More than 325 undergraduates gathered in Sayles Hall on July 31 and August 1 to present the results of their research at the annual Summer Research Symposium sponsored by the Office of the Dean of the College.
Forty-five presentations were from undergraduate researchers from the School of Engineering, or were conducted under the guidance of engineering faculty members. Many of the projects were supported by Brown's Karen T. Romer SPRINT|Undergraduate Teaching and Research Awards (UTRAs).
Brown’s Dean of the College and Abbott Gleason University Professor of History and Slavic Studies Ethan Pollock opened the symposium Thursday by addressing the participants who had filled Sayles Hall with their poster presentations. “This work represents an exceptional range of topics in the life sciences and physical sciences. Students are advancing research in cancer, Alzheimer's, and Parkinson's, while also confronting the ethical and societal challenges posed by emerging technologies. These projects reflect depth and critical thinking, and together demonstrate the power of undergraduate research to address complex questions at the intersection of the sciences and society.”
“This summer, I gained hands-on experience in computational modeling using MATLAB, developing simulations to explore biological systems at a mechanistic level,” said Dariana Alvarez-Ortiz ‘27, who worked in the lab of Associate Professor Vikas Srivastava. “In the process, I deepened my understanding of key mathematical frameworks and saw how theoretical simulations can drive real insights in biomedical research. Beyond just the technical skills I gained this summer, this was also my first real experience in academic research.”
Kathy Le ’27, who spent the summer in Associate Professor Nora Ayanian’s laboratory, added “Through my experience working in the lab, I was exposed to the process of real academic research, including working with graduate students and other classmates. I gained a better understanding of the possibilities we can achieve with the technology we have at our disposal. This experience gave me a better understanding of the type of work I would want to be involved with after graduation.”
Following are projects presented by engineering students and/or advised by professors in the School of Engineering:
Karunmay Aggarwal ’28 (mechanical engineering) presented the poster “Creating Actively Controlled Turbulence in the Wind Tunnel,” advised by Professor Kenny Breuer and supported by a SPRINT|UTRA.
This project presents the design and construction of an Active Turbulence Grid (ATG) for controlled turbulence generation in wind tunnel experiments. The ATG consists of a modular array of diamond-shaped vane elements, each actuated by high-speed servo motors that enable dynamic, programmable manipulation of airflow.
Dariana Alvarez-Ortiz ’27 (mechanical engineering) and Valeria Quero ’27 (engineering, computer science) presented “The QAO Kinetic Framework: An Experimentally-Driven PDE Model to describe the Metabolics of the Tumor Microenvironment.” They were advised by Associate Professor Vikas Srivastava, and supported by a SPRINT|UTRA.
Despite decades of observing the Warburg effect in cancer cell metabolism, its regulatory mechanisms are still poorly understood and current theoretical models are limited. Moreover, hypoxia and acidity, as microenvironmental stressors, can shape cellular phenotype. Thus, a model that captures the abiotic factors in the tumor microenvironment and their interplay is needed.
Soujanya Aryal ’27 (electrical engineering), Advait Mishra ’28 (chemical engineering, applied math), Qizhi Sun ’26 (environmental science, economics, geology), and Noreen Chen ’27 (computer science, environmental science) presented the research “WebGIS for the Arctic - Visualizing Environmental Risks and Indigenous Observations.” They were mentored by Assistant Professor of the Practice of Earth, Environmental, and Planetary Sciences Seda Salap-Ayca and supported by a SPRINT|UTRA award.
The Arctic, increasingly threatened by industrial activities including mining and hydrocarbon extraction, is a region whose ecosystems and communities are prone to serious environmental risks. The threat of these risks is exacerbated by a lack of accessible, community-driven monitoring systems. This project seeks to create an interactive Web GIS (Geographic Information Systems) tool to improve hazard monitoring in the Arctic.
Olivia Baptiste ’26 (biomedical engineering) worked on the “Development of an Implantable Talus Replacement for Measuring In Vivo Loads in the Ankle Joint.” Baptiste was advised by Professor of Orthopedics and Professor of Engineering (Research) J.J. Trey Crisco, and supported by a SPRINT|UTRA.
Understanding the loads that occur across musculoskeletal joints is critical for advancing knowledge of joint function, pathology, implant design, and biomechanical modeling. While hip and knee joint loading have been studied extensively, in vivo load measurement in the foot and ankle remains underexplored. The talus plays a key role in load transmission between the tibia and calcaneus, making it essential for studying foot biomechanics and pathologies such as post-traumatic osteoarthritis and avascular necrosis. This project aims to design and assess the feasibility of a novel instrumented talus implant to measure real-time in vivo loads within the ankle joint.
“High Throughput Design of Experiment Optimization of Lipid Nanoparticles for Macrophage Delivery” was the topic of Anneliese Baum’s ’27 (biomedical engineering) research, supported by a SPRINT|UTRA and advised by Manning Assistant Professor Theresa Raimondo.
Immunotherapies for solid tumors often prove ineffective due to an immunosuppressive tumor microenvironment (TME). Among the various cell types that contribute to this type of TME are tumor associated macrophages (TAMs). TAMs typically display an anti-inflammatory, “M2-like” phenotype, contributing to tumor growth and immune evasion. One non-receptor tyrosine kinase (NRTK) has been identified as a key positive regulator of the M2 phenotype and is associated with tumor fibrosis, macrophage migration, activation, and autophagy. Thus, researchers hypothesized siRNA knockdown of this kinase in TAMs can produce an anti-tumor effect.
Ella Blanco ‘28 (design engineering, classics) presented research on “Enhancing Ground Robot Perception through Multi-Sensor Integration and Custom Hardware Design.” Blanco was advised by Associate Professor Nora Ayanian and supported by a SPRINT|UTRA.
