molecule transport compare to the work of Dane Wittrup's group?

E.g., https://www.ncbi.nlm.nih.gov/pubmed/19825804

Submitted by jmhaugh on Tue, 03/17/2020 - 13:37

Comment

What tools do you use? What computer languages? How much computing power you use?

Submitted by jbarhak on Tue, 03/17/2020 - 13:40

Comment

You just mentioned using the objective function to account for adverse effects. How about adding a coupled, dynamical model of toxicity effects, and using its output in your objective function? 

Do you anticipate that a toxicity model needs to be dynamically coupled to the growth/treatment model? Or can it be done post hoc after the simulation as as a simple AUC (similar to your current dosing constraints)? 

Submitted by mathcancer on Tue, 03/17/2020 - 13:41

Comment

I am curious to know if you have discussed your modeling approach with oncologists as it might be useful for them. 

Submitted by nkkchem on Tue, 03/17/2020 - 13:46

Comment

I was hoping to email back each of you to start larger conversations, but it appears these comment entries do not list your contact info! So I would love for each of you to email me directly so we can have a more substantive back/forth.

ajarrett@utexas.edu

Submitted by Angela (not verified) on Tue, 03/17/2020 - 13:57