Maz Abulnaga is a PhD candidate whose work is changing the way researchers and physicians study the placenta with the goal of improving health outcomes for mothers and babies. Specifically, Maz develops algorithms for studying the placental shape in fetal magnetic resonance imaging, an imaging modality that is being used to track the health and function of the placenta, identify pathology, and support patient outcomes. Maz has made several contributions to the theoretical and algorithmic foundations of volumetric geometry processing and has successfully applied his models to clinical research of the placenta. With the support of his MathWorks Fellowship, Maz will develop geometry processing and machine-learning algorithms to quantify placental shape and functional changes throughout pregnancy, which are necessary to identify pathology and plan pregnancy outcomes. He has recently developed a novel approach to provide a standardized representation of placental shape that is being used by researchers globally. He envisions developing a set of open-source tools to enable clinical studies to improve our understanding of placental and fetal development with the goal of revolutionizing fetal-maternal health.

Bernardo is a PhD student in mechanical engineering whose research examines the synthesis of robust strategies for grasping and dexterous manipulation. Specifically, he is examining the problem of caging, a geometric property by which a set of fingers manipulate an object by trapping it rather than immobilizing it. He has developed an explicit optimization approach to formulate the caging condition as a convex mixed-integer optimization problem. This was the first of its kind, since caging had previously been approached as either a topological or computational geometry problem. More recently, he has extended the formulation to synthesizing grasping strategies with certificates of correctness, a problem that is of high interest today both in academia and industry. He uses MATLAB’s Robotics Toolbox, Optimization Toolbox, and Deep Learning Toolbox as the main prototyping interface. In addition, his work has resulted in several tools that have been made open to the robotics and MathWorks communities, including ABB ROS Interface, Contact-TrajectoryOptimization Models, and Certified Grasping Toolbox. Bernando earned a BSc in electronics engineering from Universidad Simón Bolívar, Venezuela

Shashank is a PhD student in mechanical engineering whose research primarily focuses on the development of reduced-order, rate-dependent methods for a variety of granular intrusion problems, such as meteorite impacts and animal and vehicular locomotion in sands and deserts (a field referred to as terramechanics). He uses large-scale, detailed numerical simulations to understand the physics of such scenarios, which in turn allows him to develop robust reduced-order models that can be run in real time. He uses MATLAB for various purposes in his research, most importantly for the implementation of reduced-order models for granular intrusion, allowing for the real-time simulation of diverse intrusion scenarios with different system properties and intruder shapes. Shashank earned a BTech in mechanical engineering fromIITGandhinagar, India and an SM in mechanical engineering from MIT.He also worked as a scientist at the Defense Research and Development Organization, India before joining MIT

Mohammad is a PhD student in civil and environmental engineering whose research is focused on explaining and predicting biodiversity changes using mathematical models. Specifically, his goal is to provide closed-form solutions using computational methods in order to skip the impossible task of simulating all possible parameter values and combinations and to establish rigorous and testable estimates about the probability of persistence of species. He is developing his own code and integrating libraries from MATLAB with the goal of generating a computational package that will allow researchers to work with different ecological models, derive the range of parameter values (or combinations of parameter values) compatible with the persistence of a given species in an ecological system, and estimate the probabilities (or risks) of species extinctions or species invasions. Mohammad earned a BS in mathematics and a BS in electrical engineering from Rensselaer Polytechnic Institute and an SM in civil and environmental engineering, an SM in electrical engineering and computer science, and an SM in computation for design and optimization from MIT.

Pasquale is a PhD student in aerospace engineering. He is developing general-purpose tools for estimation that are robust to corrupted information and adversarial attacks, run in real-time, and require minimal tuning. His interests include safe and trustworthy perception with application to single and multi-robot autonomous systems. He uses MATLAB on a regular basis, including the Computer Vision Toolbox and Robotics System Toolbox for his work on robust estimation. He earned his SB in computer science from University of Pisa, Italy, an SM in embedded computing systems from Scuola Superiore Sant’Anna of Pisa, Italy and an SM in aeronautics and astronautics from MIT. Prior to MIT, he was a research scientist at the United Technologies Research Center, Ireland.