Contemporary ground robots, including advanced platforms such as the Boston Dynamics Spot, typically rely on integrated sensor suites comprising onboard cameras and LiDAR systems for environmental perception and mapping. However, these standard configurations frequently demonstrate limitations in capturing comprehensive spatial and contextual information necessary for complex autonomous operations. This research investigates the enhancement of real-time perception capabilities through the strategic integration of supplementary vision sensors within a sophisticated multimodal neural network architecture.
“Liposomal Nanoparticles for Hyphal Inhibition to Mitigate Candida albicans Pathogenicity in Vulvovaginal Candidiasis” was investigated by Serly Chohmalian ’26 (biomedical engineering). Chohmalian was advised by Elaine I. Savage Professor of Engineering Anita Shukla through a SPRINT|UTRA award.
Vulvovaginal candidiasis (VVC) affects 75 percent of women at least once in their lifetime, with five percent experiencing recurrent infections annually. Candida albicans is the primary pathogen responsible for VVC. A key driver of C. albicans pathogenicity is its yeast-to-hyphal transition, which enables tissue invasion and the release of candidalysin, a toxin that damages host cells and triggers hyperinflammatory responses. While nanoparticles (NPs) have been widely explored as antifungal drug carriers, their potential to directly modulate fungal morphology as a therapeutic strategy remains underexplored. This study aimed to develop a nanoparticle-based strategy to inhibit C. albicans hyphal formation and enhance antifungal drug delivery.
Terrell Davis ’27 (materials engineering) and Melissa Truong ’28 (mechanical engineering) were supported by the Presidential Scholars Program (PSP) and were mentored by Associate Professor Vikas Srivastava. The researchers studied “Quasi-Static to High Strain Rate Experiments to Connect Tissue Deformation and Cell Injury in a Continuum-Scale Model of Bodily Injury.”
Understanding how cells embedded within soft tissue respond to mechanical stress is essential for advancing trauma research and improving biomaterial design. While many studies have focused primarily on chemical and biological injury mechanisms, the complex effects of pure mechanical loading, especially at high strain rates, remain inadequately understood. This knowledge gap is particularly critical given its implications in real-life situations such as blunt trauma, compression injuries, and pressure-induced tissue damage. This research aims to explore the relationship between mechanical deformation and cell injury using a reliable in vitro model. In the study, three-dimensional collagen-based hydrogels embedded with fibroblast cells were used to create biomimetic tissue surrogates that closely replicate natural conditions.
Jake Delesky ’27 (biology) presented the research “Validation of Lipid Nanoparticle Formulation for Delivery of Immune Checkpoint siRNA to Macrophages.” Delesky was advised by Manning Assistant Professor Theresa Raimondo and supported by a SPRINT|UTRA.
In recent years, lipid nanoparticles (LNPs) have become recognized as a highly successful method of delivering nucleic acids, especially in light of the development of the LNP-mRNA vaccines for use against the coronavirus disease. LNPs consist of ionizable lipids, phospholipids, cholesterol, and PEG-lipid conjugates. In acidic conditions, they encapsulate RNA through charged interactions. Following cellular uptake via endocytosis, the acidic endosome protonates the ionizable lipids. This causes membrane destabilization and promotes endosomal escape, allowing the RNA to be released. To maximize therapeutic efficacy, LNP formulation must be optimized to the target cell type, the target organ or tissue, and the loaded cargo. This can be done by altering the molar ratios of the lipid components, which changes their physical characteristics and can help achieve desired particle properties such as stability, encapsulation efficiency, cellular uptake, and promotion of endosomal escape [3]. This work aims to validate an LNP formulation for the delivery of macrophage-checkpoint targeting siRNA to IL-4-treated human THP-1-derived macrophages.
“A Public Land Record Analysis of Aspen, Colorado’s Silver Boom and Bust” was presented by Arianna Gerola ’27 (computer science), under the mentorship of Professor Rashid Zia. Gerola was supported by a SPRINT|UTRA award.
U.S. land records represent a large open dataset of immense historic value and public interest. These detailed records provide a wealth of information on how communities are formed and evolve. When integrated with other public data, including the decennial census, state corporation documents, and modern parcel maps, such records can provide relevant information into historical economic and social dynamics. Pitkin County, Colorado, and its primary city, Aspen, served as an interesting case study for Gerola.
Lilly Goldfarb ’27 (design engineering) presented “Restoring Expression: A Virtual Avatar System for Locked-In Syndrome Communication.” Supported by a SPRINT|UTRA award, Goldfarb’s research was under the direction of L. Herbert Ballou University Professor Leigh Hochberg.
