NeurIPS and APS DFD conferences

Congrats to Hunor for being selected to present his paper at NeurIPS (a very prestigious and leading deep learning and AI conference). Hunor’s work is titled “Modeling Coupled 1D PDEs of Cardiovascular Flow with Spatial Neural ODEs” and is based on his summer internship and collaboration with the Los Alamos National Lab.

Dr. Arzani, Hunor, and Siva will also present on scientific machine learning modeling of blood flow at the APS Division of Fluid Dynamics Conference. Dr. Arzani is part of this year’s invited speakers and will present at the Fluid Dynamics in Clinical Imaging Minisymposium. Nov 19-21 Washington, DC.

Wildfire modeling grant!

We have received an NSF EAGER grant (one year) from CBET for our proposal “EAGER: Understanding complex wind-driven wildfire propagation patterns with a dynamical systems approach”. We will use our lab’s expertise in modeling convective transport and dynamical systems to study wildfire transport. PI: Amir Arzani, Co-PI: Dr. Rob Stoll (Utah), Dr. Ali Tohidi (SJSU), and Dr. Fatemeh Afghah (Clemson).

Summer conferences (SB3C and USNCCM) + DDPS talk

Hunor will present his work titled “Enhancing Corrupt Cardiovascular Flow Data With Machine Learning” at the Summer Biomechanics, Bioengineering, and Biotransport (SB3C) conference. Dr. Arzani will chair the “Machine Learning in Biofluids” session (June 4-8, Vail, CO). Later this summer, Dr. Arzani will present our latest research on explainable AI (XAI) titled “Towards Interpretable and Generalizable Deep Learning in Mechanics” at the U.S. National Congress on Computational Mechanics (USNCCM) conference (July 23-27, Albuquerque, NM ). Dr. Arzani’s XAI presentation can be viewed below. Also, a longer version of this talk was recently delivered at the DDPS seminar series and can be viewed here.

Internships

Our undergrad researcher Ethan Shoemaker will start an internship at NASA Glenn Research Center this summer where he will continue his research on physics-informed neural networks (PINN). Ethan will then move to UC Irvine for grad school. Also, our PhD student Maryam has been going through multiple internship interviews at Apple and Google.

NSF CAREER Award!!

Dr. Arzani has received the highly prestigious NSF CAREER Award. The five-year project (~$508k) is titled “CAREER: Synergistic physics-based and deep learning cardiovascular flow modeling” and is funded by the NSF Office of Advanced Cyberinfrastructure Program. The award will build a foundation for scientific machine learning and fluid dynamics research and will enable lifetime leadership for integrating the related research with education activities.

APS DFD conference: back to in-person!

Dr. Arzani, Maryam, and Mostafa will present their work at APS DFD conference. Dr. Arzani is part of the organizing committee (abstract sorting and in charge of assigning session chairs) and will chair the “Biological Fluid Dynamics: Data-driven Hemodynamics” session. This is a new session at APS that we created for the first time due to the growing interest in using data science and machine learning in blood flow modeling. Nov 21-23 Phoenix, AZ.

New NSF grant

We have received a new Collaborative NSF grant (ECCS-Comms Circuits & Sens Sys Program) to support our scientific machine learning research. Specifically, we will be enhancing 4D flow MRI data by data assimilation, deep learning, and image processing. In collaboration with Roshan D’Souza at UW-Milwaukee (NAU PI: Amir Arzani; UWM PI: Roshan D’Souza). Thanks NSF!

SB3C + USNCCM conferences

Summer Biomechanics, Bioengineering, and Biotransport (SB3C):

Sara will present her work titled “Role of coherent structures in airflow mediated infectious disease spread with expiratory particles”, and MohammadReza will present his work titled “Aortic valve dynamics coupled with growth and remodeling due to aging and calcification”. Dr. Arzani will co-chair the “Patient Specific Flow and Physiology II” session.

U.S. National Congress on Computational Mechanics (USNCCM):

Dr. Arzani will present his work titled “Hybrid Physics-based and Data-driven Modeling of Near-wall Blood Flow with Physics-Informed Neural Networks”. The recorded talk could be accessed here:

Ali’s graduation

Ali defended his MS thesis. He published three first-author journal papers, presented at multiple conferences, and contributed to other projects in our lab. He has got three fellowships (UC Berkeley, Michigan, UIUC). Ali will join University of Michigan to do his PhD in data-driven modeling in fluid mechanics. He also won NAU’s 2020 Grad College Graduate Research Assistant Award ($1000).

Ali’s goodbye lunch. Photo taken right before the pandemic!