Ninety per cent of the world's data have been generated in the last 5 years ( Machine learning: the power and promise of computers that learn by example . Report no. DES4702. Issued April 2017. Royal Society). A small fraction of these data is collected with the aim of validating specific hypotheses. These studies are led by the development of mechanistic models focused on the causality of input–output relationships. However, the vast majority is aimed at supporting statistical or correlation studies that bypass the need for causality and focus exclusively on prediction. Along these lines, there has been a vast increase in the use of machine learning models, in particular in the biomedical and clinical sciences, to try and keep pace with the rate of data generation. Recent successes now beg the question of whether mechanistic models are still relevant in this area. Said otherwise, why should we try to understand the mechanisms of disease progression when we can use machine learning tools to directly predict disease outcome?

Resting state functional magnetic resonance imaging (rsfMRI), and the underlying brain networks identified with it, have recently appeared as a promising avenue for the evaluation of functional deficits without the need for active patient participation. We hypothesize here that such alteration can be inferred from tissue damage within the network. From an engineering perspective, the numerical prediction of tissue mechanical damage following an impact remains computationally expensive. To this end, we propose a numerical framework aimed at predicting resting state network disruption for an arbitrary head impact, as described by the head velocity, location and angle of impact, and impactor shape. The proposed method uses a library of precalculated cases leveraged by a machine learning layer for efficient and quick prediction. The accuracy of the machine learning layer is illustrated with a dummy fall case, where the machine learning prediction is shown to closely match the full simulation results. The resulting framework is finally tested against the rsfMRI data of nine TBI patients scanned within 24 h of injury, for which paramedical information was used to reconstruct in silico the accident. While more clinical data are required for full validation, this approach opens the door to (i) on-the-fly prediction of rsfMRI alterations, readily measurable on clinical premises from paramedical data, and (ii) reverse-engineered accident reconstruction through rsfMRI measurements.

A modular pipeline for improving the constitutive modelling of composite materials is proposed.The method is leveraged here for the development of subject-specific spatially-varying brain white matter mechanical properties. For this application, white matter microstructural information is extracted from diffusion magnetic resonance imaging (dMRI) scans, and used to generate hundreds of representative volume elements (RVEs) with randomly distributed fibre properties. By automatically running finite element analyses on these RVEs, stress-strain curves corresponding to multiple RVE-specific loading cases are produced. A mesoscopic constitutive model homogenising the RVEs’ behaviour is then calibrated for each RVE, producing a library of calibrated parameters against each set of RVE microstructural characteristics. Finally, a machine learning layer is implemented to predict the constitutive model parameters directly from any new microstructure. The results show that the methodology can predict calibrated mesoscopic material properties with high accuracy. More generally, the overall framework allows for the efficient simulation of the spatially-varying mechanical behaviour of composite materials when experimentally measured location-specific fibre geometrical characteristics are provided.

Evolutionary Algorithms are powerful optimization techniques which have been applied to many different problems, from complex mathematical functions to real-world applications. Some studies report performance improvements through the combination of d

Traditionally, clinical devices are designed, tested and improved through lengthy and expensive laboratory experiments and clinical trials [1]. More recently, computational methods have allowed for rapid testing, speeding up the design process and enabling far more complete searches of design space. While computational models cannot fully capture the complexities of biological systems, they provide valuable insights into crucial underlying mechanisms, such as the effects of fluid-structure interactions (FSIs). In this paper we present a modular, partitioned, computational FSI pipeline whereby 2D reduced order models guide the 3D design of the problem of interest. This framework is applied to the problem of hydrocephalus shunt occlusion. Hydrocephalus is a medical condition characterised by an excess of cerebrospinal fluid (CSF) in the brain, and is commonly treated with the insertion of a shunt system. This system includes a ventricular catheter component - a hollow tube with inlet holes arranged in the tube wall close to the closed tip - which is positioned in the lateral ventricles of the brain. Despite recent improvements in the catheter material, this treatment still has high failure rates, most often due to the blockage of the catheter by the Choroid Plexus (ChP) tissue. We use an idealised FSI model to compare existing catheter designs by considering the deformation of the ChP under CSF flow in the ventricle environment in an hydrocephalus scenario. To the best of our knowledge, this is the first computational framework to directly incorporate the deformation of the ChP to discriminate between catheter designs. The faster 2D model is used in a comprehensive parameter sweep of the catheter design domain, and motivates a new design, then confirmed to be an improvement when tested in the full 3D domain. This approach demonstrates the success of using reduced order methods to guide the design of a more complex problem.