Locked-in syndrome (LIS) is a devastating neurological condition in which individuals remain fully conscious yet almost entirely paralyzed, often retaining control of only vertical eye movements or blinking. Traditionally, people with LIS have communicated by using these limited eye movements to select letters or words from a visual chart, spelling out messages letter by letter. Unfortunately, as the condition progresses, even these small motor abilities can be lost, rendering communication nearly impossible. Recent advances in computational neuroscience and brain-computer interfaces (BCIs) are opening new doors for restoring communication in these extreme cases. By implanting electrodes and using neural decoding systems, researchers can now detect and interpret distinct brain wave patterns, allowing locked-in individuals to control external devices without any physical movement. However, these neural decoders vary significantly in accuracy and resolution, often depending on patient-specific factors and calibration. While some individuals can produce up to thirty distinguishable signals—enough to represent a full alphabet and symbols—others may only reliably generate four unique signals. To address this, BrainGate has developed a virtual keyboard interface that can be operated with just four unique commands. This four-way gesture decoder enables participants to navigate a virtual keyboard, select letters, and construct words or phrases by repeatedly choosing among four options. Once a word or sentence is formed, the system uses a Redis stream to pass the text to a text-to-speech (TTS) engine, which vocalizes the participant’s intended message. This technology has already made significant strides in restoring basic communication abilities for people who otherwise have no means to express themselves. 185 Yet, verbal output alone does not fully capture human communication, which also relies on facial expressions and gestures to convey emotion and nuance. To address this gap, our project explores the development of a virtual speaking avatar that animates alongside the TTS output. By synchronizing facial animations with the spoken words, the avatar can provide an additional layer of expressiveness and social presence for the user. Preliminary work has begun in Blender to design the avatar’s visuals and basic movements. The next phase will focus on improving the avatar’s realism, developing a robust animation system, and integrating it with real-time text-to-speech output. This project bridges human-computer interaction, accessibility design, and computational neuroscience, aiming to restore not only the ability to speak but also the expressive qualities that make communication profoundly human.
Thomas Gordon ’26 (computer engineering) was mentored by Professor Sherief Reda and presented the research “Adapting Quantization Techniques For Multi-Task Learning Models.” Gordon was supported by a SPRINT|UTRA.
While deep neural networks present immense potential for tasks such as computer vision and machine learning, their demand on both computational load and energy consumption provide logistical challenges, especially for augmented reality (AR) devices. AR devices require many sensor inputs for processing, which greatly increases their computational load; but they also often operate in embedded or real-time environments, where their available energy and processing power is significantly limited. These networks often cannot function quickly enough for real-time visual data processing. There is therefore a crucial need for machine-learning models that have minimal computational and power load, but still provide accurate calculations. These models can lower these resource requirements through quantization, where a continuous range of values are approximated by a smaller set of discrete possibilities—trading some non-essential accuracy for power efficiency and quicker computation. The technique works by “rounding” highly accurate floating point data into less precise formats, sometimes down to a single bit. The method in which quantization can be most effectively used is an ongoing field of research in machine learning, and advancements here can result in faster, smaller, and more power-efficient models. Low- and mixed-precision quantization is a well-established method of optimization for neural networks, and a substantial body of research has already been done in examining the tradeoffs between quantized architectures and their impacts on accuracy. However, existing quantization research overwhelmingly applies to models trained on single tasks. This project observes whether the flexibility of multi-task learning models could allow new optimization strategies, and looks to adapt existing techniques (such as blockwise quantization and quantization of elementwise operations) to these MTL models. Improved efficiency techniques for these models would be a step towards successful AR frameworks on resource-constrained devices.
Luke Guan ’27 (mathematics, computational biology) presented the poster “ENGN Machine Learning Model Comparison and Validation for Lipid Nanoparticle (LNP) Optimization.” Guan was supported by a SPRINT|UTRA, and mentored by Manning Assistant Professor Theresa Raimondo.
The recent success of mRNA vaccines for COVID-19 has driven increased interest in RNA therapeutics, a class of promising new medications that rely on the delivery of RNA molecules to target cells. To prevent RNA degradation and facilitate drug targeting, RNA molecules are often encapsulated in lipid nanoparticles (LNPs). Our LNPs consist of a mixture of four lipid components: an ionizable lipid (MC3), a phospholipid (DSPC), a polyethylene glycol conjugated lipid (PEG), and cholesterol. Previous research finds that each individual lipid component can have different or opposing effects on a variety of clinically important outputs such as uptake, cell viability, particle localization, and siRNA-driven gene silencing. Since these components form a mixture, the effect of each lipid component cannot be isolated by independently changing its molar ratio. This makes the task of finding the best LNP formulation a complex optimization problem with four interacting input variables and multiple competing outputs. To solve this optimization problem, we fit various machine learning (ML) and linear regression models to our data and compared their predicted response surfaces and optimal formulations.
“Understanding the Dynamics of Intercalation of Potassium Ions in a Graphene Homo-bilayer for Clean Energy Storage Applications,” was presented by Julius Hochberg ’27 (engineering physics). Hochberg was mentored by Physics Assistant Professor Matthias Keuhne and supported by a SPRINT|UTRA.
As the world takes necessary steps towards fully adopting sustainable energy, a significant challenge is posed by storing it. Currently, most clean energy storage systems depend on batteries that make use of lithium ions, putting significant pressure on lithium’s environmentally costly extraction. In order to facilitate the transition away from fossil fuels, it is crucial to develop better battery technologies–and ideally ones that do not require a critical resource like lithium. This project seeks to explore the potential of using readily available potassium ions for effective clean energy storage solutions.
Paul Hudes ’27 (applied math-economics, English literature) presented the poster “Identifying Small-Scale Ocean Features from Very-Sparse Lagrangian Data using Network Science.” Hudes was advised by Thomas J. and Alice M. Tisch Assistant Professor Monica Martinez Wilhelmus and supported by a SPRINT|UTRA.
Oceanic eddies play a crucial role in transporting nutrients, heat, microplastics, freshwater, and sea ice in the global ocean circulation. While eddies are relatively well characterized in mid-latitude regions, where 197 their scales typically range from O(10) to O(100) km, eddies in polar regions tend to be smaller, often below O(10) km. This poses a challenge for detection, as higher spatial resolution satellite observations are required. In addition, in situ measurements are sparse due to harsh environmental conditions. To address these limitations, the Wilhelmus group has demonstrated that floating sea ice plates (floes) can serve as effective Lagrangian tracers for measuring surface oceanic eddies in polar regions. Prior work has focused on analytical and simulation-based approaches to characterize the kinematics of surface eddies. In this project, the viability of applying machine learning to identify eddy characteristics from sparse particle trajectory data is investigated.
