This interdisciplinary project developed a framework for implementing digital twins in European healthcare, focusing on secure, standardized data infrastructures and interoperable European Health Record (EHR) systems. It involved students from aerospace, strategic design, clinical medicine, and mechanical engineering. Analysis between Sweden and the Netherlands revealed Sweden excels in governance and trust, whereas the Netherlands is stronger in technical infrastructure. Overall adoption of digital twins is hindered by data privacy concerns and consent models. The public sector demonstrated the highest readiness, while patient groups showed lower willingness. The outcome includes an interactive dashboard, a readiness index, and an actionable implementation blueprint.

This study investigates the impact of shape-memory alloy (NiTiNol) parameter variability on the predictive reliability of finite element (FE) models for transcatheter aortic valve implantation (TAVI). A fully verified and validated FE model was developed following ASME V&V 40 guidelines, integrating experimental tests, numerical verification, and uncertainty quantification to isolate material effects from numerical artifacts. A Monte Carlo parameter fitting approach was employed using three independent tensile datasets to calibrate the NiTiNol model, resulting in a quantified uncertainty range of ±5.34%. Simulation results revealed that only Austenite Young’s modulus significantly influenced the mechanical response under physiological loading, while other transformation parameters had negligible effects. In contrast, literature-derived parameters introduced up to 30% deviation, highlighting the inadequacy of non-case-specific data. This work emphasizes the critical need for validated material parameters to ensure simulation credibility, particularly in clinical and regulatory contexts where simulation outcomes increasingly inform patient-specific treatment and device design.

This study presents a novel latent diffusion framework for generating clinically accurate virtual synthetic thrombi for use in in-silico clinical trials (ISCTs). The framework combines a variational autoencoder (VAE) with a diffusion model to generate three-dimensional representations of thrombi, including their shape, non-contrast computed tomography (NCCT) Hounsfield units, and computed tomography angiography (CTA) Hounsfield units. The model is trained on 1901 automatically segmented thrombi from the MR CLEAN registry. Additionally, we propose a method for incorporating thrombus composition characteristics into the virtual thrombi based on the relationship between NCCT imaging characteristics and red blood cell content, derived from in-vitro thrombus analogue histological analysis. The generated synthetic thrombi demonstrate clinically plausible characteristics across multiple dimensions, evaluated using key imaging metrics: thrombus attenuation increase (TAI) quartiles measuring perviousness, NCCT and CTA standard deviations quantifying heterogeneity, and median NCCT values indicating thrombus density. The synthetic samples show comparable distributions to the training data for these metrics, as well as for volume and surface area, while maintaining physiologically plausible ranges. The uniqueness of the samples is confirmed by low Intersection over Union (IoU) values between synthetic and real thrombi. The estimated histological compositions of synthetic thrombi align well with those observed in retrieved patient thrombi, though limitations exist in representing very low red blood cell content thrombi. This framework provides a foundation for generating realistic virtual thrombi that can be used to enhance ISCTs, particularly in the development and testing of new treatment options for acute ischemic stroke. Future work should focus on establishing connections between synthetic thrombi and vessels, and further exploring the relationship between imaging characteristics and thrombus composition through controlled in-vitro experiments.

