The MICDE PhD Student Seminar Series showcases the research of students in the Ph.D. in Scientific Computing. These events are open to the public, but we request that all who plan to attend register in advance.

Applications of the phase-field method to polycrystalline materials

Phase-field modeling is a common diffuse interface method for simulating microstructure evolution due to its ability to capture complex morphologies without the need for explicitly tracking phase interfaces. A typical application of the phase-field method is polycrystalline grain growth during annealing, where grain boundaries migrate toward their centers of curvature. Recent studies have shown abnormally large grains can be grown in shape memory alloys during cyclic annealing due to additional driving forces generated during the growth and dissolution of second-phase precipitates. In this work, we model grain growth via a phase-field model that considers stored energy generated during the cyclic heat treatments. Applications of the phase-field method to experimentally acquired grain microstructures will also be discussed.

Zach Croft, Applied Physics

Zach is a PhD student in the Applied Physics program. He works in the field of computational materials science with an emphasis on phase-field modeling of polycrystalline evolution and solidification of alloys under Professor Katsuyo Thornton.

Using causal inference to estimate counterfactual disparity measures for access to weight management treatments

Type 2 Diabetes (T2D) is a prevalent condition with significant variation in outcomes based on race and ethnicity, underscoring the need for more improved prevention practices. Because effective weight management is a key component of T2D prevention, increasing access to evidence-based treatments for those most at-risk for developing T2D is imperative. Yet, existing population health management approaches do not typically measure disparities in access to treatments or do so in ways that do not account for the increased risk experienced by certain patient populations. This talk will (1) describe how causal inference was used to calculate counterfactual estimates of disparities in referral to weight management treatments among a population of adults with obesity, (2) compare counterfactual estimates generating from the standard approach vs. a risk-based approach , and (3) share UM research, computing, and other resources that supports this research.

Cassie Turner, Health Infrastructures and Learning Systems

Cassie has a joint appointment at Michigan Medicine and the Ann Arbor Veteran Affairs Health System, where she contributes to health research and practice focused on improving metabolic health through leveraging analytics, novel care models, and learning health systems approaches.