Akari Inami ’27 (chemical engineering) presented the poster “In-Situ Remediation of PFAS-Impacted Groundwater: Experimental Evaluation of Particulate Amendment Delivery and Retention in the Subsurface.” Inami was mentored by Professor Kurt Pennell and supported by a SPRINT|UTRA award.
Per-and polyfluoroalkyl substances (PFAS) are a class of synthetic chemicals widely used in consumer products and industrial applications. Often called “forever chemicals”, PFASs are resistant to degradation due to exceptionally strong carbon-fluorine bonds. As a result, PFAS persist in the environment, migrate in water and soil, and accumulate in the human body. PFAS exposure has been associated with serious health issues, including increased risk of cancer and damage to the immune/reproductive systems. Conventional treatment of PFAS-impacted groundwater often relies on “pump and treat” methods that involve extraction and above-ground treatment. However, this approach can be energy-intensive and expensive. In-situ remediation using injectable particulate amendments has recently emerged as a potential alternative. In this approach, particulate amendments are directly injected into the subsurface via injection, where they form a permeable adsorptive barrier (PAB) that is designed to immobilize PFAS and prevent further migration. However, published data on amendment transport and retention under realistic conditions are scarce and provide insufficient guidance for effective field implementation. To address this knowledge gap, a combination of batch, column and aquifer cell studies are being conducted to assess transport and retention of polymer stabilized powdered activated carbon and ion exchange resin in representative porous media. Relevant properties of the soils were also collected. The findings aim to improve the fundamental understanding of amendment delivery and retention, which can be utilized to support remediation design and optimize in-situ PFAS treatment strategies.
Kira Ivanova ’28 (engineering, economics) presented “Tracing the Metal Contamination in an Urban Watershed: Integrating Data and Reactive Modelling in Historically Polluted Site in Providence.” Ivanova was advised by Assistant Professor of Earth, Environmental, and Planetary Sciences and Environment and Society, and Assistant Professor of Engineering Daniel Ibarra and supported by a SPRINT|UTRA award.
Urbanization and industrial activities have introduced persistent and toxic trace metals into urban watersheds, altering natural hydrological and biogeochemical processes. Understanding the mobility, speciation and long-term fate of these metals is critical for effective risk assessment and remediation. This study investigates coupled hydrological and geochemical controls on metal transport at the Centredale Manor Superfund site, a historically contaminated watershed subject to legacy pollution from direct waste disposal and urban runoff.
“Characterizing the Impact of Stretch on Structure and Function of Cardiomyocytes on Second-Generation Bioelectric Thread” is the topic of Benjamin Jamal ’27 (biomedical engineering) summer research. Jamal was advised by Associate Professor Kareen Coulombe and supported by a SPRINT|UTRA award.
Heart attack survivors face permanent heart damage and higher risk of developing arrhythmias and heart failure. Implanting engineered heart tissues (EHTs) to restore contractility to the damaged myocardium is a promising therapy, but is challenged by poor integration with the host heart. To address this, we engineered an electrically conductive, or bioelectric, thread to facilitate electrical coupling between two tissues.
“Engineering Profiling Circulating Tumor Cell Heterogeneity using Computer Vision and Machine Learning” is the project Novia Jiang ’28 (biomedical engineering, political science and East Asian studies) undertook this summer. Jiang was advised by Associate Professor Ian Y. Wong and Ph.D. student Alexxa Cruz-Bonilla, and was supported by a SPRINT|UTRA.
In a patient with cancer, metastasis may take place, where cancer cells detach from the original tumor, travel to other parts of the body, and form new tumors. This process is associated with later stages of cancer and cancer-related deaths. Circulating tumor cells (CTCs) are such cells involved with metastasis. Like other cancer cells, CTCs display great heterogeneity in terms of shape and biomarker expression. This makes profiling them with computer vision and machine learning difficult. This project aims to investigate morphological changes and proliferation rates of CTCs based on the microenvironment.
Hania Khan ’27 (mechanical engineering) presented research on “Mechanical Counterpressure: The Spacesuit of the Future.” Khan was advised by Adjunct Associate Professor Rick Fleeter and supported by a SPRINT|UTRA.
An alternative method of pressurization, Mechanical Counterpressure, is an elegant solution to issues of mobility, bulkiness, fatal failure modes and so on. Within the Mechanical Counterpressure literature, Khan found that fabric patterning as a method of eliminating longitudinal stress was largely unexplored. Hence, they focused on designing a prototype demonstrating that directing longitudinal stress along the skin’s “Lines Of Non-extension” would allow the comfortable usage of fabrics with high elastic moduli in a Mechanical Counterpressure spacesuit.
Owen Landry ’26 (biomedical engineering) presented the research “Low Strain Rate Material Response of a Soft Tissue Surrogate of the Marine Mammal Melon.” Landry served as an Undergraduate Research Assistant in the lab of Associate Professor Vikas Srivastava, who advised on the project.
Certain marine mammals known as the Odontoceti, or toothed whales, possess an ovoid melon tissue located on their skull, which is involved in the production and reception of sound as part of their ability to use echolocation. These marine mammals can be exposed to the effects of underwater explosions in a variety of circumstances, resulting in potential damaging of their melon tissue. Previous studies have shown that tests designed to mimic the forces potentially experienced by marine mammal melon tissue by underwater explosions have shown that the damage caused by the impact resulted in a significant reduction in the speed of sound transmission through the tissue, the ultrasonic velocity through the specimen decreasing compared to the control group which received zero strain. This project aimed to observe the effects of exposure to different compressive strain rates and percentages on the speed of sound transmission through surrogate materials, as a result of the complications with obtaining and testing marine mammal melon tissue.
