Cells such as myocytes and adventitial fibroblasts are responsive to mechanical cues in their local environment. In response to mechanical loads, a variety of mechano-transduction mechanisms and signaling pathways are activated to regulate their response to the altered conditions.

In order to define mechano-signaling networks and their role in cellular function and remodeling, we have adapted and refined previously published systems models of myocyte hypertrophy. Using uncertainty quantification, we first found that the model accuracy was robust to parameter changes over a wide range with model outputs being least sensitive to time constants and most affected by uncertainty in reaction weights. We also found epistemic uncertainty in the reaction logic of the model could greatly affect model accuracy while uncertainty in the validation data had a modest effect on model accuracy.

As a step forward toward understanding myocyte response to external loading, including direction-dependent pathways, we extended this previous network model to include the transcriptional regulatory networks controlling gene expression as well as protein translation, and introduce a mass-action method to model quantitative gene expression. By incorporating RNA-sequencing data, this new approach displayed high accuracy with 69% agreement overall and 72% agreement for predicted differentially expressed genes in response to longitudinal stretch. We further found that the difference between transverse and longitudinal stretch responses in cardiomyocytes could be related to the sensitivity of directional mechanotransduction, with the sensitivity of longitudinal stretch being greater than transverse. Upon analyzing genes regulated by multiple TFs, we found that expression of these genes didn’t monotonically change with the number of TFs, which indicates TF regulation effects may saturate faster when multiple TFs coregulate gene expression. Moreover, we identified AT1 and ET1 receptors as main regulators of the stretch induced responses through receptor inhibition simulations and subsequent experiments.

A similar approach was used to study mechanical signaling and remodeling responses in PAAFs. In the current work, we have modified an existing systems model of cardiac fibroblast signaling to PAAFs and the cellular regulation of profibrotic signaling by combining both in-vitro and in-silico models of cell signaling in response to altered mechanical conditions. A UQ analysis on this model highlighted parameters to be optimized and network modules to be elucidated with more experiments. The signaling model in PAAFs and the subsequent experiments identified that both stretch and increased substrate stiffness regulated profibrotic genes, while no interaction effect was found between stretch and stiffness for several key genes studied. In addition, the activation of fibronectin expression by stretch in PAAFs may be angiotensin-independent when the cells are adhered on stiff but not soft substrates.

While these signaling network models can help distinguish regulators and their sensitivity to different mechanical stimuli, it is not known how these regulators participate in gene regulation of in-vivo hypertrophy. In the future, these signaling network models can be used to identify key regulators of hypertrophy-related heart failure and tissue fibrosis and provide support for drug discovery.

Modeling the hearts of patient with hypoplastic left heart syndrome (HLHS) has remained an ongoing challenge due to its unique single ventricular morphology. Severe distortions have been observed when using generic biventricular shape templates to create three-dimensional models of HLHS anatomy, which can lead to inaccurate or unconverged numerical analysis. Thus, in this work, we developed a HLHS-specific biventricular template to capture single ventricle morphology in this cohort more accurately and robustly. We applied the HLHS-specific template to reconstruct HLHS patient-specific models from cardiac magnetic resonance images. Models created using the HLHS-specific template resulted in significantly higher accuracy compared with using the generic template to model the same cohort of patients. The models created using the cohort-specific template were also qualitatively more anatomically realistic with the size of left ventricle and relative location of the apex well captured. When patient-specific models are used for finite element analysis of biomechanics or electrophysiology, it is important to ensure that all elements in the mesh are within an acceptable level of distortion. Element distortion metrics, element edge angles and aspect ratios, confirmed that element distortion in meshes created using the HLHS-specific template is comparable with that of biventricular anatomic models previously used in finite element analysis. A semi-automated workflow, developed to use HLHS-specific template to create patient-individual models from manually segmented contours, will allow future users to create 3D models consistently and efficiently. Accurate 3D patient-specific models will serve as useful tools for statistical shape and physiological analysis to provide insights into new associations between ventricular morphology and clinical outcomes.

Cardiac resynchronization therapy (CRT) is an effective treatment for dyssynchronous heart failure. Most patients perform better during clinical tests of cardiac function and may even have their hearts reverse remodel to a more normal state. As many as 30-40% of patients, however, do not respond. A focus of current research is identifying which patients will respond to CRT, though current clinical indications for CRT are based on dyssynchrony and heart failure and not on validated predictors of CRT response. In this thesis, five patient-specific computational models based on non-invasive data were developed and compared with eight previous patient-specific models that were based on more detailed and invasive clinical information to test whether the dyssynchrony metrics from the original study were still predictive in the new group. Model properties such as the volume fraction of negative work (VFNW) and coefficient of variation of work (COVW), that had correlated with patient outcomes in the original cohort, did not correlate with reverse remodeling as measured by reduction in end-systolic volume in the new group. The sensitivity analysis showed that these quantities were sensitive to the parameters of ventricular filling mechanics that could not be included in patient-specific models based on non-invasive data. Non-invasive estimates of filling parameters are available, but their reliability has been questioned. We conclude that it is likely to be necessary to obtain invasive measurements of diastolic pressures for patient-specific models to predict CRT outcomes. However, these measurements are available by cardiac catheterization, which could be justified for CRT patients.