Background: Ischemic stroke remains a major global health threat and a leading cause of mortality. Although advancements have been made in acute ischemic stroke treatment, challenges persist in achieving efficient recanalization and improved outcomes. Research on the mechanical characterization of ex vivo thrombi is limited. This paper aims to investigate the compositional and mechanical characterization of acute ischemic stroke thrombi retrieved in a per pass method and determine the relationship between composition and mechanical properties.
Methods: Thrombi fragments were collected per pass and segmented into samples. Larger fragments were sectioned into 1 mm pieces for cyclic unconfined compression testing, assessing hysteresis and tangent modulus as a measure of stiffness. The remaining pieces post-sectioning were combined into a single offcut sample, while smaller fragments were labeled as extra samples and kept for histology. Following mechanical testing, compositional analyses quantified the percentages of red blood cells, fibrin, and platelets in each sample.
Results: In 34 passes, a total of 108 thrombus fragments from 20 patients were successfully retrieved. Compositional heterogeneity was observed both between patients and within a single patient. Higher red blood cells (RBC) and lower fibrin/platelet contents were associated with Pass 1. The results showed a mean hysteresis area of 1.6 kPa and a mean hysteresis ratio of 37%. The median low strain stiffness was 1.76 kPa and the median high strain stiffness was 576.4 kPa. RBC and fibrin/platelets concentration had a linear relationship with high strain stiffness. Low strain stiffness, hysteresis area, and ratio changed with varying quantities of fibrin/platelets, RBC, and platelets composition.
Conclusion: This study enhances understanding of the relationship between thrombus composition and mechanical properties. The findings highlight the significant heterogeneity in thrombi and the non-linear viscoelastic behavior. Stiffness is highly corelated to strain levels, and the hysteresis amount depends on structural changes and varying compositional concentrations.

Stroke, caused by large vessel occlusions, is the second leading cause of death worldwide. Large vessel occlusions are the result of thrombosis, the undesired coagulation of blood within the vasculature. Due to this coagulation, a vessel can get blocked, prohibiting blood flow to areas distal from the occlusion, with all the serious consequences that may ensue. Since 2015, mechanical thrombectomy procedures have been widely accepted as a successful treatment technique to remove thrombi from the vasculature. However, complete reperfusion is only reached in 50% of acute ischemic stroke cases.

To successfully remove a thrombus from the vasculature, it is necessary to apply a specific retrieval force. This force must exceed the opposing forces, including the impaction force generated by the blood pressure gradient across the thrombus and the interaction forces between the thrombus and the vessel wall. Multiple studies have focused on tensile and compressive properties of thrombi. However, little is known about shear loading of the thrombus and about the interaction properties of the thrombus-vessel wall interface. Therefore, the aim of this study is to gain a better comprehension of thrombus biomechanics by studying the shear behavior of thrombi and the interaction of the thrombus-vessel wall interface in vitro.

In order to perform these in vitro studies, two custom-made test setups have been designed and developed. The friction test setup contains a plate of which the angle can be inclined slowly. By placing a thrombus-vessel wall sample on top of this plate, it is aimed to determine the static and kinetic coefficient of friction of the thrombus-vessel wall interface. Furthermore, it is aimed to determine the effect of time on this interaction. The thrombus-vessel wall interface was created by obtaining a piece of vein and blood from pigs. To study the shear behavior of thrombi a shear test setup has been developed. Two thrombi types have been utilized for this study, red blood cell- and fibrin-rich. Within the shear test setup it is possible to perform shear experiments under different normal loading conditions. A comprehensive analysis of the data acquired from experiments performed with both test setups has been conducted. Furthermore, a computational model has been developed to fit towards the experimental data obtained from the shear experiments.

The friction experiments suggest that time positively influences the bonds formed between a thrombus and the vessel wall as the coefficients of friction increase with an increased waiting time. Furthermore, a strong positive correlation was found between the static and kinetic coefficient of friction. This result was also found when doing an extensive analysis of the data obtained from the shear experiments. Additionally, the shear experiment showed that the thrombus composition influences its mechanical properties. Higher shear moduli and kinetic coefficients of friction were found for the fibrin samples, compared to the red blood cell samples.

The results obtained from the friction and shear experiments provide valuable insights into thrombus biomechanics. By extending the performed studies a better comprehension on thrombus mechanics and the thrombus-vessel wall interaction can be achieved.

