71446 research outputs found
Sort by
Lower Limb Locomotor Sensing and Estimation for Athletic Performance and Gait Rehabilitation
No full textMobility is a cornerstone of independence and overall health, yet it is often compromised in both clinical and athletic populations. This dissertation investigates how wearable sensors and data-driven modeling can estimate gait metrics and analyze gait patterns across two distinct use cases: recreational running and post-stroke walking with exoskeleton assistance. In both contexts, the goal is to extend biomechanical assessment beyond the laboratory and support personalized gait monitoring in real-world environments.
Inertial measurement units (IMUs) offer a wearable alternative to traditional motion capture systems, enabling kinematic measurements outside of controlled laboratory settings. When paired with machine learning and ground truth force plate data, data from IMUs can estimate kinetic metrics, including joint torques and ground reaction forces (GRFs). Throughout this dissertation, IMUs serve as the primary sensing modality for estimating gait metrics during both running and post-stroke walking across varied environments.
We first investigated the feasibility of using data from IMUs to measure overstriding during running, i.e. the horizontal distance between the foot and the body’s center of mass at foot strike. Overstriding is associated with fatigue and running-related injury, and real-world measurement could inform training or technique adjustments. Linear models using sagittal segment angles from 10 recreational runners accurately estimated overstriding during treadmill and overground running and explained over 80% of the variance in peak braking force, demonstrating potential to monitor landing and loading mechanics.
Extending this work, we developed machine learning models to estimate kinetic metrics during running. A model trained on treadmill IMU and GRF data from 15 runners estimated braking and propulsion forces during overground running, achieving a root mean squared error (RMSE) of 4.3% bodyweight (%BW), which improved to 2.6%BW with individual fine-tuning using eight strides of overground data. With this method, we demonstrated the transferability of treadmill-trained models to predict braking and propulsion forces during overground running and highlighted the benefit of model individualization. When extrapolated to outdoor track running, the fine-tuned model better captured expected relationships between impulse and speed than the generalized model. These results provide insights into the accuracy and generalizability of kinetic metric estimation from treadmill-based, IMU data-driven models during overground running in and out of the laboratory.
Applying this approach in a clinical context, we estimated gait metrics and identified adaptation to exoskeleton assistance post-stroke during human-in-the-loop optimization (HILO) in a community setting. Treadmill-trained models achieved RMSEs of 0.98° for trailing limb angle, 0.076 Nm/kg for peak ankle torque, and 1.2%BW for peak propulsion during overground walking. Results revealed heterogeneous responses to exoskeleton assistance during HILO: some participants increased walking speed with greater paretic limb engagement, while others showed limited gait changes or benefit from assistance. These findings underscore the importance of understanding individual gait adaptation patterns to personalize and optimize rehabilitation.
This dissertation concludes with a review of lower limb exoskeletons for locomotor assistance, emphasizing key research priorities for advancing gait rehabilitation. Developing accessible tools for gait analysis and providing personalized feedback or assistance through wearable technologies will be essential for effective gait training and performance evaluation. Looking ahead, the integration of scalable wearable sensing with robust estimation models holds promise for transforming how mobility is monitored and enhanced across clinical, athletic, and everyday mobility use cases.Engineering and Applied Sciences - Engineering Science
The House of Panic
No full textIn this work, the author examines — or at least tries his very hardest to examine — the modern condition through a series of events that will remain, for reasons of both discretion and narrative suspense, unspecified. Through a careful balance of storytelling, armchair psychology, elementary mathematics, sociological speculation, and the application of particle physics on the blockchain, the project attempts to locate meaning in places it almost certainly isn’t. The resulting text aspires to say something profound about contemporary life but will settle, if necessary, for merely sounding like it almost does. The embargoed contents herein remain classified until further notice, or until someone asks nicely.Extension Studie
Enumeration in stochastic processes and polyhedral geometry
Get PDFThis dissertation explores the combinatorics of Markov chains and polyhedral geometry, with a focus on the asymmetric simple exclusion process (ASEP) and the Ehrhart theory of polytopes.
The first part addresses the stationary distribution of stochastic models, including the open ASEP, the Arndt-Heinzel-Rittenberg (AHR) model and the doubly ASEP (DASEP).
We give a two-layer simple random walk interpretation for the open ASEP model, a tableaux formula for the AHR model, and show that the DASEP exhibits homomesy phenomenon.
The second part of the dissertation studies the Ehrhart theory of positroid polytopes and alcoved polytopes.
We present combinatorial formulas for the -polynomials of positroid polytopes and alcoved polytopes.
We also prove a connection between our shelling formula with decorated ordered set partitions.Mathematic
Leveraging Samples with Diverse Ancestries to Better Understand the Genetic and Molecular Architecture of Complex Traits and Improve Polygenic Risk Scores
Get PDFLarge-scale genome-wide association studies (GWAS) have identified numerous genetic variants associated with human complex traits and diseases. However, the vast majority of these studies have been conducted in individuals of European ancestry, which has led to missed opportunities for biological discovery in non-European ancestry populations. Moreover, this imbalance could result in unequal benefits of precision medicine, as polygenic risk scores (PRS) based on large-scale genetic studies in European populations have high predictive power of clinical outcomes in European samples but poor predictive power in non-European samples.
