An active lung model for novel disease insight

This project aims to unveil fundamental lung mechanical behaviour that dictate why certain lung pathologies impact treatment outcomes and quality of recovery. The COVID-19 pandemic has raised more questions about the variability of effects experienced by the population and the fallout in terms of long term effects (scarring, fibrosis). This project will use computational modelling and novel experimental procedures to recapitulate lung adaptation during different stages of lung disease and contribute novel insight to different disease pathways. The study will make use of additive manufacture, novel polymer formulations, including stimuli responsive polymers, to create high-fidelity active lung models with dynamic motion, validating computational models. Such devices will advance the capabilities to adapt to new respiratory threats in the future and train medical practitioners.

Active lung models: Explore current literature on stimuli responsive polymers, biomaterials and biomimetics. Synthesize a variety of different dynamic phase changing and adaptive polymers using methodologies from existing literature. Experiment with the polymer chemistry, whereby monomer quantities are the independent variables in well constrained methodologies.

Assess the mechanical properties of synthesized polymer samples across phases using various mechanical testing methods (e.g. tensile testing). Digital image correlation will be used to collect spatial variability of response during simple mechanical tests as well as under variable stimuli control. Data will enable the development of a computational toolbox using deep learning techniques to correlate monomer quantities with the tensile properties.

Develop mimics for lung tissues based on their properties in medical literature and the aligning quantities of monomers from the toolbox. Use the polymer tissue mimics to experiment with local stimuli response. Develop simplified one dimensional lung experiments before moving into three dimensional geometries to fine tune basic physiological behaviour. The one dimensional test platform aims to address permeability and mechanical variability in disease (benchmarked with clinical data). Examples such as scarring and oedema will be simulated and optimised in this simple configuration before being integrated into a three dimensional setup.

Explore manufacturing methods for more complex geometries e.g. moulding and additive manufacture, building on expertise in the group with advanced high-resolution imaging of lung. Develop and test increasingly more advanced three dimensional lung models validated against mechanical ventilation type experiments.

Computational modelling: Further develop workflows for the segmentation of CT scans of lungs into 3D models. Explore identification of disease tissues from CT and integration of these disease pathways into the computational mode framework. Isolating disease mechanisms responsible for a given CT greyscale/observation is a significant radiological challenge, which can be addressed here through controllable experiments (fixing/changing mechanical properties in the physical model benchmarks) - comparing clinical, simulation and physical experimental data.

Further develop existing codes to produce simulations that can inform fabrication and vice versa (use the physical models as a controlled known model benchmark). Use data from the simulations to inform parameters for the experimental testing for a targeted pathology. Analyse data from the computational model and compare findings with the experimental results recorded from sensors integrated into the experimental setup; and clinical data.