In silico trials for treatment of acute ischemic stroke: Design and implementation

Systems-level computational modeling in ischemic stroke: from cells to patients

Ischemic stroke, a significant threat to human life and health, refers to a class of conditions where brain tissue damage is induced following decreased cerebral blood flow. The incidence of ischemic stroke has been steadily increasing globally, and its disease mechanisms are highly complex and involve a multitude of biological mechanisms at various scales from genes all the way to the human body system that can affect the stroke onset, progression, treatment, and prognosis. To complement conventional experimental research methods, computational systems biology modeling can integrate and describe the pathogenic mechanisms of ischemic stroke across multiple biological scales and help identify emergent modulatory principles that drive disease progression and recovery. In addition, by running virtual experiments and trials in computers, these models can efficiently predict and evaluate outcomes of different treatment methods and thereby assist clinical decision-making. In this review, we summarize the current research and application of systems-level computational modeling in the field of ischemic stroke from the multiscale mechanism-based, physics-based and omics-based perspectives and discuss how modeling-driven research frameworks can deliver insights for future stroke research and drug development.

Challenges and Potential of Artificial Intelligence in Neuroradiology

PURPOSE

Artificial intelligence (AI) has emerged as a transformative force in medical research and is garnering increased attention in the public consciousness. This represents a critical time period in which medical researchers, healthcare providers, insurers, regulatory agencies, and patients are all developing and shaping their beliefs and policies regarding the use of AI in the healthcare sector. The successful deployment of AI will require support from all these groups. This commentary proposes that widespread support for medical AI must be driven by clear and transparent scientific reporting, beginning at the earliest stages of scientific research.

METHODS

A review of relevant guidelines and literature describing how scientific reporting plays a central role at key stages in the life cycle of an AI software product was conducted. To contextualize this principle within a specific medical domain, we discuss the current state of predictive tissue outcome modeling in acute ischemic stroke and the unique challenges presented therein.

RESULTS AND CONCLUSION

Translating AI methods from the research to the clinical domain is complicated by challenges related to model design and validation studies, medical product regulations, and healthcare providers' reservations regarding AI's efficacy and affordability. However, each of these limitations is also an opportunity for high-impact research that will help to accelerate the clinical adoption of state-of-the-art medical AI. In all cases, establishing and adhering to appropriate reporting standards is an important responsibility that is shared by all of the parties involved in the life cycle of a prospective AI software product.

Modelling midline shift and ventricle collapse in cerebral oedema following acute ischaemic stroke

In ischaemic stroke, a large reduction in blood supply can lead to the breakdown of the blood-brain barrier and to cerebral oedema after reperfusion therapy. The resulting fluid accumulation in the brain may contribute to a significant rise in intracranial pressure (ICP) and tissue deformation. Changes in the level of ICP are essential for clinical decision-making and therapeutic strategies. However, the measurement of ICP is constrained by clinical techniques and obtaining the exact values of the ICP has proven challenging. In this study, we propose the first computational model for the simulation of cerebral oedema following acute ischaemic stroke for the investigation of ICP and midline shift (MLS) relationship. The model consists of three components for the simulation of healthy blood flow, occluded blood flow and oedema, respectively. The healthy and occluded blood flow components are utilized to obtain oedema core geometry and then imported into the oedema model for the simulation of oedema growth. The simulation results of the model are compared with clinical data from 97 traumatic brain injury patients for the validation of major model parameters. Midline shift has been widely used for the diagnosis, clinical decision-making, and prognosis of oedema patients. Therefore, we focus on quantifying the relationship between ICP and midline shift (MLS) and identify the factors that can affect the ICP-MLS relationship. Three major factors are investigated, including the brain geometry, blood-brain barrier damage severity and the types of oedema (including rare types of oedema). Meanwhile, the two major types (stress and tension/compression) of mechanical brain damage are also presented and the differences in the stress, tension, and compression between the intraparenchymal and periventricular regions are discussed. This work helps to predict ICP precisely and therefore provides improved clinical guidance for the treatment of brain oedema.

