Welcome To OpenBU
OpenBU is Boston University’s digital institutional repository for scholarly articles, theses and dissertations, preprints, and grey literature. This repository enables BU researchers to share, disseminate, and preserve their scholarship, and makes their research more accessible
If you are looking for information on BU's opt-out open access policy, please visit the BU Open Access Policy page.
Communities in OpenBU
Select a community to browse its collections.
Recent Submissions
Implant survival rate in sites previously augmented with guided bone regeneration using titanium mesh: systematic review
(2026) Chan, Yu-Ting; Price, Albert
Alveolar ridge defects following tooth loss often require bone augmentation prior to dental implant placement. Titanium mesh has been increasingly used in guided bone regeneration (GBR) due to its rigidity, biocompatibility, and ability to maintain space for bone formation. This systematic review evaluated implant survival rates following ridge augmentation using titanium mesh and assessed mesh exposure rate and mesh thickness as secondary outcomes.A systematic literature search of PubMed, Embase, and Web of Science was conducted for studies published up to April 30, 2024, using Medical Subject Headings (MeSH) and relevant keywords related to dental implants, guided bone regeneration, and titanium mesh. Study selection was performed in accordance with PRISMA guidelines. Only human clinical studies reporting implant survival rate following ridge augmentation with titanium mesh were included. Sixteen studies met the inclusion criteria, with combined totals of 344 participants, 356 augmented sites, and 679 implants. Implant
survival rates ranged from 81.8% to 100%, with an overall mean survival rate of 98.2%. Titanium mesh exposure was reported in most studies, with a mean exposure rate of 32.02%. Mesh thickness was inconsistently reported, ranging from 0.1 mm to 0.3 mm.
Titanium mesh-assisted GBR appears to be a predictable technique for ridge augmentation prior to implant placement. However, the limited availability of randomized controlled trials and heterogeneity among included studies warrant cautious interpretation and highlight the need for further high-quality prospective research.
Beyond first steps: examining associations between early motor skills and language outcomes in late talkers (LT)
(2026) Quigley, Caitlyn Carrington; Iverson, Jana M.; Maffei, Marc
In the first three years of life, children experience rapid growth in expressive language, progressing from simple speech sounds and gestures to complex utterances and conversational exchanges. However, late talkers (LTs), fail to meet these milestones, and may go on to experience persistent language disorders or receive diagnoses including autism spectrum disorder (ASD). The present study aimed to examine (1) the associations between 24-month abilities (gross motor, fine motor, and language skills) and expressive language outcomes at 36 months (total number of words [TNW], number of different words [NDW], and mean length of utterance [MLU]) and (2) whether 24-month motor skills (fine and gross) account for additional variance in 36-month expressive language outcomes beyond that explained by 24-month language ability. Children with late language emergence (n = 21), drawn from a longitudinal cohort of children at elevated likelihood for autism spectrum disorder, were followed from 24 to 36 months to examine associations between early motor and language abilities and later expressive language outcomes. Expressive language at 36 months was assessed using natural language samples to derive TNW, NDW and MLU. Results indicated that 24-month fine motor skills were significantly associated with all 36-month expressive language outcomes, such that stronger early fine motor abilities corresponded to greater vocabulary size, lexical diversity, and morphological complexity. In contrast, gross motor skills were not significantly associated with later language outcomes. Hierarchical regression analyses revealed that fine motor skills accounted for the largest amount of additional variance in expressive language outcomes beyond that explained by 24-month language ability and gross motor skills, particularly for lexical diversity, although models did not reach statistical significance, likely due to the small sample size. Overall, findings suggest that early fine motor development plays a meaningful role in supporting later expressive language growth in late talkers and may serve as an early indicator of language trajectories.
Neuromechanics underlying force control in children with unilateral cerebral palsy
(2026) Kim, HyunJoon; Ferre, Claudio L.
