Growing Matter Lab
Computational Biomechanics at University of Manchester
We turn medical images and clinical data into patient-specific predictions.
Our purpose
We develop simulation and machine-learning methods for the mechanics of the human body — the aorta, the eye and the soft tissues that carry load. With clinicians and industry partners, we work to make high-fidelity modelling fast and reliable enough to inform individual patient care.
We use patient-specific models to understand how disease and damage develop in blood vessels, the eye and other soft tissues.
Quantitative biomechanical measures extracted from routine clinical imaging.
Machine-learned surrogates that bring high-fidelity analysis within clinical timescales.
Impact & spin-offs
Our computational research also creates tools and companies beyond biomechanics. Precica shows how a challenge encountered in the lab can lead to technology for industry.
University of Manchester spin-out
Launched 2026.
Making large datasets easier to store and use.
Founded by Ajay B. Harish, Precica is a University of Manchester spin-out developing data compression that keeps information searchable and ready for analysis.
The idea grew from the challenge of storing and exploring large simulation datasets, taking research from the lab into commercial development.
From the lab
Our paper Neck geometry and shape compactness as additional haemodynamic modulators in abdominal aortic aneurysms, by Nandurdikar et al., is now published in Royal Society Open Science.
Our paper Development of data-driven Bayesian shape optimisation framework for reverse flow reduction over a rounded ramp, by Yudianto et al., is now published in Computers & Fluids.
Our paper Exploring Neural Network Surrogates for High-Order Mesh-Free Interpolants, by Starepravo et al., has been accepted for publication in Computational Particle Mechanics. Read the paper.
Our paper Learning mesh-free discrete differential operators with self-supervised graph neural networks, by Starepravo et al., is now published online in Computer Methods in Applied Mechanics and Engineering (CMAME). It is assigned to the January 2027 issue.