Kathy Le ’27 (computer engineering) and Devesh Kumar ’28 (applied math, computer science) worked with Associate Professor Nora Ayanian on “Using Reinforcement Learning to Train Autonomous Quadrotors.” The work was supported by a SPRINT|UTRA award.
Le and Kumar worked to develop a neural controller through reinforcement learning that enabled autonomous flight on a variety of drones. Being able to deploy neural controllers with the latest AI, that generalize across robots, would allow for more complex, ambitious projects which are difficult to model but easy to simulate such as carrying novel payloads, resistance to faults, and environmental challenges.
Thatcher LeBlanc ’27 (environmental engineering) presented “Model for Removal of PFAS from Water Using Foam Fractionation.” LeBlanc’s work was supported by a SPRINT|UTRA and advised by Professor Kurt Pennell.
Per- and polyfluoroalkyl substances (PFAS) present major environmental and health issues, from decreasing fertility rates to increasing cancer rates. Foam fractionation has emerged as a cost-effective technology to rapidly remove PFAS from groundwater. Due to their higher surface activity, long-chain PFAS readily accumulates on the surface of air bubbles and can be removed easily using this method. However, effective removal of short-chain PFAS requires the addition of a cationic co-surfactant, which acts to reduce the bubble size and ion-pair with negatively charged PFAS to be removed from a system. Although the foam fractionation has been studied in detail, mathematical modelling of the foam fractionation process for PFAS is limited. To address this knowledge gap, a mathematical model was created in MATLAB that incorporates two of the key processes that occur during removal; adsorption of PFAS and co-surfactants at bubble surface and the complexation of PFAS with the more surface active co-surfactants to enhance adsorption.
“Projected Gradient Ascent for Convex Maximization” was the project presented by Heon Lee ’26 (computer science, mathematics). Lee was advised by Professor Pedro Felzenszwalb, and supported by a SPRINT|UTRA.
Constrained optimization involves finding the best solution within a feasible region. We study the maximization of convex functions over convex domains. While standard methods typically require multiple steps to reach a solution, we demonstrate that in linear cases, a single projected step suffices for an arbitrarily good approximation. We further demonstrate the effectiveness of this approach on the Max-Cut problem, achieving strong approximation guarantees using only one projection. For more general convex functions, we analyze the behavior of projected gradient ascent and show that it converges to first-order stationary points. These results naturally connect the projected gradient ascent method to the conditional gradient algorithm through a limiting process.
Samuel Lorenzo ’26 (biomedical engineering) presented “Investigating Metabolic Stress Origins for Arrhythmogenesis in Non-PRKAG2 Wolff-Parkinson-White Syndrome using Patient-derived Atrial and Ventricular Engineered Cardiac Tissues.” Lorenzo was advised by Associate Professor Kareen Coulombe and supported by an Advanced Undergraduate Research Fellowship.
Wolff-Parkinson-White (WPW) syndrome is a heart condition characterized by accessory electrical pathways that bypass normal conduction between the atria and ventricles. This often results in premature ventricular contractions that can progress to life-threatening arrhythmias including ventricular fibrillation and cardiac arrest. WPW affects 1-3 per 1000 people worldwide and can be familial or non-inherited. While familial WPW is commonly associated with mutations in the PRKAG2 gene, causing cardiac glycogen overload and abnormal conduction, most WPW patients do not possess the PRKAG2 mutation, suggesting a non-monogenic etiology. To investigate the origins of non-monogenic WPW, our lab generated human induced pluripotent stem cells (hiPSCs) from a non-PRKAG2 WPW patient and successfully differentiated them into atrial and ventricular cardiomyocytes using modulation of the Wnt signaling pathway followed by lactate-based metabolic selection.
Jayanth Mani ’28 (computer engineering) presented the research “Walking Property Lines: Using Shapes to Identify Historical Land Transactions,” under the advisement of Professor Rashid Zia. Mani was supported by a SPRINT|UTRA.
Land records can serve as a window into the past, providing information about how communities formed and evolved over time through both public and private action. In the United States, these land deeds are often publicly available with complete data available as early as 1680. Many counties and towns offer digitized records, which allow historical land deeds to be searched by the grantor and grantee of the land. However, these databases rarely associate the land in each historical deed with specific geolocations. This poster presents methods to match survey descriptions contained in historical deeds to the shapes of modern GIS parcels. Methods we have investigated include: shape distances, angle comparisons, etc.
Peyton Marcotte ’27 (mechanical engineering) presented “PMARC Precision Motion and Resistance Cube (PMARC).” The presentation was supported by a Space Grant/NASA under the mentorship of Adjunct Associate Professor Rick Fleeter.
PMARC is a puzzle cube engineered for both spaceflight and terrestrial use, featuring a novel adjustable external tensioning mechanism that enables a 30-newton range of resistance. Designed to reduce exercise training time for astronauts through fine motor hand and wrist resistance, PMARC also offers therapeutic benefits on Earth, aiding in physical rehabilitation where muscle and bone density loss is common due to illness or inactivity.
“Determination of PFAS Distribution Between Bulk ice and Melting Water Using Batch Reactors,” was the project of Maya Marquez-Sturm ’26 (environmental engineering). Marquez-Sturm was supported by a SPRINT|UTRA and advised by Professor Kurt Pennell.
Per- and polyfluoroalkyl substances (PFAS) are a class of persistent, carcinogenic chemicals that have been increasingly recognized as major groundwater contaminants. As both public and regulatory awareness of PFAS grows, there is an increasing drive to understand its behavior in the environment. One area that remains poorly understood is how PFAS behave under freezing and thawing conditions. This research project set out to investigate how freezing impacts uptake, leaching and fate of PFAS. Understanding these dynamics is critical for refining EPA and other regulatory guidelines to account for seasonal changes that may enhance PFAS mobility in cold-climate conditions.
