Aotearoa Fellow, Auckland Bioengineering Institute
Hi, my name is Claire Miller and I am an applied mathematician in the area of computational biology. Currently, I am an Aotearoa Fellow at the Auckland Bioengineering Institute, University of Auckland, New Zealand developing spatiotemporal mathematical models to better understand endometriosis onset and early growth.
Before my fellowship, I was a postdoctoral researcher at the University of Amsterdam as a part of the INSIST project, with the goal of conducting in silico clinical trials for acute ischemic stroke. In my PhD (at the University of Melbourne) I developed a multiscale model of epidermal (skin) tissue to understand how the tissue regulates its height. I have found that the challenge, and the appeal, of computational biology is building a mathematical system that is able to sufficiently, and demonstrably, represent highly complex biological systems with a limited knowledge of the parameters.
Prior to my PhD I worked on the development of fire progression models at CSIRO in Melbourne, Australia. There is an obvious need for these types of models in Australia as bushfires are so common and can be so devastating. The interesting feature of this project was the need for creative ways to solve problems where the balance between computational speed and model accuracy is critical to the value of the model.
My experience in research has motivated me to pursue work in multidisciplinary teams/projects, and has shown the great benefits of connections between academia and industry. I believe it is through such teams and connections that we are able to have a targeted impact on society. I am also passionate about problems in womens health, and this is an area I am excited to pursue in my current research.
For further information, I have details on each research project below, refer to my CV, or feel free to contact me.
The following is a list of all my publications in reverse chronological order. For citation information and abstracts please refer to my Google Scholar.
Claire Miller, Raymond M. Padmos, Max van der Kolk, Tamás I. Józsa, Noor Samuels, Yidan Xue, Stephen J. Payne, and Alfons G. Hoekstra. “In Silico Trials for Treatment of Acute Ischemic Stroke: Design and Implementation”. Computers in Biology and Medicine 137 (Oct. 2021), p. 104802. DOI: 10.1016/j.compbiomed.2021.104802.
Giulia Luraghi, Sara Bridio, Claire Miller, Alfons Hoekstra, Jose Felix Rodriguez Matas, and Francesco Migliavacca. “Applicability Analysis to Evaluate Credibility of an in Silico Thrombectomy Procedure”. In: Journal of Biomechanics 126 (Sept. 2021), p. 110631. DOI: 10.1016/j.jbiomech.2021.110631.
Claire Miller, Edmund Crampin, and James Osborne. “Multiscale Modelling of Desquamation in the Interfollicular Epidermis”. PLOS Computational Biology 18(8): e1010368 (Aug. 2022). https://doi.org/10.1371/journal.pcbi.1010368
Claire Miller, Edmund Crampin, and James M. Osborne. “Maintaining the Proliferative Cell Niche in Multicellular Models of Epithelia”. Journal of Theoretical Biology 527 (Oct. 2021), p. 110807. DOI: 10.1016/j.jtbi.2021.110807.
Claire Miller, Matt Plucinski, Alec Stephenson, Carolyn Huston, Kay Charman, Mahesh Prakash, Andrew Sullivan, Simon Dunstall. Electrically caused wildfires in Victoria, Australia are over-represented when fire danger is elevated. Landscape and Urban Planning 167:267-274 (2017).
James Hilton, Claire Miller, Jason Sharples, Andrew Sullivan. Curvature effects in the dynamic propagation of wildfires. International Journal of Wildland Fire 25(12):1238-1251 (2016).
James Hilton, Claire Miller, Andrew Sullivan. A power series formation for two-dimensional wildfire shapes. International Journal of Wildland Fire 25(9):970-979 (2016).
James Hilton, Claire Miller, Andrew Sullivan, Chris Rucinski. Effects of spatial and temporal variation in environmental conditions on simulation of wildfire spread. Environmental Modelling and Software 67:118-127 (2015).