Effective force control is a key component of successful motor performance in activities of daily living. Early brain injury, which commonly results in unilateral cerebral palsy (UCP), leads to unilateral motor impairments, limits activities and participation in age-appropriate activities. These impairments persist through adulthood. This dissertation comprises three studies designed to provide mechanistic information that can inform rehabilitation strategies for children with UCP, by characterizing force-control behavior and its corticospinal substrates, and by evaluating neuromodulatory interventions informed by those mechanisms. In Study 1, we aimed to determine force control patterns and their association with neurophysiology in neurotypical (NT) adults in uni- and bimanual conditions. In Study 2, we aimed to characterize force control patterns in NT children and children with UCP, in relation to underlying neurophysiology. In both Study 1 and 2, we used submaximal isometric force control tasks and single-pulse transcranial magnetic stimulation (TMS) to measure hand grip force control and neurophysiology, respectively. In Study 3, we aimed to identify potential rehabilitation approaches that might be informed by the mechanistic information gained from Study 1 and 2. Thus, we performed a meta-analysis to examine the efficacy of movement-based rehabilitation protocols combined with therapeutic brain stimulation, or, transcranial direct current stimulation (tDCS). Study 1 included 28 NT adults, and Study 2 included 12 children (10 NT, two UCP). Study 3 included twenty individual studies with 378 children with CP. Study 1 identified force-level dependent differences in submaximal force control parameters. In addition, during the bimanual condition in submaximal force control, we observed a clear bilateral deficit-like phenomena at higher force levels compared to lower force levels with respect to force production, variability, and regularity. Correlation analyses between force control parameters and neurophysiological outcomes revealed that higher cortical excitability was associated with amplification of force fluctuation properties, whereas spatial characteristics were associated with greater force production and stability. Representational outputs were associated with modulation of force fluctuations. Together, these results suggest that corticospinal mapping, a measure of functional organization, rather than cortical excitability, underlies stable force control with greater force production in NT adults, highlighting brain topography as a candidate target for rehabilitation approaches aimed at improving motor function. In Study 2, we observed lower force production and stability with greater interlimb asymmetry in children with UCP compared to NT children. Similar to NT adults, indices of corticospinal spatial organization showed positive associations with improved force control in NT children. Additionally, more lateral and anterior center of gravity (COG) positions, a measure of the amplitude-weighted center of corticospinal representation, showed potential association with better force control. Representational output of corticospinal activity was also associated with force production and irregularity of force output. Correlation analyses across children with UCP and NT children revealed that greater force production, accuracy, stability, irregularity, and synergy were associated with better motor performance. Together, these findings suggest that rehabilitation targeting force control capabilities and interlimb coordination may be a strategy to enhance upper limb motor function in children with UCP. In Study 3, a meta-analysis demonstrated positive effects of tDCS on motor function in children with CP, with the largest effect sizes observed for anodal tDCS, a form of stimulation known to increase excitability, on M1. These findings suggest that anodal tDCS targeting M1 may effectively improve motor function in children with CP. Taken together, these findings suggest that anodal tDCS targeting the primary motor cortex (M1) may represent a potential therapeutic avenue for improving motor function in children with CP. Mechanistic insights from Studies 1 and 2 can inform targeted neuromodulation and therapy designs aimed at restoring force control and reducing interlimb asymmetry in children with UCP.
From staff development to student learning: AI in business research instruction
(ALA (American Library Association), 2026-05-18) Edmunds, Brock
[As generative AI tools like ChatGPT rapidly reshape academic research, librarians are stepping into uncharted territory, learning these technologies while simultaneously guiding students in how to use them responsibly. In response, Boston University Libraries launched Tech and AI Talks in January 2025: a monthly brown-bag series of informal, hour-long sessions co-led by Health Sciences Librarian Kate Silfen and I, engaging staff with emerging technologies, especially generative AI.]