Nadia Michael ’26 (biomedical engineering), Francine Ho ’26 (health and human biology) and Tenasica Barnwell ’26 (public health and data fluency) worked on the project “Understanding Effects of Maternal Smoking During Pregnancy Using the Fetal Neurobehavior Coding System (FENS).” The group was supported by both a SPRINT|UTRA award and a PLME Summer Research Assistantship. They were advised by Professor of Psychiatry and Human Behavior Laura Stroud.
With an estimated 5.4% of pregnant people in the U.S. continuing to smoke cigarettes during their pregnancy, maternal smoking during pregnancy (MSDP) is a major public health issue (Kipling et al., 2024). MSDP is associated with increased infant morbidity and mortality, pregnancy complications, and alterations in offspring neurodevelopment (Stroud et al., 2020; Wells and Lotfipour, 2023; Wehby et al., 2011). Although preliminary studies indicate effects of MSDP on neurobehavioral development in infancy and across childhood, few studies have examined the effects on fetal neurobehavioral development. Thus, research to examine the impact of MSDP on fetal neurobehavior is needed to identify potentially negative fetal developmental outcomes to increase the possibilities for intervention. The Fetal Neurobehavior Coding System (FENS), developed by Dr. Amy Salisbury and colleagues, is used to assess the four domains of fetal neurobehavior: Fetal Heart Rate, Motor Activity, Behavioral State, and Responsiveness to Extrauterine Stimuli (Salisbury et al. 2005). Within the Behavioral State and Responsiveness codes there are five additional categories: Breathing, Mouthing, Isolated Movements, Complex Movements, and Stress Signs (Salisbury et al. 2005). Size of movements, quality of movements, and repetitiveness are also tracked. This project applies the FENS coding system in a set of previously recorded ultrasound videos to analyze fetal neurobehavior in fetuses that were and were not exposed to MSDP.
Aryan Narayan ’26 (biochemistry and molecular biology) presented the research “Proteinase K-Conjugated Magnetic Beads Enhance Lipoarabinomannan Detection for Serum-Based Tuberculosis Diagnostics.” Narayan was advised by Professor Anubhav Tripathi, and supported by a SPRINT|UTRA award.
Tuberculosis (TB) remains the leading cause of infection-related mortality worldwide. Current diagnostic methods for active TB are expensive, infrastructure-dependent, and time-consuming, significantly hindering timely diagnosis and treatment, particularly in low-resource settings. Lipoarabinomannan (LAM), a mycobacterial glycolipid, has emerged as a promising biomarker for rapid, point-of-care active TB detection. However, immunometric detection of LAM in blood is complicated by its low concentration and steric interactions with high-molecular-weight serum proteins, necessitating effective sample pretreatment for accurate antigen detection. Existing LAM-based diagnostic methods exhibit poor sensitivity, lengthy processing times, and reliance on high-temperature steps to accelerate enzymatic reactions and later deactivate enzymes used for deproteinization to preserve the antibodies employed in downstream immunometric assays. To address these limitations, we developed a room-temperature, isothermal pretreatment strategy using Proteinase K conjugated to magnetic beads via carbodiimide coupling chemistry, employing a two-step reaction activated by N-hydroxysulfosuccinimide (sulfo-NHS) and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC).
“Multi-Sensor Actigraphy” is the project of Myan Nguyen ’27 (applied math, computer science) and Kevin Yang ’28 (applied math, computer science). They were co-advised by Assistant Professor (Research) Marissa Gray and Computer Science Associate Professor Jeff Huang, and supported by a SPRINT|UTRA.
Actigraphy-based sleep monitoring has gained popularity due to its affordability and unobtrusive nature compared to polysomnography (PSG), the clinical gold standard. However, existing single-sensor actigraphy methods, typically wrist-worn, exhibit limitations such as underestimating sleep-onset latency and overestimating awakenings, thus affecting the accuracy of key metrics like Total Sleep Time (TST). In response, this study introduces a novel multi-sensor approach using four Axivity AX6 sensors placed on each limb (both wrists and ankles) to enhance sleep state classification accuracy, particularly addressing shortcomings in wakefulness detection (Wake After Sleep Onset - WASO).
Chompoonek Nimitpornsuko ’27 (electrical engineering, international and public affairs) and Ethan Kim ’28 (mechanical engineering) presented the poster “The Anchor: A Multi-Use Facility in Low Earth Orbit.” Adjunct Associate Professor Rick Fleeter advised the project, and it was supported by a SPRINT|UTRA award.
This technical paper proposes an ambitious yet pragmatic design of a multi-use facility in low Earth orbit (LEO), called the Anchor. While its external structural layout conforms to the cost-efficient design principle of “form following function,” the Anchor sports a human-centered internal layout intended to provide physical comfort, psychological well-being, and social cohesion. This dual-track approach seeks to create a balance between human welfare, cost efficiency, and scalability, enabling people to not only survive but thrive in an orbital facility.
Erica Sahin ’26 (biomedical engineering) presented the research “Degradation of Electrospun Polycaprolactone Scaffolds for Engineered Human Myocardium Composites.” Sahin was advised by Associate Professor Kareen Coulombe and was supported by a SPRINT|UTRA.
