Session L09: Biofluids: Medical Devices II

Comparison of the closing kinematics of two designs of Mechanical Heart Valves (MHVs) against a Bioprosthetic Heart Valve

The Bileaflet Mechanical Heart Valves (BMHVs) are known to fully close later during the cardiac diastole compared to the natural/Bioprosthetic Heart Valve (BHVs), that fully close by the end of systole. Granted the widely reported relationship between the leaflet kinematics, hemodynamics and coagulant potential of prosthetic valves by the past research, new designs of MHVs are needed to replicate the physiological mechanics and function. This study compares the closure kinematics of a novel Trileaflet MHV (TMHV) against a BMHV and a BHV under similar numerical conditions. The flow solver employs the Curvilinear Immersed Boundary (CURVIB) method, strongly coupled to a Fluid-Structure Interaction (FSI) algorithm for the determination of the valves’ leaflet kinematics. The leaflets of the TMHV and BHV start to close during the forward deceleration phase of the cardiac cycle, and fully close by early diastole. Contrary to that, the BMHV leaflets do not begin their closing excursion until the onset of regurgitation. The hemodynamic reasons behind this behavior are explored in this study.    

Two-Step Uncertainty Quantification Methodology for Medical Device Design: Influence of Input Parameter Probability Distribution on Output Uncertainty

When representing an input parameter for uncertainty quantification (UQ), use of a Gaussian distribution is ubiquitous, yet for some input parameters other types of probability distributions may be more appropriate. In this study, we compare the effect of input probability distribution choice on output uncertainty as predicted by nondeterministic finite element simulations of an implantable nitinol inferior vena cava (IVC) filter. Computational modeling and simulation are commonly combined with experimental fatigue testing data to predict the fatigue resistance of such cardiovascular devices. For applications where model credibility is crucial, characterization of model uncertainty is required to understand failure and safety factors. In our UQ methodology, data on three categories of input parameter – material properties, geometry, and experimental conditions – is gathered and statistically characterized across multiple manufacturing lots. After a sensitivity analysis step to exclude uninfluential parameters, three probability distributions – Gaussian, gamma, and uniform – are fit to the remaining input parameter datasets. Each of the three probability distributions are sampled using Latin hypercube sampling which serves as inputs to nondeterministic FE simulations that predict local strains and strain amplitudes. In this presentation, we will present this broadly applicable UQ methodology as applied to a medical device.   

Evaluation of flow structures formed in a blood pump using Large Eddy Simulations

Extracorporeal membrane oxygenation (ECMO) is a last resort treatment for critically ill patients in need of lung and/or heart support. In ECMO, the patient's blood is pumped through a circuit composed of a blood pump, oxygenator, tubing, cannulae, and connectors exposing the blood components to highly unsteady flow fields. The formation of blood clots can be triggered by these local flow conditions in which areas characterized by high shear and prolonged residence time are particularly problematic. This study focuses on characterizing flow structures formed in the diagonal ECMO pump DP3 (Xenios Ag., Heilbronn, Germany). Large Eddy Simulations (LES) was used to assess pump performance during different operational conditions, identifying canonical flow structures formed within the pump. With an increased understanding of the flow structures developing in the pump, pump designs could be improved and potentially reduce complication risks and thus contribute to provide a gentler treatment to the patient.   

The FDA Generic Centrifugal Blood Pump Model as a Benchmark for CFD Validation

Computational fluid dynamics (CFD) is widely used in the design and assessment of medical devices. To be relied upon to inform regulatory decisions, however, CFD credibility must be established by performing verification and validation. Toward this end, FDA and academic collaborators established a benchmark centrifugal blood pump model for CFD validation. Interlaboratory validation experiments were conducted with the pump across a wide range of operating conditions to measure the pressure head and the velocity field using particle image velocimetry. Hemolysis experiments were also performed in a single laboratory to quantify the damage to red blood cells caused by the pump. Additionally, an open interlaboratory CFD study was conducted in which blinded participants from around the world submitted CFD simulation results of the pump for comparison with the experimental measurements. In this presentation, we provide an overview of the benchmark blood pump validation case and we summarize the results of the interlaboratory CFD study. The blood pump model and the experimental data are publicly available for use as a benchmark data set for CFD validation. This study also provides insight into the accuracy of CFD simulations of mechanical circulatory support devices from a wide range of users in the medical device community.   