Claire Miller, Max van der Kolk, Raymond Padmos, Tamás Józsa, and Alfons Hoekstra. “Uncertainty Quantification of Coupled 1D Arterial Blood Flow and 3D Tissue Perfusion Models Using the INSIST Framework”. In: Computational Science – ICCS 2021. Lecture Notes in Computer Science. Cham: Springer International Publishing, 2021, pp. 691–697. ISBN: 978-3-030-77980-1. DOI: 10.1007⁄978-3-030-77980-1_52.
Max van der Kolk, Claire Miller, Raymond Padmos, Victor Azizi, and Alfons Hoekstra. “Des-Ist: A Simulation Framework to Streamline Event-Based In Silico Trials”. In: Computational Science – ICCS 2021. Lecture Notes in Computer Science. Cham: Springer International Publishing, 2021, pp. 648–654. ISBN: 978-3-030-77967-2. DOI: 10.1007⁄978-3-030-77967-2_53.
Claire Miller, James Hilton, Andrew Sullivan, Mahesh Prakash. SPARK – A bushfire spread prediction tool. Environmental Software Systems; Infrastructures, Services and Applications, pp. 262-271 (2015).
James Hilton, Claire Miller, Matt Bolger, Lachlan Hetherton, Mahesh Prakash. An Integrated Workflow Architecture for Natural Hazards, Analytics and Decision Support. Environmental Software Systems; Infrastructures, Services and Applications, pp. 333-342 (2015).
Gary Delaney, James Hilton, Paul Cleary, Claire Miller. The role of inter-grain friction in determining the mechanical and structural properties of superellipsoid packings. Powders and Grains 2013: Proceedings of the 7th International Conference on Micromechanics of Granular Media (2013).
In silico clinical trials (postdoc)
Supervisor: Prof. Alfons Hoekstra
There are many potential clinical uses for mathematical models, one of which is in silico clinical trials. Traditional clinical trials are becoming increasingly expensive–—several difficulties relate to their reliance on large cohorts to provide sufficient statistical power for regulatory approval. In silico clinical trials are an emerging concept in which a cohort of virtual patients are used to reduce the required number of real participants or refine the inclusion criteria for patients in a trial.
I am part of an EU Horizon 2020 project, INSIST, developing in silico clinical trials for the treatment of acute ischemic stroke (the occlusion of an artery in the brain). My research focuses on, firstly, how we combine the diverse range of models required to model stroke injury and treatment and, secondly, how we prove the credibility of the trials such that they can be used as evidence in clinical trial submissions to regulators. In order to model stroke injury and treatment it is necessary to incorporate models for: the generation of virtual patients, arterial blood flow, blood perfusion in the brain, brain tissue death, thrombolysis (IV drug) and thrombectomy (surgical removal) treatments, and predicted functional outcome. The spatiotemporal scales and the methodologies used for these models are diverse, and include empirical, statistical, and finite element approaches. Consequently, integrating them is a challenge—one which is assisted by the expertise of clinicians who can provide physiological insight. In order to run these diverse models within a single modelling framework we use an event-based modelling approach: we model three events—pre-stroke, stroke, and treatment—and assume that the system instantaneously reaches a new steady state at each event. This allows models to be run consecutively, for each event, and avoids the need to integrate models solved at second time scales over the hour time scales between stroke onset, treatment, and final patient functional outcome. Further details on this approach can be found in this paper.
The second aspect of my research focuses on proving credibility of the trial such that it can be used clinically, which involves a high level of collaboration with clinical researchers. Assessing the credibility and uncertainty of the models we use is critical for translating research to a clinical setting, which is a long-term goal of my research. There is currently no regulatory standard for proving trial credibility, and development of an appropriate framework is still an open area of research and discussion in the in silico clinical trial community, and within regulatory organisations. In addition to performing in-depth validation studies of both individual models and the full trial, there are several other aspects to proving credibility such as analysing context of use, identifying and quantifying epistemic and aleatoric errors, and evaluating the allowable level of uncertainty in model outcome. To this end, I am performing uncertainty analyses on the trial, in which the effect of uncertainty in model inputs and parameters is determined; their effect on the uncertainty surrounding the predicted outcome is analysed; and the parameters with greatest effect identified. I perform these studies on a subset of the models included in the trial, using a quasi-Monte Carlo approach, and a sensitivity analysis, using Sobol indices, to determine the effect of unknown parameters on the outcome of interest. Results of this work for uncertainty in the blood flow model, and its effect on the predicted volume of dead tissue, can be found here.