During a heart attack, restricted blood flow to the heart causes the death of one billion cardiomyocytes. The resulting scar tissue formation and ventricular remodeling impede contraction and limit cardiac output. The Coulombe Lab has developed engineered human myocardium (EHM) to promote cardiac regeneration by delivering human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) to the site of the infarction. An anisotropic electrospun polycaprolactone (PCL) scaffold has been engineered to be embedded within the tissue patch to lend stability during epicardial implantation and mechanical support for the heart’s pumping. To avoid increased inflammation and a foreign body response, the PCL scaffold is designed to degrade over time. Optimally, the scaffold will degrade at the same rate that the heart regenerates; as the synthetic material breaks down, newly deposited extracellular matrix (ECM) will take its place. The implanted hiPSC-CMs mature in vivo and provide active contractile support to the infarcted region. To fabricate the EHM composite, the PCL scaffold is cultured within a cardiomyocyte-loaded hydrogel for 1 week before implantation. Therefore, this project characterized the in vitro degradation rate of electrospun PCL scaffolds in solution over 28 days.
Roberto Serrano-Pomales ’26 (biomedical engineering) presented the research “Evaluating the mechanism of metabolite-based lipid nanoparticle (LNP) internalization in macrophages.” Serrano-Pomales was supported by the Shukla Lab, and advised by Elaine I. Savage Professor of Engineering Anita Shukla.
Staphylococcus aureus is a leading cause of bacterial death globally. S. aureus can form persistent biofilms that are not susceptible to clearance using our traditional antibiotic approaches. The complex biofilm structure hijacks the immune system by impairing and altering innate immune cells such as macrophages, involved in pathogen clearance. Therefore, there is a crucial need for novel therapeutics that can target and manipulate host immune responses to effectively clear bacterial biofilm infections. We hypothesize that modulating macrophage polarization from an anti-inflammatory (M2) towards a pro-inflammatory (M1) phenotype can enhance bacterial clearance. Lipid Nanoparticles (LNPs) offer a versatile platform with tunable composition, efficient drug encapsulation, and low toxicity that is promising for this application. This project aimed to fabricate metabolite-based LNP formulations and investigate the mechanism of nanoparticle uptake by bone-marrow derived macrophages.
Ziqi Shen ’27 (cognitive neuroscience, computer science) presented the research “Towards Restoring Upper Limb Movement for Individuals with Paralysis Using a Posture-tracking Unity Game.” Shen was advised by L. Herbert Ballou University Professor Leigh Hochberg and Assistant Professor of Neuroscience Carlos Vargas-Irwin, and was supported by a SPRINT|UTRA award.
Individuals with tetraplegia, such as those affected by Amyotrophic Lateral Sclerosis (ALS) or spinal cord injuries, face significant challenges in overall limb mobility. Previous studies have primarily focused on restoring single-joint movements, such as those of the shoulder or wrist. However, there has yet to be a task that incorporates movements across multiple joints simultaneously. This BrainGate soft robotics project aims to restore upper limb movement for these individuals, specifically targeting shoulder and elbow motion through an interactive Unity game. In this game, participants maneuver a character, a "banana man," to follow a "ghost man."
“A Bacteria-Responsive Hydrogel forTreatment of Topical Infections” is the poster presented by Jared Sonkin ’26 (biomedical engineering). Elaine I. Savage Professor of Engineering Anita Shukla advised the project, which was supported by the Shukla Lab.
Wound infections are a major clinical problem with 8.2 million cases in the United States in 2014. Topical antibiotics are the most common treatment for these infections, but their effectiveness is becoming increasingly limited due to the emergence of antibiotic resistance. The spread of antibiotic resistance can be limited by a more selective administration of antibiotics, which could be accomplished through a bacteria-responsive wound dressing. Hydrogels, which consist of a hydrophilic polymer network, have been established as suitable materials for wound dressings due to their ability to hydrate wounds and absorb wound exudate. Additionally, their capacity for functionalization and drug loading makes them well-suited for developing responsive wound dressings. Here, we designed a hydrogel-based wound dressing that degrades in the presence of esterases. These enzymes that are universally produced by bacteria have been studied as biomarkers for diagnosis of topical infection, but there has been limited research investigating their use as a stimulus for responsive delivery of therapeutics. We hypothesize that degradation of this hydrogel by esterases could trigger the release of encapsulated nanoparticle cargo for treatment of topical infections.
Hailey Stone ’28 (biochemistry) presented the poster “Design and Construction of Water Flume Experimental Equipment.” Stone was advised by Professor Kenny Breuer and supported by a SPRINT|UTRA.
Projects in design and construction of various elements for bio-inspired experiments in the Breuer Lab water flume were conducted this summer. This included the fabrication of three mesh frames: two for upstream flow turbulence grids and one for downstream outtake protection. Primary focus was placed on long term use and durability, while also taking into account the necessity to accommodate different flume configurations and anti-scratch restrictions for installation.
Mehmet Topal ’26 (computer engineering, economics) presented the research “Open-source, low-cost, motorised light sheet sample holder.” The research was advised by Thomas J. Watson Sr. Professor Kimani Toussaint, and supported by a SPRINT|UTRA award.
Light sheet microscopy (LSM) has become a powerful imaging modality in biomedical research due to its ability to produce high-resolution, volumetric data while minimizing photodamage and photobleaching. However, current commercial light sheet systems are often prohibitively expensive and inflexible, limiting accessibility for researchers. We identified critical limitations in the current sample holder solutions, which either relied entirely on manual positioning or were a part of a full-sized microscope with a custom light generation and capture setup, and could not adapt to existing lab setups. To address this, we designed an open-source, low-cost sample manipulation platform compatible with standard optical breadboards and the ability to integrate into diverse imaging setups tailored for light sheet imaging.