Should individual timeline and serial CT/MRI panels of all patients be presented in acute brain insult cohorts? A pilot study of 45 patients with decompressive craniectomy after aneurysmal subarachnoid hemorrhage

PURPOSE

Our review of acute brain insult articles indicated that the patients' individual (i) timeline panels with the defined time points since the emergency call and (ii) serial brain CT/MRI slice panels through the neurointensive care until death or final brain tissue outcome at 12 months or later are not presented.

METHODS

We retrospectively constructed such panels for the 45 aneurysmal subarachnoid hemorrhage (aSAH) patients with a secondary decompressive craniectomy (DC) after the acute admission to neurointensive care at Kuopio University Hospital (KUH) from a defined population from 2005 to 2018. The patients were indicated by numbers (1.-45.) in the pseudonymized panels, tables, results, and discussion. The timelines contained up to ten defined time points on a logarithmic time axis until death ([Formula: see text]; 56%) or 3 years ([Formula: see text]; 44%). The brain CT/MRI panels contained a representative slice from the following time points: SAH diagnosis, after aneurysm closure, after DC, at about 12 months (20 survivors).

RESULTS

The timelines indicated re-bleeds and allowed to compare the times elapsed between any two time points, in terms of workflow swiftness. The serial CT/MRI slices illustrated the presence and course of intracerebral hemorrhage (ICH), perihematomal edema, intraventricular hemorrhage (IVH), hydrocephalus, delayed brain injury, and, in the 20 (44%) survivors, the brain tissue outcome.

CONCLUSIONS

The pseudonymized timeline panels and serial brain imaging panels, indicating the patients by numbers, allowed the presentation and comparison of individual clinical courses. An obvious application would be the quality control in acute or elective medicine for timely and equal access to clinical care.

MRI-based parameter inference for cerebral perfusion modelling in health and ischaemic stroke

Cerebral perfusion modelling is a promising tool to predict the impact of acute ischaemic stroke treatments on the spatial distribution of cerebral blood flow (CBF) in the human brain. To estimate treatment efficacy based on CBF, perfusion simulations need to become suitable for group-level investigations and thus account for physiological variability between individuals. However, computational perfusion modelling to date has been restricted to a few patient-specific cases. This study set out to establish automated parameter inference for perfusion modelling based on neuroimaging data and thus enable CBF simulations of groups. Magnetic resonance imaging (MRI) data from 75 healthy senior adults were utilised. Brain geometries were computed from healthy reference subjects' T1-weighted MRI. Haemodynamic model parameters were determined from spatial CBF maps measured by arterial spin labelling (ASL) perfusion MRI. Thereafter, perfusion simulations were conducted in 75 healthy cases followed by 150 acute ischaemic stroke cases representing an occlusion and CBF cessation in the left and right middle cerebral arteries. The anatomical fitness of the brain geometries was evaluated by comparing the simulated grey (GM) and white matter (WM) volumes to measurements in healthy reference subjects. Strong positive correlations were found in both tissue types (GM: Pearson's r 0.74, P<0.001; WM: Pearson's r 0.84, P<0.001). Haemodynamic parameter tuning was verified by comparing the total volumetric blood flow rate to the brain in healthy reference subjects and simulations (Pearson's r 0.89, P<0.001). In acute ischaemic stroke cases, the simulated infarct volume using a perfusion-based estimate was 197±25 ml. Computational predictions were in agreement with anatomical and haemodynamic values from the literature concerning T1-weighted, T2-weighted, and phase-contrast MRI measurements in healthy scenarios and acute ischaemic stroke cases. The acute stroke simulations did not capture small infarcts (left tail of the distribution), which could be explained by neglected compensatory mechanisms, e.g. collaterals. The proposed parameter inference method provides a foundation for group-level CBF simulations and for in silico clinical stroke trials which could assist in medical device and drug development.

In-silico enhanced animal study of pulmonary artery pressure sensors: assessing hemodynamics using computational fluid dynamics

