July 5th a mini-symposium on In-Silico Trials for Acute Ischemic Stroke was held at the PASC 2021
Description
Computer models and HPC simulations are becoming a very important approach to assist clinicans in treating diseases and devising new therapies. The goal of this minisymposium is to investigate several challenges associated with this approach. We will propose a selection of talks inspired by the problems raised in the H2020 project INSIST, for instance the estimation of success scores of a medical intervention through modeling (e.g. thrombolysis and thrombectomy processes in the treatment of a stroke, the impact of the lack of oxygen in the brain), and the way to build a virtual population of patients to propose new treatments and avoid in-vivo and in-vitro experiments. The problem of validation and uncertainty quantification of the numerical models will also be considered.
Design and Implementation of In-Silico Trials for Acute Ischemic Stroke
Description
In silico trials are patient-specific simulations on a cohort of virtual patients to support the development and evaluation of medical devices, drugs and treatment. In silico trials have the potential to refine, reduce cost, and partially replace current clinical trials or animal experimentation. We present the design and implementation of an in silico trial for treatment of acute ischemic stroke, and demonstrate running a trial. The in silico trial is implemented as an event-based simulation for stroke and treatment, and is designed to be compartmental to aid development and reproducibility. A statistical population model will be used to generate cohorts of virtual patients. Sophisticated 3D simulations can be replaced by surrogate and statistical models such that the framework can be run for large cohorts of patients. Ways to overcome some of the challenges and difficulties in setting up such an in silico trial are discussed.
Thrombolysis in Virtual Patients
Presenter : Remy PetkantchinDescription
Ischemic stroke disease is the 1st cause of disability and 3rd cause of death in civilized countries. Treatment options are thrombolysis and thrombectomy. The latter consists in the mechanical removal of the blood clot, through surgery. The former, consists in the chemical breakdown of the clot, by injecting a fibrin-lysing product, commonly called tPA. However, treatment success rates are still very low: only 30% of the patients will have fully recovered 3 months after treatment. Our goal is to better understand the thrombolysis process in patients, using numerical modeling and HPC simulations. We consider 3D Lattice Boltzmann flow simulations on top of which we add the processes of tPA transport and fibrin dissolution. Blood flow is computed in patient-specific arteries, with patient-specific blood clots acting as porous media whose porosity increases as the lysis takes place. We propose a local permeability model which allows us to deal with heterogeneous clot composition, as observed in patients. The permeability factor and lysis properties can be specified at each point of the clot, and yield a realistic global behavior, even during the lysis regime.
In-silico Modelling of the Impact of Ischaemic Stroke on the Human Brain
Description
The rising clinical incidence of cerebrovascular diseases, such as stroke and dementia, mean that understanding cerebral blood flow and its control in both healthy and diseased states has become an important clinical problem. However this is made highly challenging by the spatial and temporal limitations of human brain imaging and the number of blood vessels involved in a what is a highly complex dynamic regulating system over many length scales. In this talk I will discuss the work that we have done to construct models of both cerebral blood flow and oxygen transport that cover the many length scales involved, running from a few micrometres to a few millimetres. I will show how we have developed such models from the bottom up, through highly detailed modelling of the microcirculation and the application of homogenisation and optimisation methods, in order to generate a full-brain model of blood flow and oxygen transport. I will then focus on how this can be applied (and validated) in the context of ischaemic stroke to simulate the impact of both ischaemic stroke and its treatment on the human brain.
A Low Dimensional Surrogate Model of High-Fidelity Simulations of Intra-Arterial Thrombectomy
Description
The technique of surrogate modelling consists in reproducing the behavior of a physical system by finding a mapping between a set of input and output data produced by the system. In this work, the surrogate modelling technique was applied to reproduce the behavior of high-fidelity simulations of the clinical procedure of intra-arterial thrombectomy (IAT), the treatment to acute ischemic stroke (AIS), a pathology where a blood clot, or thrombus, obstructs a cerebral artery preventing the perfusion of cerebral tissues. IAT consists in mechanically removing the thrombus by means of a stent-retriever positioned at the occlusion site through a micro-catheter. The IAT procedure fails if the thrombus escapes from the stent inside the vessel or if the stress and strain state in the thrombus causes its fragmentation, leading to embolization. High-fidelity simulations of IAT procedures allow to predict the outcome and to evaluate the stress and strain state of the thrombus, but they are computationally demanding. The aim of this work is to develop a surrogate model of high-fidelity in-silico IAT able to predict the evolution of strain in the thrombus during the procedure