Organ-on-chips (OoCs) are micro-fabricated cell culture platforms that mimic the function and structure of human organs. Limitations arose in the use of OoCs when they lacked sensing abilities and real-time monitoring. Throughout this thesis, an attempt is being made to address these complications by developing an \textit{in situ} glucose sensor for OoCs in collaboration with Bi/ond. The foundation of this proof-of-concept is an etalon, consisting of two reflective layers separated by a dielectric layer that allows light to enter and resonate between the reflective layers. The dielectric layer for this purpose will be microgels, with particular emphasis on the biomedical potential of the thermoresponsive polymer poly-(N-isopropylacrylamide) (pNIPAAm).
This sensor was developed on silicon and PDMS substrate using evaporation of Cr/Au, spin-coating pNIPAAm-co-10%-AAc and pNIPAAm-co-10%, and mid or post process modification with APBA. Plasma treatment was employed to create a monolithic layer, resulting in a successfully completed fabrication process verified through SEM. Reflectance spectroscopy confirmed the functionality and response, along with an effective generation of a calibration curve for glucose for both substrates and modification processes. The post process approach emerged as more preferable while exhibiting resilience to sterilization procedures that even improved its response to glucose due to removal of unbound APBA during sterilization. Post process samples were used to detect changes in glucose levels in cell medium that had been in contact with C2C12 mouse myoblast cells over 1-, 3-, and 7-day periods. The results showed that the concentration of glucose declined approximately 16 mg/dL after day 1, 130 mg/dL after day 3 and 227 mg/dL after day 7, in comparison to the control measurement. Additionally, there was optical confirmation of the non-toxic effects of the microgel-beads on cell culturing. The investigation of this proof-of-concept of glucose responsive microgel-based etalons was successful, and the next stage would involve integrating this sensor into a chip of Bi/ond.

Stroke, currently ranked as the second leading cause of mortality and the most prevalent source of enduring disability, imposes significant societal and economic consequences. These consequences encompass the necessity for extended care, rehabilitation, and the economic burdens associated with lost productivity. In the diagnostic process, physicians typically employ imaging techniques to assess cerebral blood vessels, ruling out alternative conditions when stroke is suspected. Subsequent to di agnosis, treatments like Intravenous Thrombolysis (IVT) and Endovascular Thrombectomy (EVT) are administered to either dissolve or retrieve thrombi. However, with EVT emerging as the new standard of stroke care and an increasing body of evidence indicating the substantial impact of thrombus me chanical properties on thrombectomy success, the use of clot analogues is gaining prominence as a means of facilitating efficient mechanical testing, both in terms of time and finances. By establishing a link between mechanical properties and image characteristic parameters, medical practitioners gain a better understanding of intracranial thrombi before embarking on surgical procedures. This knowledge can serve as a theoretical foundation for optimizing surgical instruments and devising effective surgical protocols in the future. In pursuit of these objectives, this study involved 29 patients and entailed several key steps: firstly, the extraction of imaging parameters from pre-EVT CT images; secondly, the creation of thrombus analogues from the patient’s blood; thirdly, the execution of five distinct mechanical tests to derive me chanical properties, encompassing cyclic compression, compression relaxation, tensile test to failure, tensile relaxation, and fracture testing. Additionally, the hyperelastic Yeoh model was utilized to simu late the behavior observed from cyclic compression and tensile to failure test. Prony series (n=2) was fitted to stress relaxation curves. In summary, this investigation revealed noteworthy observations pertaining to mechanical proper ties. Notably, hysteresis areas demonstrated robust and statistically significant correlations with com pression high-strain stiffness, low-strain stiffness, and tensile stiffness along with the ultimate tensile stress, implying that the thrombus analogue exhibits both hyperelastic and viscoelastic characteristics. In the domain of imaging parameters, the evident interdependence among the densities of the three regions of interest within both CTA and NCCT is apparent, respectively. However, it is essential to un derscore that the relationship between CTA and NCCT was not deemed statistically significant. Lastly, a noteworthy revelation was the strong negative monotonic correlation (ρ=-0.82, P=0.011) observed between the analogue contraction ratio of the thrombus analogue and the density of the middle part of NCCT in actual thrombi. This insight suggests that, in addition to thrombus composition, microstructure plays a pivotal role as well in determining their properties.