To assess whether in-silico models can be used to predict the risk of thrombus formation in pulmonary artery pressure sensors (PAPS), a chronic animal study using pigs was conducted. Computed tomography (CT) data was acquired before and immediately after implantation, as well as one and three months after the implantation. Devices were implanted into 10 pigs, each one in the left and right pulmonary artery (PA), to reduce the required number of animal experiments. The implantation procedure aimed at facilitating optimal and non-optimal positioning of the devices to increase chances of thrombus formation. Eight devices were positioned non-optimally. Three devices were positioned in the main PA instead of the left and right PA. Pre-interventional PA geometries were reconstructed from the respective CT images, and the devices were virtually implanted at the exact sites and orientations indicated by the follow-up CT after one month. Transient intra-arterial hemodynamics were calculated using computational fluid dynamics. Volume flow rates were modelled specifically matching the animals body weights. Wall shear stresses (WSS) and oscillatory shear indices (OSI) before and after device implantation were compared. Simulations revealed no relevant changes in any investigated hemodynamic parameters due to device implantation. Even in cases, where devices were implanted in a non-optimal manner, no marked differences in hemodynamic parameters compared to devices implanted in an optimal position were found. Before implantation time and surface-averaged WSS was 2.35 ± 0.47  Pa, whereas OSI was 0.08 ± 0.17 , respectively. Areas affected by low WSS magnitudes were 2.5 ± 2.7  cm2 , whereas the areas affected by high OSI were 18.1 ± 6.3  cm2 . After device implantation, WSS and OSI were 2.45 ± 0.49  Pa and 0.08 ± 0.16 , respectively. Surface areas affected by low WSS and high OSI were 2.9 ± 2.7  cm2 , and 18.4 ± 6.1  cm2 , respectively. This in-silico study indicates that no clinically relevant differences in intra-arterial hemodynamics are occurring after device implantation, even at non-optimal positioning of the sensor. Simultaneously, no embolic events were observed, suggesting that the risk for thrombus formation after device implantation is low and independent of the sensor position.

In silico thrombectomy trials for acute ischemic stroke

BACKGROUND AND OBJECTIVE

In silico trials aim to speed up the introduction of new devices in clinical practice by testing device design and performance in different patient scenarios and improving patient stratification for optimizing clinical trials. In this paper, we demonstrate an in silico trial framework for thrombectomy treatment of acute ischemic stroke and apply this framework to compare treatment outcomes in different subpopulations and with different thrombectomy stent-retriever devices. We employ a novel surrogate thrombectomy model to evaluate the thrombectomy success in the in silico trial.

METHODS

The surrogate thrombectomy model, built using data from a fine-grained finite-element model, is a device-specific binary classifier (logistic regression), to estimate the probability of successful recanalization, the outcome of interest. We incorporate this surrogate model within our previously developed in silico trial framework and demonstrate its use with three examples of in silico clinical trials. The first trial is a validation trial for the surrogate thrombectomy model. We then present two exploratory trials: one evaluating the performance of a commercially available device based on the fibrin composition in the occluding thrombus and one comparing the performance of two commercially available stent retrievers.

RESULTS

The Validation Trial showed the surrogate thrombectomy model was able to reproduce a similar recanalization rate as the real-life MR CLEAN trial (p=0.6). Results from the first exploratory trial showed that the chance of successful thrombectomy increases with higher blood cell concentrations in the thrombi, which is in line with observations from clinical data. The second exploratory trial showed improved recanalization success with a newer stent retriever device; however, these results require further investigation as the surrogate model for the newer stent retriever device has not yet been validated.

CONCLUSIONS

In this novel study, we have shown that in silico trials have the potential to help inform medical device developers on the performance of a new device and may also be used to select populations of interest for a clinical trial. This would reduce the time and costs involved in device development and traditional clinical trials.

Computational modelling of cerebral oedema and osmotherapy following ischaemic stroke

In ischaemic stroke, a large reduction in blood supply can lead to the breakdown of the blood brain barrier and to cerebral oedema after reperfusion therapy. Cerebral oedema is marked by elevated intracranial pressure (ICP), tissue herniation and reduced cerebral perfusion pressure. In clinical settings, osmotherapy has been a common practice to decrease ICP. However, there are no guidelines on the choice of administration protocol parameters such as injection doses, infusion time and retention time. Most importantly, the effects of osmotherapy have been proven controversial since the infusion of osmotic agents can lead to a range of side effects. Here, a new Finite Element model of brain oedema and osmotherapy is thus proposed to predict treatment outcome. The model consists of three components that simulate blood perfusion, oedema, and osmotherapy, respectively. In the perfusion model (comprising arteriolar, venous, and capillary blood compartments), an anatomically accurate brain geometry is used to identify regions with a perfusion reduction and potential oedema occurrence in stroke. The oedema model is then used to predict ICP using a porous circulation model with four fluid compartments (arteriolar blood, venular blood, capillary blood, and interstitial fluid). In the osmotherapy model, the osmotic pressure is varied and the changes in ICP during different osmotherapy episodes are quantified. The simulation results of the model show excellent agreement with available clinical data and the model is employed to study osmotherapy under various parameters. Consequently, it is demonstrated how therapeutic strategies can be proposed for patients with different pathological parameters based on simulations.

