Data Dialogue

Title: Computational Techniques for Biomechanical Imaging: a Physics-based Data Analytics Approach

Inverse problems arise in many applications in science and engineering. For instance, in medicine, the mechanical properties of soft tissue are discriminant features of cardiovascular disease, malignant tumors, and liver disease, among others. For this reason, there has been wide interest in the last few decades in the development of non-invasive techniques (ultrasound and MRI-based) for estimating mechanical properties of human tissue. I will discuss our ongoing work on computational techniques for inverse problems in the context of biomechanical imaging. I will present general inverse problem formulations based on optimization with operator equations constraints (e.g. boundary-value problems) as well as general software development strategies amenable to general, large-scale inverse problems.