Digital mirrors – mathematical refinements of digital twins – are predictive models that simulate real-time states of physical systems and are recognised as crucial for personalised healthcare in the UK. Although their adoption in healthcare lags behind other sectors, on an individual basis, these models have advanced to the point of implementation for performance optimisation of medical devices; likewise, progress has been made for using these models to enhance orthopaedic surgical planning and postoperative care. As mentioned, advancements have largely been at the individual-scale. However, creating customised models is expensive and has significant environmental impacts caused by the heavy demands of data collection and model training/development [1]. A more efficient method involves sharing information between similar systems to build a network (or ecosystem [2]) of models, which also supports the UK’s efforts to reduce carbon emissions [3].

This project aims to adapt digital-mirror technologies to ecosystems (for example, from a single hip implant to a group of hip implants). Data and expert knowledge will be used to build detailed statistical models that reflect the diverse characteristics and relationships within groups. For similar systems, a multilevel approach will be implemented, pooling parameters across models to facilitate knowledge transfer. Emphasis will be placed on understanding the unique physical phenomena specific to each system. Machine-learned mappings/functions that leverage the abstraction of these physical and digital systems to geometric spaces [4] will then be learnt to connect to systems that are more diverse, utilising intermediate structures/models (e.g., physics-based models) as needed, to allow for transfer across the entire ecosystem. This functionality will allow informed predictions and decisions for structures/groups across ecosystems for which there are no, or limited, data available. For instance, it may be beneficial to transfer information from a demographic of elderly hip replacement patients, for whom there are extensive data, to a younger cohort, for whom data are limited. These concepts have shown promise for monitoring the health of infrastructure but remain completely unexploited for the healthcare sector.

While this concept of scaling from individuals to ecosystems is pertinent for all digital-mirror applications, this project will focus on developing technologies to address critical problems in orthopaedics. Orthopaedic conditions differ significantly from one individual to another, and are influenced by factors such as age, bone density, muscle strength, and lifestyle. Personalised healthcare can allow for treatment plans to be tailored to the unique characteristics of each patient, thus improving patient outcomes. In procedures such as hip replacements, digital mirrors could be utilised preoperatively to determine the most suitable implant size, reducing both the duration of surgery and material waste. Post-surgery, these models can also be used to project the course of recovery and to guide management strategies, facilitating the monitoring of the healing process and the long-term health of the bone-implant interface and adjacent tissues and structures. By developing a digital-mirror framework that enhances prediction accuracy while reducing costs and environmental impact, this project seeks to enable wider adoption and accessibility of personalised healthcare in orthopaedics. The candidate will liaise with healthcare professionals to identify and acquire appropriate data from multiple sources such as wearable devices, bone scans, etc. Considering these multiple data sources, they will develop new digital-mirror technologies for orthopaedic concerns, using physics-based, statistical, and mathematical modelling approaches. The ideal candidate will have a background in engineering, computer science, mathematics, and/or physics. While specific expertise in all aspects of the project is not required, they should have a strong desire to expand their knowledge in unfamiliar areas (with support from supervisors and in-house collaborators).

The successful candidate will have the chance to join the globally-renowned Dynamics Research Group (DRG). We are committed to promoting diversity, supporting flexible working arrangements, and recognising individual achievements. Our group is highly collaborative and promotes a vibrant social atmosphere.

Interested candidates are strongly encouraged to contact the project supervisors to discuss your interest in and suitability for the project prior to submitting your application.

Please refer to the EPSRC DLA webpage for detailed information about the EPSRC DLA and how to apply.