Project Description

Mechanical cues are key drivers for adaptation of human tissues. A perfect example are wounds. In vitro, animal, and clinical evidence have shown that excessive stress leads to maladaptation such as hypertrophic scar and contracture. Moreover, mechanical features of the wound environment dictate the subsequent remodeling or regeneration of the tissue. However, we are still unable to predict wound healing in routine clinical settings, let alone in more complex scenarios such as healing after lumpectomy or diabetic wounds. The postdoctoral fellow working on this project would help address two key challenges that have broad implications for integration of engineering in medicine: i) there is still a lack of comprehensive data regarding wound mechanics and mechanobiology; ii) due to the limited number of observations that can be collected for an individual patient, design of treatment for human patients requires pooling of population-level data and the integration of mechanistic knowledge from animal and in vitro evidence. The objective of this project is to create a physics-informed machine learning model of wound healing that predicts scarring with quantified uncertainty based on previous mechanistic knowledge, and new data on wound healing in response to treatments that directly control the mechanical environment, in particular lumpectomy.

Start Date

08/01/2021

Postdoc Qualifications

Experience in machine learning algorithms is required, such as basic knowledge of neural networks and Gaussian processes.
Experience in physics-based modeling is required, such as solution of partial differential equations for mass transport or mechanical equilibrium.

Not required: knowledge of wound healing biology is not required, although some exposure to biological system modeling would be advantageous. 

Co-Advisors

Adrian Buganza Tepole, abuganza@purdue.edu, Mechanical Engineering
https://engineering.purdue.edu/tepolelab/

Sherry Voytik-Harbin, harbins@purdue.edu, Biomedical Engineering

References

Tepole AB. Computational systems mechanobiology of wound healing. Computer methods in applied mechanics and engineering. 2017 Feb 1;314:46-70.

Lee T, Bilionis I, Tepole AB. Propagation of uncertainty in the mechanical and biological response of growing tissues using multi-fidelity Gaussian process regression. Computer Methods in Applied Mechanics and Engineering. 2020 Feb 1;359:112724.

Sohutskay DO, Buno KP, Tholpady SS, Nier SJ, Voytik-Harbin SL. Design and biofabrication of dermal regeneration scaffolds: role of oligomeric collagen fibril density and architecture. Regenerative Medicine. 2020 Mar;15(2):1295-312.