Marker-based motion analysis is used to evaluate human movement and identify biomechanical features related to injury risk. However, these studies require extensive expertise, are labor intensive, time consuming, and high cost, making them impractical for routine use. Open-source smartphone-based motion capture systems such as OpenCap offer scalable, low cost, and automated alternatives to conventional motion capture systems. The goal of this study was to compare OpenCap to gold standard clinical marker-based motion capture for movement screening tasks. Sixty-two retroreflective markers were placed on ten healthy collegiate female athletes who completed a set of tasks commonly used to assess movement quality. A musculoskeletal model was used to estimate hip adduction, hip flexion, hip rotation, knee flexion, and ankle flexion with both OpenCap and marker-based inverse kinematics. OpenCap was approximately 12 times faster for data collection and processing, with only 2% of trials deemed unusable. RMSEs ranged from 3.5 to 11.3° across all joints and movement tasks. Pearson’s correlation coefficients for peak joint angles for select movements ranged from moderate to very strong for hip, knee, and ankle flexion and weak to very strong for hip adduction and hip rotation. Bland Altman plots showed varying trends, bias, and limits of agreement depending on the movement and joint angle. This suggests that smartphone markerless motion capture offers a time and cost-effective alternative for capturing movement and should be further evaluated for suitability to specific clinical questions. Moreover, this platform can be further developed to increase accuracy with enhanced training of algorithms and become a standard tool for routine, high-throughput movement screening in athletes.

Tetralogy of Fallot (TOF) is the most common cyanotic congenital heart disease accounting for about 10% of all congenital cardiac malformations. Due to improvements in surgical technique, individuals with repaired TOF (rTOF) are surviving into adulthood but face residual pulmonary regurgitation that can lead to adverse ventricular remodeling and, ultimately, heart failure. Cardiovascular magnetic resonance (CMR) imaging is the gold standard for evaluation in patients with rTOF, but the wealth of information available in CMR images is under-utilized. We sought to use computational tools to extract atlas-based biomarkers of regional biventricular shape and mechanics from standard of care CMR images to better aid clinical management of rTOF. Specifically, the aims of this dissertation were to 1) rapidly build biventricular shape models from CMR data using machine learning; 2) formulate shape atlases from these models to identify clinically meaningful shape variation within the population; and 3) construct biventricular biomechanics models to quantify shape marker relationships with differences in myocardial mechanical properties. Herein we demonstrate a fully automated, end-to-end pipeline that can robustly create biventricular shape models for the challenging anatomies present in rTOF. We also discovered specific markers of biventricular shape that are better discriminators of clinical prognosis than standard imaging indices. These markers of biventricular shape were also partial determinants of systolic function and may also be surrogate measures for altered myocardial contractility. Overall, patients with rTOF may benefit from routine atlas-based analyses of biventricular shape and mechanics that can supplement clinical decision making and provide insight into mechanisms underlying myocardial pathophysiology.

The lesser Egyptian jerboa, Jaculus jaculus, is a small bipedal rodent with unique morphological features such as disproportionately long hindlimbs and tail, fused metatarsal bones, and the loss of medial and lateral digits. These contribute to an extraordinary repertoire of locomotion including high accelerations and decelerations resulting in unpredictable ricochetal motion. In addition to speed, jerboa are capable of producing immense ground reaction forces, allowing for propulsion of their bodies forward and upward over ten times their hip height. This unmatched performance combined with their unique morphological characteristics separates them from other small mammals and invites great interest in the study of their biomechanics and movement. Investigating muscle interactions and how they ultimately result in whole body movement using in-vivo experimentation alone is not always practical. Therefore, implementing a detailed computational model can provide insights into the musculoskeletal dynamics of the jerboa. This study describes and evaluates the development of the first three-dimensional model of jerboa hindlimb biomechanics based on detailed anatomical measurement collected from micro computed tomography (microCT) scans. Joints are generated for all bones in the hindlimb following International Society of Biomechanics (ISB) standards, with segment mass properties of each geometry measured experimentally and calculated computationally. Tendon insertions and muscle lines of action were validated using microdissections and biomechanics experiments. A sensitivity analysis was conducted to evaluate the model's robustness to changes in input parameters and to identify muscle and joint parameters that are particularly important for locomotor performance. This model combined with measured kinematics, ground reactions forces, and contractile muscle properties can be used to better understand how anatomic and physiological adaptations in the jerboa have evolved to generate the joint moment arms and control mechanisms that give rise to extreme locomotor performance and stability.