“Bio-Based Co-Surfactants for Enhanced Removal of Short-Chain PFAS by Foam Fractionation” was presented by Oren Van Allen ’26 (design engineering). Van Allen was advised by Professor Kurt Pennell, and served as Undergraduate Researcher in the Pennell Lab this summer.
Per- and polyfluoroalkyl substances (PFAS) are a class of persistent environmental contaminants linked to cancer, immune dysfunction, and other adverse health effects. Foam fractionation has emerged as a promising, low-energy technique to treat PFAS-contaminated water by leveraging their tendency to accumulate at the air-water interface. However, the removal efficiency for short-chain length PFAS, such as perfluorobutanoic acid (PFBA) is limited by their lower surface activity compared to longer-chain length PFAS. Cationic surfactants such as cetyltrimethylammonium bromide (CTAB) have been shown to improve short-chain PFAS removal by creating smaller, more stable bubbles and promoting electrostatic attraction between their positively charged head groups and anionic PFAS. However, concerns over the toxicity and low biodegradability of CTAB hinder their practical use. The objective of this research was to identify and test alternative co-surfactants derived from natural ingredients, as potential replacements for CTAB in foam fractionation systems.
Alex Wang ’27 (computer engineering) presented research on “The Configuration-Driven Impact of Downwash on Dynamic Quadrotor Interaction.” Wang was advised by Professor Kenny Breuer and Ph.D. student Anoop Kiran, and supported by a SPRINT|UTRA.
Quadrotors have experienced a surge in recent years, particularly in applications related to search-and-rescue operations, safety inspections, autonomous product delivery, and 3D mapping. These applications rely on an understanding of aerodynamic interactions that influence the allowed proximity of aerial vehicles near other vehicles and objects, and an accurate understanding will ultimately lead to safer and more efficient autonomous flight. Downwash generated by one drone may impact the stability of others. We seek to characterize its effects using empirical measurements of (1) aerodynamic interactions based on forces and moments, (2) efficiency based on power requirements, and (3) acoustic signals based on microphone data.
“Non-Newtonian Droplet Spreading on Hydrophobic Surfaces: Effects of Shear Thinning and Viscoelasticity” was the poster presentation of Dongyue Wang ’26 (engineering physics). Wang was supported by a SPRINT|UTRA, and advised by Royce Family Professor of Teaching Excellence Roberto Zenit.
The dynamics of liquid droplet impact and spreading on surfaces play a critical role in a wide array of applications, ranging from inkjet printing to firefighting. In this study, we investigate how fluid rheology—specifically shear thinning and viscoelasticity—affects the droplet behavior upon impact with a moderately hydrophobic surface.
Rachael Yuan ’27 (biomedical engineering) presented the research “Lignin Based Hydrogel Microneedles for Chronic Wound Healing.” Yuan was supported by a SPRINT|UTRA, and advised by Elaine I. Savage Professor of Engineering Anita Shukla.
Chronic wounds are a significant barrier to recovery and create a high financial burden on the U.S. healthcare system. As the population ages, the threat of developing chronic wounds from noncommunicable diseases like diabetes will only increase, necessitating improvements to the current wound care strategies. Microneedle (MN) patches are of increasing interest for wound healing applications because of their ability to physically disrupt nonviable tissue and bacterial biofilms commonly found in these wounds, while being painless. However, commonly used MN formulations composed of materials like polylactic acid and poly(ethylene glycol) are unable to balance flexibility with mechanical strength, while maintaining biocompatibility. We examined the use of lignin, a common polymer extracted from plants, for use in the development of MN patches for wound healing applications. Existing literature has found lignin-based hydrogels to support skin wound healing and reduce bacterial adherence. However, there has been limited focus on using lignin hydrogels in MN patch fabrication.
“Computational Model to Characterize Microneedle Mechanics During Fracture on Porcine Skin With and Without Biofilm Formation” was presented by Maxwell Zhang ’27 (applied math, computer science). Zhang was advised by Elaine I. Savage Professor of Engineering Anita Shukla and supported by a SPRINT|UTRA award.
Chronic wounds are a major concern for public health. Biofilms are a common characteristic of chronic wounds. Bacterial biofilms often develop in skin wounds in a three-dimensional community where the bacteria are embedded in self-produced polymers known as the extracellular polymeric substance (EPS). Biofilms evade antibacterial treatments and impede recovery. The structure of the EPS increases the difficulty of eradicating biofilm infections due to poor antibiotic penetration, thus generating increasing interest in the field for topical treatments such as microneedles (MNs). MNs pose a potential solution to biofilm eradication by physically penetrating the EPS, but further studies are required to understand how MNs can penetrate the biofilm located on skin. There has been extensive work on modeling both the mechanical characteristics of the skin and the mechanical characteristics of biofilm in isolation, but there is room for new research in the combination of these areas, capturing the mechanical characteristics of skin affected during the progression of bacterial biofilm growth in response to MN puncture. The objective of this project was to derive a finite element analysis (FEA) model that is complemented with an experimental model that addresses this gap in the literature.
Professor Kenny Breuer also advised Jeremy Bergman from the Rose-Hulman Institute of Technology as part of the University's Leadership Alliance through its Summer Research – Early Identification Program, which helps guide outstanding students to Brown graduate programs. Bergman’s research was titled, “Engineering Design and Analysis of a Biomimetic Flapping Robot for Fluid Dynamics and Avian Flight Studies.”
Birds achieve both forward propulsion and upwards lift during flight by flapping their wings. The study of animal flight and aerodynamics is complex, involving a deep understanding of the intersection of physics and biology. In this project, we investigate flight variables such as wingbeat amplitude and frequency and their impact on fluid dynamics.
For more photos of Engineering undergraduate presenters and their advisers, see our Flickr album.