Handling Phase Offset Errors in Processing 3-point Encoded 4D-Flow MRI Using Machine Learning

4D-Flow MRI is a non-invasive in vivo time resolved three dimensional blood flow velocity measuring technique. It requires 1 reference and 3 velocity encoded scans to estimate blood velocities. It is time consuming, has low spatio-temporal resolution, is impacted by acquisition noise, and artefacts resulting from velocity aliasing and phase offset errors. We present a novel physics informed deep learning framework called Input-Parameterized Physics Informed Neural Nets (IP-PINN) to address these limitations. IP-PINN parametrizes the output of a PINN with respect to a region of interest of fixed size within a 4D-Flow MRI image. It facilitates pre-training to increase the speed and accuracy of PINNs.  Velocities, pressure, transverse magnetization, and phase offset errors are modeled as deep neural nets. The continuous outputs are processed to mimic the 4D-Flow MRI acquisition process. The data fidelity term in the loss function is formulated in the complex Cartesian image space to naturally handle velocity aliasing and phase offset artefacts. Fluid flow physics are imposed using regularization. The method only uses data from the velocity encoded scans. Tests with synthetic 4D-Flow MRI derived from computational fluid dynamics simulations of aneurysmal flows demonstrate accurate high-resolution velocity estimation while attenuating noise and eliminating artefacts. The method also enables precise estimation of lumen boundaries through the transverse magnetization neural net.   

Sensitivity of platelet activation in an ECMO pump due to different modelling approaches

Extracorporeal membrane oxygenation (ECMO) is a life-saving therapy used in the critically ill to treat heart and/or lung failure. The circuit comprises a membrane lung with heat exchanger, a blood pump, cannulae for drainage and reinfusion of blood from/to the patient, and tubing with connectors. Its use is associated with complications, bleeding, thromboembolism, and hemolysis being primary causes of morbidity and mortality.

  Mechanical stresses play a role in platelet activation, assessed as platelet activation state (PAS). CFD can be a valuable tool to investigate effects of circuit component flow structures on thrombogenic properties in different scenarios. However, a proper assessment of the influence of boundary conditions and modelling approach is needed to characterize the sensitivity of the results due to modelling choices employed in the particle tracking. In this work, Lagrangian Particle Tracking (LPT) was used where 100’000 spherical particles, representing platelets, were injected in an ECMO blood pump to record stress history and compute their activation state. The LPT was carried out as a post-processing step on a time-resolved flow field produced using Star-CCM+. Blood was treated as a Newtonian fluid. Considering particle inertia, Newton’s 2nd law was used to compute particle velocity. In this work, drag, lift and pressure gradients were considered. The impact of these forces on platelet trajectories and activation state will be studied for different particle sizes and densities.   

Continuous, Accurate, and Robust Flowmeter for Liquid Samples

Acute Kidney Injury (AKI), a condition that can often begin with no observable clinical symptoms or signs, is a prevalent and costly issue, impacting as many as 20% of all hospital admissions. Its silent onset and lack of immediately effective diagnostic methods distinguish it from other critical conditions like heart attacks or strokes, making it a complex problem to manage. The unfortunate consequences of AKI result in approximately 2 million deaths per year, along with significantly elevated inpatient costs, amounting to roughly $42,000 per hospitalization. The biomarkers currently used to detect AKI include serum creatinine levels and urine excretion rates. Despite the crucial information provided by these markers, real-time measurement of serum creatinine is not yet feasible. However, the real-time monitoring of urine volume, which can offer significant insight into kidney function, is possible. In the current scientific and medical landscape, various state-of-the-art technologies are employed for measuring urine flow. This work delves into these existing methodologies while also introducing our novel device designed explicitly for urine flow measurement. The device we have developed demonstrates a high degree of accuracy. When comparing urine volume measurements obtained from our device with those obtained through gravimetric analysis (a standard method of measuring mass), we found a strong correlation, with a coefficient of 0.98. This suggests that our device provides an accurate and reliable method of measuring urine flow, potentially offering a valuable tool in the diagnosis and management of AKI.   

Detection of antimicrobial susceptibility on microfluidics sensor by electrochemical impedance sensing.