Nicolas is a PhD student in electrical engineering and computer science whose research focuses on medical imaging technology. He is trying to solve challenging MRI problems by creating and distributing the hardware and computational tools needed for simultaneous optimization of coil geometries, encoding field patterns, and received-signal reconstruction algorithms. He has observed that the steps, or modules, that make up most MRI sequences often have conflicting patient-specific field pattern requirements. Nicolas developed a general MATLAB-based framework for use at scan time that optimizes the field patterns for each sequence module. The framework reads patient rapid-scan data from the imager, exploits the coil pre-characterization and convex formulations to efficiently compute sets of sequence-module optimized coil currents, and then transfers the coil currents sets, along with triggering conditions, to the coil driver array. Nicolas earned an SB in electrical engineering from MIT.

Toros Arikan is a PhD candidate whose interdisciplinary work integrates signal processing, machine learning, and computational imaging to significantly expand the capabilities of sensing systems for acoustic, radio frequency (RF), and optical modalities. With the support of a MathWorks Fellowship, Toros will pursue innovative research that takes a new, and potentially highly useful, view of system design. Traditionally, system designs are based on relatively low-dimensional models. Toros’s work, by contrast, embraces high-dimensional models and finds potential in that complexity for better performance and surprising new capabilities. One example is the presence of unknown complex obstacles and reflecting structure in physical environments that can improve object localization and scene imaging performance. MATLAB tools play a central role in Toros’s research, serving as his algorithm prototyping and experimental data analysis environment. The tools, methods, and insights emerging from this work hold important potential for future research and applications, including the acoustic and RF communities. To help realize these new possibilities, Toros will make the resulting tools of his research widely available through the MathWorks File Exchange.

Maanasa Bhat is a PhD candidate who is integrating interests in health, energy, and materials science to develop innovative new approaches to synthesis of nanomaterials. As a MathWorks Fellow, Maanasa is designing a flame-based synthesis process to manufacture nanomaterials for low-cost sensors and photocatalysts with a wide range of potential environmental and health applications. The goal of Maanasa’s work is to build a broader understanding of the physical and chemical mechanisms involved in the flame synthesis process. The aim is to strengthen connections between the input operating conditions and the composition and morphology of output products to achieve high controllability. MathWorks tools have been indispensable in this research. As her project progresses, Maanasa hopes to tailor the novel flame synthesis process she has developed for specific uses such as the detection of methane leaks, pollutant emissions in air, as well as breath diagnosis for hard-to-detect conditions such as endometriosis and fibroids.

Harsh G. Bhundiya is a PhD candidate who is applying structural engineering, computational modeling, and design techniques to purse his interest in space structures. His work is ushering in a new paradigm for constructing large structures in orbit, called in-space manufacturing, with an enormous potential impact on space applications. Drawing on MathWorks tools, Harsh has reached exciting milestones, including the development of a path-planning framework called Bend-Forming, which enables the creation of paraboloids, isogrid columns, and other structural elements that can be linked together to form large structures such as RF reflectors and solar sails. As a MathWorks Fellow, Harsh will characterize the compression behaviors of these structures and study the effects of processing defects on their buckling response through experiments and nonlinear finite element analysis. Simulink and Simscape Multibody will play critical roles in this work. Harsh was pleased to share previous work on MathWorks File Exchange, and will continue to share his research, as he seeks to create a new generation of large space structures for diverse applications.

A PhD student in electrical engineering and computer science, Roberto is a member of the Organic and Nanostructured Electronics Lab. He uses MATLAB’s versatile Partial Differential Equation (PDE), Global Optimization, and ParallelComputingToolboxes to solve, model, and fit the group’s experimental data sets on carrier recombination and diffusion in semiconductors for optoelectronic applications. Leveraging the PDE Toolbox, his most recent work, “AccurateDetermination of Semiconductor Diffusion Coefficient Using Confocal Microscopy,” explores common pitfalls in modeling charge carrier diffusion in semiconductors that were elucidated by simulations carried out using the PDE toolbox. He is now focused on modeling semiconductors for solar applications that present anisotropic diffusion and complex boundaries using the PDE Toolbox coupled with the Parallel Computing and the Global Optimization Toolboxes to optimize multi-variable minimization problems. The Parallel Computing Toolbox has been essential in bringing the computation times down to manageable time frames, allowing him to explore multiple schemes that could represent the complex carrier diffusion anisotropy that is observed. He earned an SB in electrical engineering and physics fromMIT.