Multiscale agent-based modelling of the epidermis (PhD)
Thesis title: Understanding the regulation of epidermal tissue structure by molecular and cellular processes using multi-scale models.
Supervisors: Dr James Osborne and Prof. Edmund Crampin.
Mechanistic mathematical models allow us to investigate aspects of biological tissues and diseases that can be expensive, difficult, or impossible to investigate in vitro or in vivo. Multiscale models specifically can lead to a better understanding of how the interactions of processes at different spatial and temporal scales interact and influence tissue structure and dynamics. This was a key focus of my PhD: I investigated how mechanisms acting at cellular and subcellular scales regulated the thickness of the epidermis (the outermost layer of the skin). The developed multiscale model used an overlapping spheres multicellular tissue model, including a novel cellular mechanism to control division direction during proliferation, and a novel enzyme kinetics model for the subcellular interactions controlling desquamation, which is the degradation of adhesion between cells that results in loss of the cells from the top of the tissue. Balancing these processes of cell proliferation and loss produced a tissue model that was able to self-regulate its thickness.
Agent-based models, as opposed to continuum models, model each cell individually. The disadvantage of these models is their high computational requirement; however, the advantage is that different cells can be in different states, allowing investigations into heterogeneities in the system. The overlapping spheres approach models cells as spheres that interact both with each other and external stimuli, with the purpose of understanding their collective behaviour. The novel cellular model I developed during my PhD was a rotational force model, motivated by the experimentally observed regulation of division orientation in the skin. This model solved a modelling issue of erroneous loss of proliferative cells from the stem cell niche, which in the epidermis is the bottom-most layer of cells. Further details on this mechanism can be found in this paper.
Subcellular models model processes occurring at subcellular spatial scales, such as protein interactions. I developed an enzyme kinetics model to model protein interactions for the degradation of the adhesion proteins between cells, which causes desquamation of the cells from the top of the tissue through external environmental forces. This model was based on interactions hypothesised in the experimental literature and was parameterised using in vitro data from the literature. Integrating this model into the multicellular model allowed me to investigate hypotheses on diseases related to excessive desquamation and analyse the effect of heterogeneity in protein levels released by the cells. The implementation and results of the subcellular and multiscale model can be found in this paper.
My model is built using the Chaste libraries. For more details I refer you to https://www.cs.ox.ac.uk/chaste/.
Bushfire progression modelling (2013-2016)
In Australian landscapes, bushfires can cause devestating damage to communities as well as fatalities. If we want to best combat these fires, understanding where they are likely to spread is critical. Getting this information as quickly as possible is vital to decision makers.
When I was at CSIRO I worked on a team developing a fire spread prediction model. The model was developed for use in operational situations and was implemented using a level set method. The rate of spread model that determined the level set function was based on empirical data. For more information about the model, I refer you to the following papers: model implementation, and a high level description of the prediction tool. Alternatively see the Spark website.
The model was also developed for fire research. This included understanding the effect of uncertainty in inputs to model output, and investigating different rate of spread functions. Papers on this research: the effect of variation in inputs, and the effect of including curvature in the spread function
A lot of my work in this area was around data collection for input into the model. This included stabilisation and conversion of video footage of fires into 2 dimensional fire lines, and sourcing of fuel data from different sources such as government data or satellite data.
In addition, I did some research into understanding why powerline fires tended to be more significant than other ignition types. The paper that came from this work can be accessed here.
Outside of my research, I love getting outdoors. I try and spend as many weekends as possible either multi-day hiking, bouldering, or skiing in the winter. I also have a website about multi-day hiking in Australia: https://www.glampliteoz.com.
My current local time is .