Antimicrobial resistance (AMR) is emerging as a global health threat; even common infections are becoming harder to treat with empirical antibiotics. Antimicrobial susceptibility testing (AST) ensures the current choice of antibiotics for treating specific infections. We explored the impedance characteristic of the bacterial cells at the cell-electrode interface for AST. Impedance changes over time when the bacterial cells grow/die in response to antibiotics. This change in impedance is correlated with the susceptibility profile of bacteria. We employed a microfluidic device with a screen-printed electrode to determine AST. Four carbon electrodes with alternate working and counter electrodes of 0.5 mm were screen printed on a glass slide. To fabricate a PDMS microfluidic device, 3D printed mould was used. The screen-printed electrode and the PDMS device were attached to plasma. The channel was functionalised with poly-L-Lysine to enhance bacterial attachment. Before bacterial attachment, 10% Nutrient Media (NM) impedance was measured. Bacterial suspension of 10⁸ CFU/mL was introduced, and unattached bacteria were washed off. Electrochemical impedance sensing (EIS) was performed with bacteria attached to the electrode of the microfluidic device. The device is then incubated for two hours without antibiotics and used as a positive control (PC). For the test device, we introduced an antibiotic (Ampicillin – 5µg/mL) and the 10% NM at t=0 hr and impedance was measured with respect to time. R_ct decreased in PC, implying bacterial growth, whereas in the test device, there was no change. In PC, R_ct difference can be attributed to the increased electrode coverage due to bacterial growth. The constant R_ct after incubation in the test device suggests the inhibition of bacterial growth in the presence of antibiotics. Using a low conductivity medium and selection of R_ct as a response signal enables the AST even with simple microfluidic geometry. The change in R_ct was related well to the bacterial growth profile. We performed the experiment with resistant bacteria and human samples and got the expected result. The proposed detection system will enable rapid AST detection and allows front-line healthcare workers to perform AST with minimal training.   

In-silico Hemodynamics Simulations to Investigate Stroke Outcomes in Patients After Left Ventricular Assist Device Implantation

Stroke is a leading cause of complications and death in advanced heart failure patients treated with implanted Left Ventricular Assist Device (LVAD). Hemodynamics significantly determines stroke risk in patients on LVAD support. However, quantitative assessment of flow variables and their relation to stroke outcomes post-LVAD implantation remains a major challenge. This inspired our in-silico study comprising patient-specific hemodynamics analyses in a set of 12 patients on LVAD support: 6 with reported stroke outcomes and 6 without. We compared hemodynamics using quantitative flow descriptors for helicity, vortex generation, and wall shear stress. Baseline flow pre-implantation was studied to analyze hemodynamic alterations from a pre-implant flow scenario that can potentially reveal hidden links to stroke outcomes during LVAD support. Finally, we analyzed differences in a Lagrangian particle transport analysis for embolism potential. Our analysis revealed key differences in flow descriptors, their ratios against baseline pre-implant flow, and Lagrangian descriptors, between the cases with and without stroke outcomes. Building upon these observations in future works can lead to significant advancements in understanding stroke etiology and pre-surgical assessment of stroke risks.   

Artificial microtubules for particle transport and delivery

Directed transport of microcargoes is essential for living organisms as well as for applications in microrobotics, nanotechnology and biomedicine. Existing delivery technologies often suffer from low speeds, limited navigation control and dispersal by cardiovascular flows. In cell biology, these issues are largely overcome by cytoskeletal motors that carry vesicles along microtubule highways. Thus inspired, here we developed an artificial microtubule (AMT), a structured microfibre with embedded micromagnets that serve as stepping stones to guide particles rapidly through flow networks. Compared with established techniques, the microcargo travels an order of magnitude faster using the same driving frequency, and dispersal is mitigated by a strong dynamic anchoring effect. Even against strong fluid flows, the large local magnetic-field gradients enable both anchoring and guided propulsion. Finally, we show that AMTs can facilitate the self-assembly of microparticles into active-matter clusters, which then enhance their walking speed by bridging over stepping stones collectively. Hence, we demonstrate a unique strategy for robust delivery inside microvascular networks and for minimally invasive interventions, with non-equilibrium effects that could be equally relevant for enhancing biological transport processes.