Peng Cao is a PhD candidate who has demonstrated innovative use of MathWorks tools to make fundamental contributions to machine learning and digital health. Specifically, Peng’s research focuses on sensing systems, data analysis, and radio signals. MATLAB is an essential tool in this work, for numerical optimization, complex signal processing, and analysis and manipulation of radio signals and electromagnetic fields. The MathWorks Fellowship will support Peng’s cutting-edge research in several domains. She has developed new approaches to learning from crowdsourcing labels, including robust information-theoretic loss functions that enable learning from multiple labels and noisy labels, as well as semi-supervised, multimodal learning. Peng has also made notable advances in human-motion modeling with a transformer-based deep learning model for generative modeling of 3-D human motion. Another project yielded a new approach to enable robots to move quickly in indoor environments without colliding with people—one of the first such projects to demonstrate indoor localization around corners using radio signals. Most recently, Peng developed the first contactless system to monitor blood oxygen by employing a neural network to assess oxygen levels within 2 to 3 degrees. This work, which was validated with Covid-19 patients, has enormous potential for telehealth applications.

Yifan (Henry) Cao is a PhD candidate working at the frontier of a new materials science field: high- entropy alloys (HEAs). HEAs offer valuable traits, such as high strength and fracture resistance. As a MathWorks Fellow, Henry will be expanding a new method he has developed to optimize machine learning interatomic potentials (MLP) in cross-scale atomistic simulations, thereby predicting HEA
properties with greater efficiency than current methods. Although HEAs are created by mixing many chemical elements, local chemical ordering (LCO), the prominent mechanism underlying many of these properties, remains poorly understood. Henry’s research seeks to close that knowledge gap. Two MathWorks tools will play a central role in the project: Simulink, which will enable the development of a well-rounded MLP training set, and MATLAB, which will be used to extract physically important features and plot relevant data obtained from HEA cross-scale simulations. Henry also plans to explore potential applications of these tools in mesoscale, atomic, and quantum mechanics modeling approaches.

Katherine is a master’s student in aeronautical and astronautical engineering and member of the Engineering SystemsLaboratory. A longtime MATLAB and Simulink user, her current research involves population modeling and simulation of pedestrian activity through transit environments. She is implementing agent-based modeling and simulation in order to capture the emergent phenomena of population groups moving across complex sites, both here on Earth and on Mars. She has used MATLAB to numerically simulate pedestrian flows in hallways using a social-force model and is currently developing a Java-based integrated, flexible platform to simulate agent behavior across multiple agent pools and simulation scenarios. Katherine earned a BS in aerospace engineering fromUniversity of Illinois.

Florian Chavagnat is a PhD candidate whose research sits at the boundary of applied physics and computational methods. Specifically, he is developing physics-based models that represent the complex nature of boiling heat transfer with the objective of supporting NASA’s potential moon and Mars explorations. As a MathWorks Fellow, Florian will advance his cutting-edge study of the management and transfer of cryogenic fuel. His achievements include: the development of a novel method to deliver experimental diagnostics for cryogenic boiling; a low-cost approach to demonstrate the applicability of boiling diagnostics; and the creation of a flight-ready apparatus to study the influence of reduced gravity on microscale boiling parameters. MATLAB is central to Florian’s research, enabling him to postprocess and analyze data and optimize the formulation of the constitutive relations using the Curve Fitting Toolbox, Global Optimization Toolbox, and Statistics and Machine Learning Toolbox. In addition to providing NASA with a first generation of closure models, Florian’s ongoing research is generating new data in a broad range of conditions and advancing the fundamental understanding of the boiling mechanism.

Cécile is a PhD student in materials science and engineering. Her research is focused on the creation of new methods for manufacturing high-performance polymer and composite materials in a rapid, energy-efficient manner. Specifically, she studies interfacial polymerization (IP), a process by which a polymer is formed at the interface between two immiscible liquids (often water and an organic solvent), each containing one type of reactive species (initiator or monomer). She uses MATLAB to enable experimental data analysis and prediction of IP-based process kinetics, as well as future perspective on the integration of MATLAB algorithms in manufacturing and 3D printing. Cécile earned an MS in materials science and engineering from MINES ParisTech, France.