Essays in contract theory, information, and stochastic analysis
(2026) Zhang, Yuyang; Xing, Hao; Salins, Michael
This dissertation studies stochastic systems arising in economic decision-making and mathematical analysis, with a primary focus on delegated investment under information frictions. The first part develops a unified framework for delegated investment and optimal contracting in a noisy rational expectations equilibrium, building on and extending foundational models in the literature. Portfolio managers acquire costly information about asset payoffs while incurring portfolio management costs, and make investment decisions on behalf of delegated investors. Within this framework, optimal contracts incorporate benchmarking components to mitigate agency frictions induced by these costs. The model endogenizes information acquisition and characterizes how equilibrium market conditions determine the precision of private signals. We show that private and public information act as substitutes in the partial equilibrium, and that higher portfolio management costs lead to greater reliance on benchmarking, reduced aggregation of private information into prices, and lower public price informational efficiency. From a welfare perspective, we further show that a social planner who places sufficient weight on direct investors or liquidity providers can improve price efficiency relative to the decentralized equilibrium. The second part investigates stochastic partial differential equations, focusing on finite-time explosion for a broad class of superlinear stochastic parabolic equations on bounded spatial domains. We consider general second-order self-adjoint elliptic operators, a wide range of space-time colored noise structures, and various boundary conditions, including Neumann, periodic, and Dirichlet cases. Under suitable conditions on the growth of the nonlinearity, we establish nonexplosion results in arbitrary spatial dimensions and extend existing theory by identifying sharp conditions on admissible growth rates. Although these two parts address distinct applications, they are unified by a common focus on stochastic modeling and the role of randomness in both discrete and continuous systems.
David Halberstam: defying an American president to change American journalism
(2026-05-19) Green, Leia
Climate change and health disparities: assessing local, regional, and national health inequities in the United States
(2026) Black-Ingersoll, Flannery Jean; Nori-Sarma, Amruta
Climate change is a critical global public health concern with unevenly distributed health hazards. In the U.S., persistently marginalized populations face the highest burden of hazardous environmental exposures and subsequent health disparities. This dissertation includes a novel method for health impact assessments of environmental health disparities and two quasi-experimental study designs evaluating understudied populations and health outcomes.In my first study, I addressed a gap in publicly available, highly spatially resolved population data, constructing an address-level synthetic population (‘matched allocation’) using publicly available U.S. Census and tax property data. I compared this population with a random assignment of households to addresses. From these comparisons, I found that strong interactions between modifiers and exposures contributed to differences in health impacts. In my second study, I used a quasi-experimental study design with a difference-in-differences estimator to investigate psychiatric emergency services use for 1) hot days with and without alerts, and 2) hot days after compared to before the implementation of a citywide climate resilience plan. I found no appreciable evidence of change in the incidence rate ratios (IRRs) for: services use on hot days with compared to without alerts (IRR: 1.06, 95% CI: 0.93, 1.19) or services use on hot days after compared to before the plan (IRR: 1.03, 95% CI: 0.97, 1.09). Alert systems alone may be insufficient at preventing adverse outcomes in vulnerable populations. In my third study, I used a time-stratified case-crossover study design to examine associations between summertime heat index and mental-health related hospitalizations among adult Medicaid beneficiaries. I derived climate, sex, age, and race/ethnicity group-specific effect estimates and conducted a meta-analysis across climate regions. Models with lags (0-6 days) showed increased risks at the 1st (IRR: 1.24, 95% CI 1.21, 1.28) and 99th (IRR: 1.07, 95% CI 1.04, 1.11) percentiles in the continental group and decreased risk in the 95th percentile (IRR: 0.94, 95% CI 0.91, 0.97) in the tropical/temperate group. Further stratifications revealed greater magnitudes of effects among males, older age groups, and non-White race/ethnicity groups.
These three studies implement a range of methods to identify and examine climate change-related health disparities.