A key insight from genetic discovery is the widespread pleiotropy of common variants associated with disease, which has been underutilized in elucidating distinct genetic mechanisms underlying multiple diseases. A fundamental challenge in genetics is connecting disease-associated variation to biological mechanisms, as individual variants exert modest effects and often have unclear functional consequences. While aggregated associations across a disease can provide insights, the presence of numerous independent mechanisms has limited the interpretability of gene enrichment strategies.
In Chapter 1, I conduct GWAS for 36 quantitative traits in a large Korean cohort, a major biobank effort that broadens the population diversity of genetic studies in East Asia. I identify 301 novel genetic loci, compare the genetic architecture across East Asian and European ancestry populations, and pinpoint novel causal variants through statistical fine-mapping.
In Chapter 2, I develop multi-ancestry PRS for venous thromboembolism (VTE) using European and African ancestry samples. By evaluating the prediction performance of these models across diverse ancestry groups, I demonstrate that multi-ancestry PRS for VTE outperform population-specific PRS, particularly in African ancestry populations with smaller GWAS sample sizes.
In Chapter 3, I investigate genetic drivers of heterogeneity in thyroid cancer pathophysiology using a multi-ancestry GWAS meta-analysis. By leveraging cross-trait associations of 66 independent thyroid cancer-associated variants across 151 phenotypes, I identify five distinct mechanistic clusters of thyroid cancer, each representing biologically meaningful pathways and robust associations with disease outcomes in an independent dataset.Population Health Science
Ending the Stalemate: Iran, the United States, and the Use of Applied History to Facilitate Changes in Diplomacy
Get PDFThis thesis explores the complex and adversarial relationship between the U.S.
and Iran since 1979, including the impacts of sanctions, Iran's pursuit of nuclear
weapons, regime change efforts, and alliances aimed at cooperation in the Middle East.
Examining historical parallels highlights how applied history can inform diplomatic
strategies. Sanctions, the primary U.S. approach toward Iran, have largely failed to
influence Iranian policies and have caused significant negative effects. Historical
examples, such as sanctions against Fascist Italy, reveal ineffectiveness in changing
aggressive behavior. I suggest the Peace of Westphalia from 1648 as a model for creating
a balanced Middle Eastern order involving major powers to reduce sectarian tensions and
promote cooperation. Iran's nuclear program is crucial for its national security as a
deterrent, and although Iran has not yet weaponized its capabilities, it is nearing threshold
status. Therefore, I propose that Iran could use nuclear weapons to negotiate for sanctions
relief, similar to South Africa's experience with its nuclear program. I also argue that
political change in Iran should not rely on external military intervention, as past U.S.
efforts in Iraq, Libya, and Afghanistan led to prolonged instability. The recommendation
is that the U.S. avoid consideration of regime change as a strategy and focus on
diplomatic engagement, learning from past mistakes. Analyzing historical cases reveals
that a diplomatic approach with realistic negotiations offers the best path to regional
stability, encouraging cooperation and mutual recognition to foster peace in the Middle
East. Both nations should study history to avoid repeating errors, as understanding past
conflicts helps clarify regional and ideological dynamics.Extension Studie
Robust Inference in Financial Markets
No full textThis thesis studies thick tails, found across financial markets, which present challenges for modern financial theory. Notably, traditional tools in financial econometrics deteriorate in the presence of thick-tailed predictors. The standard Koenker-Basset estimator of linear quantile regression is an example of a model with an unbounded influence function and is thus vulnerable to those predictors. To address thick tails, this thesis proposes a robust estimator of the linear quantile that is computable using existing software and has a bounded influence function, making it less vulnerable to non-Gaussian conditions frequently found in financial markets. Statistical properties of this estimator and empirical results when this estimator is used on market data are also described. Bounded influence functions are useful outside of purely financial econometric applications. This thesis connects the robust properties of the estimator developed to differential privacy, a mathematical privacy tool that protects individuals from identification under data analysis, and proposes differentially private and pseudo-private estimators. This estimator works well in comparison with existing private and pseudo-private regression methods, reducing noise under the same privacy budget constraints. Specifically, this paper presents three separate differentially private algorithms, and shows their utility in simulated experiments and real world data. In addition, this paper presents a pseudo-private algorithm for use in cases with small sample size where differential privacy is not feasible. Using the motivating example of Opportunity Atlas data, the estimator decreases noise significantly.Applied Mathematic
Several problems in extremal combinatorics
Get PDFThis thesis presents three distinct contributions to extremal combinatorics. First, it
resolves a conjecture by Bollobás, Brightwell, and Leader on the prevalence of unate k-SAT
functions, proving that for all fixed k≥2, almost all k-SAT functions on n variables are
unate (i.e., monotone after negating certain variables).
Second, it addresses a question posed by Sapozhenko on the enumeration of q-ary t-error
correcting codes. Improving upon previous bounds, it demonstrates that for a broad range
of parameters, the number of such codes is tightly bounded by the Hamming bound.