Justin Chen is a PhD candidate whose research explores the intersection of algorithms, machine learning, and data analysis. Specifically, Justin is making important contributions in the emerging area of learning-augmented algorithms. As a MathWorks Fellow, Justin will build upon several recent projects in this domain, including research on counting triangles in a data stream (a fundamental tool of network analytics); the classic optimization problem of online bipartite matching; and offline algorithms for fundamental graph problems. His interests also include new applications of algorithms for data science problems. In these and other projects, Justin is making innovative use of MATLAB to develop powerful new algorithms that improve speed and efficiency, with possible applications spanning from Google’s ad market to the kidney exchange program. His work has also yielded broader benefits, such as a new framework of reductions for learning-based algorithms and major progress in an open question posed by Sealfon (related to differentially private computation of shortest graph paths) that enhances his own work and holds the potential to advance many paths of discovery.

Jaehun Choe is a PhD candidate whose research bridges the fields of materials science, mechanics, and robotics to advance cutting-edge medical robotics applications. As a MathWorks Fellow, Jaehun is developing a guidewire robot capable of navigating blood vessels in the brain to treat stroke patients remotely and to deliver life-saving care faster. His project builds on recent advances in soft, thread-like robots made of ferromagnetic material that can navigate a life- sized model of vasculature with the enhanced mobility required for endovascular surgery. Using MathWorks tools like MATLAB and Simulink, Jaehun is developing a semi-autonomous control system that uses 3-D positional data of vessels to calculate the directions of desired branches, and suggests optimal position and orientation of the robot. His goal is to develop a fully autonomous system, and to share custom MATLAB codes underlying his system. Jaehun’s research is yielding important contributions to medical robotics research and has the potential to bring life-saving treatment to millions of stroke victims.

Ximo is a PhD student in aeronautics and astronautics conducting research at the intersection of numerical modeling and physics of nanoscale engineering devices. A frequent MATLAB user, he is working on a problem related to the numerical analysis of the shapes of small liquid menisci stressed under very high electric fields to evaporate ions. As these ions fly away, they produce thrust, hence a rocket can be constructed under this principle. This is difficult to analyze since the scales change dramatically from millimeters to nanometers, which creates a set of numerical challenges as equations describing the balance of electric, surface tension, and hydraulic stresses, including thermal loads that need to be solved simultaneously on meshes with very strong dimensional gradients. He earned a BS in aeronautics and astronautics and physics from Universitat Politècnica de Catalunya, Spain

Clement is a PhD student in aeronautics and astronautics working on aviation automation problems at the MITInternational Center for Air Transportation. His PhD topic involves investigating high-level aviation automation in dynamic and stochastically varying environments. In particular, he is focusing on automated situation awareness and decision making with regards to airborne trajectory prediction. He is investigating data analytics approaches, usingMATLAB tools, for mining large sets of aircraft surveillance (ADS-B) data to infer aircraft behavior in structured and unstructured airspaces around airports. Clement earned a BS in mechanical and aerospace engineering from Princeton University and an SM in aeronautics and astronautics from MIT.

Benjamin Dacus is a PhD candidate who is exploring how various materials respond in a fusion reactor environment at micro and macro levels. The goal of his work is to identify or fabricate materials well-suited to hostile conditions, from currently available metals to state-of-the-art radiation tolerant materials. With the support of a MathWorks Fellowship, Ben will expand his contributions to a world-first in situ ion irradiation transient grating spectroscopy (I3TGS) sys- tem. MATLAB-based analysis tools—deployed and/or created by Ben—play a fundamental role in the I3TGS system. This includes non-linear fitting and curve-fitting capabilities, enabling data acquisition one million times faster than previous capabilities, with one thousand times the data throughput rate. This tool led to key observations and discoveries, including the precipitation of radiation-induced secondary phases as they grow on the macroscale. His next goal is to augment the I3TGS system to replicate the conditions of the first commercial fusion power reactors. By integrating cutting-edge hardware with powerful MATLAB tools, Ben’s research provides a valuable new technique to test and develop materials in a wide range of hostile environments and conditions.