Depositor coordination and the dynamics of bank runs
(2026) Yuan, Ziqi; Xing, Hao
This dissertation studies how idiosyncratic shocks, strategic interactions among depositors, and bank funding structures jointly shape the timing and severity of bank runs. Chapter 2 develops a continuous-time model of bank runs using a mean-field game framework that captures strategic complementarities among depositors who receive idiosyncratic shocks. Depositors observe private states and choose when to withdraw, while the probability of bank failure increases with the aggregate mass of withdrawals. The model generates two empirically relevant features of run dynamics: an initial "run-in-cluster'' followed by gradual withdrawals from remaining depositors. The analysis further shows how idiosyncratic shock dynamics, including long-run expectations about fundamentals, create a trade-off between short-term stability and fragility once a run begins. Chapter 3 provides empirical evidence from the collapse of Silicon Valley Bank in March 2023. Using an event-study and a difference-in-differences design, we show that banks with higher exposure to uninsured deposits experienced significantly larger declines in market-to-book valuations following the crisis. Consistent with the model's predictions, banks with greater reliance on uninsured deposits also experienced greater subsequent deposit outflows. In summary, the theoretical and empirical results highlight how depositor coordination and bank liability structures interact to generate financial fragility in modern banking systems.
Articulatory impairments in Parkinson's disease: insights from acoustics, kinematics, and listener perception
(2026) Dragicevic, Daria Alessandra; Stepp, Cara E.
Articulatory impairments are common in people with Parkinson’s disease (PwPD) and contribute to reduced speech intelligibility and communicative participation. Reductions in articulatory range of motion have been well-documented in PwPD; however, despite being described as one of the most deviant perceptual qualities in the speech of PwPD, articulatory precision remains relatively unexplored. This dissertation examined how articulatory range of motion and articulatory precision, measured using acoustic and kinematic metrics, contribute to articulatory impairments in PwPD. The first study evaluated whether clear speech cueing results in greater changes in articulatory range of motion or articulatory precision. Thirty-one PwPD completed speaking tasks during which simultaneous acoustic and kinematic data were collected under both typical and clear speaking conditions. Results showed that articulatory range of motion measures exhibited significantly greater differences between speaking conditions than precision measures, with no differences between acoustic and kinematic measures. The second study examined which objective metrics were most strongly associated with listener-perceived articulatory precision. Thirty-two PwPD completed speaking tasks, and fifteen listeners rated articulatory precision from these speaker recordings. Results indicated that range of motion measures were the strongest predictors of listener ratings of articulatory precision, whereas precision measures explained minimal variance. Acoustic and kinematic models both showed strong predictive performance and performed similarly well. Together, these findings enhance our understanding of articulatory precision and range of motion, measured through acoustic and kinematic measures, and how these metrics contribute to articulatory impairments in PwPD.
Continuous-time trading with multiple insiders and price impact
(2026) Li, Boyi; Robertson, Scott
We study imperfect competition among risk-averse insiders in a continuous time Kyle-Back model framework with price impact, allowing informed traders with heterogeneous CARA risk aversion to enter the market sequentially. We show that, unlike in the risk-neutral case with perfectly correlated signals, a linear equilibrium exists even when multiple insiders observe the same signal. Our main application considers a market in which one insider observes the asset value at time 0, while a second insider receives the same signal at a later date. We characterize the equilibrium explicitly and compare it with the single-insider benchmark. Before the second insider enters, the first insider trades more aggressively in anticipation of future competition, which accelerates early information revelation and lowers market depth. After entry, however, the first insider’s trading intensity declines because of the risk-sharing effect. As a result, price informativeness is always higher before entry than in the single-insider benchmark, whereas after entry it may be either higher or lower, depending on the timing of the second insider’s arrival. Market depth displays a similar pattern: it is lower before entry, while after entry it may remain lower the benchmark or cross it, depending on the timing of entry. We also show that the first insider’s welfare is lower than in the single-insider benchmark, but it increases as the second insider enters the market later. More generally, we characterize how differences in insiders’ risk aversion shape their equilibrium trading behavior.