Finally, it investigates the commonness and uncommonness of systems of linear equations
over finite fields. Answering questions of Kamčev, Liebenau, and Morrison, it shows that all
2*k linear systems with k even and girth k-1 are uncommon over sufficiently large finite
fields, and provides a near-complete classification of common 2*5 linear systems.
This thesis provides new insights into the structure of various combinatorial objects and
contributes to the broader understanding of extremal combinatorics.Mathematic
Essays on the Economics of Social Insurance and Labor Markets
Get PDFThis dissertation investigates the labor market and fiscal consequences of social insurance reforms, drawing on large-scale administrative data from Germany. Across three chapters, I examine how programs designed to insure against economic risk - whether through wage subsidies, public insurance expansions, or regulatory wage floors - affect employment, workforce composition, and broader economic outcomes.
The first chapter analyzes the employment effects of Germany’s Short-Time Work (Kurzarbeit) program during the COVID-19 crisis. Using matched administrative data on program participation and the universe of Social Security records, I show that while Short-Time Work mitigated immediate job losses in 2020, participating firms experienced persistently lower employment growth through 2022. Instrumental variables estimates suggest that extensive program use led to substantial job hoarding, merely delaying, but not averting, adjustments. These findings highlight the short-run benefits and long-run costs of broad-based job retention subsidies.
The second chapter evaluates the impact of Germany’s 2022 Healthcare Development Act, which introduced wage floors in long-term care (LTC). Using a difference-in-differences design comparing LTC nurses to hospital nurses, I find that the reform raised wages particularly at the bottom, without reducing employment or firm entry. Wage compression resulted from both genuine wage gains and positive selection, as lower-paid workers were more likely to exit and newly hired staff experienced accelerated wage growth.
The third chapter studies the long-run labor market effects of Germany’s 1995 introduction of universal LTC insurance. Leveraging regional variation in pre-reform coverage and rich individual work histories, I show that the insurance expansion spurred large-scale job creation in nursing homes without displacing employment in other sectors. Employment gains were concentrated among lower-skilled workers, and local unemployment declined. A structural model reveals that the fiscal and welfare effects depend critically on pre-existing labor market frictions and tax wedges, with the reform effectively paying for itself through fiscal spillovers.
Together, these studies offer empirical evidence on how well-designed social insurance programs can support vulnerable workers, stimulate labor demand, and shape earnings distributions, underscoring the importance of targeting and institutional context for ensuring long-run efficiency.Public Polic
Evolutionary motif swapping of human dihydrofolate reductase rewires the enzymatic cycle
Get PDFDespite common descent, enzymes often rescue poorly when expressed across domains
of life. Causes include factors external to the enzyme, such as differences in codon usage,
sensitivity to proteases, and transcriptional or post-translational regulatory differences, yet
often the underlying cause remains unclear. Dihydrofolate reductase (DHFR) catalyzes the
same metabolic conversion across the tree of life. Nevertheless, human DHFR (hsDHFR)
does not effectively rescue growth of DHFR-deficient E. coli despite similar in vitro kinetics.
This phenomenon has been previously attributed to inhibition of hsDHFR by its oxidized
cofactor, NADP+. To understand this phenomenon, we designed mutants based on deep
sequence divergences across the tree of life, yielding variants which outperform both wild-type
enzymes in vitro and which rescue growth of E. coli. Remarkably, a single, ancient sequence
insertion underlies gain of function, not by modulating product inhibition, but by redirecting
ligand flux – the non-equilibrium sequence of steps binding and unbinding product, cofactor,
and substrate. We find that deleting this insertion decouples the dynamics of a substrate
binding loop from subdomain motion, thereby controlling a critical enzymatic parameter
orthogonal to catalytic proficiency.Biology, Molecular and Cellula
Characterization and Optimization of a Murine Engineered CAR T cell Surrogate for the Treatment of Autoimmune and Autoinflammatory Conditions
No full textThe adaptive immune system can occasionally generate cells that are autoreactive,
leading them to trigger autoimmune responses led by diminishing function or numbers of
Tregs. Regulatory T cells (Tregs) are a rare subpopulation of T cells that play a critical
role in the maintenance of homeostasis and immune tolerance. While Chimeric Antigen
Receptor (CAR) T cell therapy is already an important therapeutic modality for tackling
complex autoimmune conditions, the generation of a CAR Treg may prove more
effective and specific, treating the autoimmune condition more directly without the off
target toxicity effects. This thesis evaluates preclinical surrogates for CD19 CAR T cells
which, following optimization, were compared to CD19 CAR Tregs in a murine
inflammatory model in terms of safety, engraftment, off-target effects, and efficacy.
Following optimization, the newly engineered CAR T cell was an effective control tested
in co-culture assays with B cells and T cells, showcasing the off-target immune response
that was not observed with the specific CAR Treg cultures. In an in vivo model of
systemic lupus erythematosus, CAR T cells generated a significant immune response
compared to the CAR Tregs. This suggests that a specific CAR Treg would be a safer
alternative to a CAR T cell-based therapy.Extension Studie