The cardiovascular system is regulated through numerous control systems, which operate on various time scales. The baroreflex operates on the scale of seconds to minutes, responding to acute changes in pressure by altering heart rate, heart contractility, and blood vessel diameter. Blood volume control operates on a time scale of minutes to hours, in which the kidneys regulate blood pressure by means of blood plasma osmolarity and volume in the circulatory system. When pressure is altered on a longer time scale of days to years, growth and remodeling occur in the heart and blood vessel walls.

In this thesis, a multi-scale model was created to improve the capacity for study of the summed or individual effects of different time scales. This time scale consideration resulted in a more physiological approach to modeling the cardiovascular system than previous models. The time scales modeled were an acute seconds-to-minutes scale, during which the baroreflex acts, and a minutes-to-hours scale during which blood volume control occurs. The model was then optimized to allow for future coupling to computationally expensive growth and remodeling models. Arterial pressure regulation was demonstrated in both normal conditions and the pathophysiological condition of aortic valve disease. The work of this thesis successfully reproduced the local effects of the disease, and demonstrated the physiologic response on multiple time scales to the reduction in arterial pressure at an aortic stenosis.

Cardiac fibrosis is the excessive accumulation of extra-cellular matrix that is mainly

regulated by the activation of cardiac fibroblasts and their differentiation into myofibroblasts.

Mechanical forces are important regulators of cardiac fibroblast activation. However, it is not

clear which fibrotic signaling pathways are activated by the specific mechanical cues. Therefore,

we used in vitro stretch models as well as stiff and soft hydrogels to examine how mechanical

stretch and stiffness impacts the pathways involved in cardiac fibroblasts’ ability to generate

fibrotic phenotypes. Treatment of cardiac fibroblasts on plastic with transforming growth factor

β receptor I inhibitor resulted in lower mRNA expression levels for key myofibroblast gene

markers. Transforming growth factor β receptor I inhibitor also eliminated the stretch induced

upregulation of key fibrotic genes for fibroblasts on soft gels but only eliminated upregulation of

the smooth muscle α-actin gene on stiff gels. Surprisingly, inhibition of Rho kinase did not

impact expression levels of pro-fibrotic genes for cardiac fibroblasts on plastic and hydrogels.

Due to complications with the hydrogel stiffness, atomic force microscopy showed that

hydrogels can change their stiffness after fabrication depending on the environment the gel is in,

explaining the unusual cellular response we were experiencing with the fibroblasts on hydrogels.

Overall, transforming growth factor β signaling does significantly influence how cardiac

fibroblasts generate pro-fibrotic phenotypes. However due to the revelation of the changing

stiffness with the hydrogels, more work is needed to determine whether this pathway is more

involved in stretch responses or stiffness responses. More research is needed to determine

whether inhibiting transforming growth factor β signaling in cardiac fibroblasts can be used to

maintain a freshly isolated phenotype while culturing large quantities of cardiac fibroblasts for

longer periods on plastic tissue culture substrates.

Medical devices enable crucial insight into fundamental biological processes, development and disease. In this dissertation we present three medical devices for cellular discovery and diagnostics. In chapter I, we demonstrate a label- free image-encoded microfluidic cell sorter with a scanning Bessel beam. Microfluidics- based cell sorters offer a convenient, high information content, disposable solution that overcomes many of the disadvantages of conventional cell sorters. However, flow confinement in the microfluidic channel is generally one-dimensional via sheath flow. Consequently, the equilibrium cell distribution spreads outside the focal plane of commonly used Gaussian laser excitation beams, giving rise to a high number of blurred images that hinder subsequent cell sorting based on cell image features. To address this issue, we present a Bessel–Gaussian beam image-guided cell sorter with an ultra-long depth of focus, enabling focused images of >85% of passing cells. This system features label-free sorting capabilities based on features extracted from the output temporal waveform of a photomultiplier tube (PMT) detector. In chapter II, we demonstrate an automated, high throughput diagnostic device suited for high-traffic settings. Here we demonstrate a multiplex reverse-transcription loop-mediated isothermal amplification (RT-LAMP) coupled with a gold nanoparticle-based lateral flow immunoassay (LFIA) capable of detecting up to three unique viral gene targets in 15 min. RT-LAMP primers associated with three separate gene targets from the SARS-CoV-2 virus (Orf1ab, Envelope, and Nucleocapsid) were added to a one-pot mix. A colorimetric change from red to yellow occurs in the presence of a positive sample. Positive samples are run through a LFIA to achieve specificity on a multiplex three-test line paper assay. Positive results are indicated by a characteristic crimson line. The device is almost fully automated and is deployable in any community setting with a power source. In chapter III, we demonstrate a novel prototype and concept for cold plasma – based live cell lithography. A plasma jet is formed using a custom dual – lead nozzle, high voltage alternating current transformer, nitrogen feed gas, and frame. The concept aims to leverage the batch processing and spatial resolution advantages of lithography to identify, preserve, and isolate cells of interest while preserving their spatial information.