Quantification of hypoxic regions distant from occlusions in cerebral penetrating arteriole trees

The microvasculature plays a key role in oxygen transport in the mammalian brain. Despite the close coupling between cerebral vascular geometry and local oxygen demand, recent experiments have reported that microvascular occlusions can lead to unexpected distant tissue hypoxia and infarction. To better understand the spatial correlation between the hypoxic regions and the occlusion sites, we used both in vivo experiments and in silico simulations to investigate the effects of occlusions in cerebral penetrating arteriole trees on tissue hypoxia. In a rat model of microembolisation, 25 μm microspheres were injected through the carotid artery to occlude penetrating arterioles. In representative models of human cortical columns, the penetrating arterioles were occluded by simulating the transport of microspheres of the same size and the oxygen transport was simulated using a Green’s function method. The locations of microspheres and hypoxic regions were segmented, and two novel distance analyses were implemented to study their spatial correlation. The distant hypoxic regions were found to be present in both experiments and simulations, and mainly due to the hypoperfusion in the region downstream of the occlusion site. Furthermore, a reasonable agreement for the spatial correlation between hypoxic regions and occlusion sites is shown between experiments and simulations, which indicates the good applicability of in silico models in understanding the response of cerebral blood flow and oxygen transport to microemboli.

The brain function depends on the continuous oxygen supply through the bloodstream inside the microvasculature. Occlusions in the microvascular network will disturb the oxygen delivery in the brain and result in hypoxic tissues that can lead to infarction and cognitive dysfunction. To aid in understanding the formation of hypoxic tissues caused by micro-occlusions in the penetrating arteriole trees, we use rodent experiments and simulations of human vascular networks to study the spatial correlations between the hypoxic regions and the occlusion locations. Our results suggest that hypoxic regions can form distally from the occlusion site, which agrees with the previous observations in the rat brain. These distant hypoxic regions are primarily due to the lack of blood flow in the brain tissues downstream of the occlusion. Moreover, a reasonable agreement of the spatial relationship is found between the experiments and the simulations, which indicates the applicability of in silico models to study the effects of microemboli on the brain tissue.

A Review of the Advancements in the in-vitro Modelling of Acute Ischemic Stroke and Its Treatment

In-vitro neurovascular models of large vessel occlusions (LVOs) causing acute ischemic stroke (AIS) are used extensively for pre-clinical testing of new treatment devices. They enable physicians and engineers to examine device performance and the response of the occlusion to further advance design solutions for current unmet clinical needs. These models also enable physicians to train on basic skills, to try out new devices and new procedural approaches, and for the stroke team to practice workflows together in the comfort of a controlled environment in a non-clinical setting. Removal of the occlusive clot in its entirety is the primary goal of the endovascular treatment of LVOs via mechanical thrombectomy (MT) and the medical treatment via thrombolysis. In MT, recanalization after just one pass is associated with better clinical outcomes than procedures that take multiple passes to achieve the same level of recanalization, commonly known as first pass effect (FPE). To achieve this, physicians and engineers are continually investigating new devices and treatment approaches. To distinguish between treatment devices in the pre-clinical setting, test models must also be optimized and expanded become more nuanced and to represent challenging patient cohorts that could be improved through new technology or better techniques. The aim of this paper is to provide a perspective review of the recent advancements in the in-vitro modeling of stroke and to outline how these models need to advance further in future. This review provides an overview of the various in-vitro models used for the modeling of AIS and compares the advantages and limitations of each. In-vitro models remain an extremely useful tool in the evaluation and design of treatment devices, and great strides have been made to improve replication of physiological conditions. However, further advancement is still required to represent the expanding indications for thrombectomy and thrombolysis, and the generation of new thrombectomy devices, to ensure that smaller treatment effects are captured.