Spring 2019 TEER Talks

Roarke Horstmeyer joined Duke on July 1, 2018, as an assistant professor of biomedical engineering.

Dr. Horstmeyer is interested in exploring new ways to capture and process biomedical images. He develops microscopes, cameras and computer algorithms for a wide range of applications, from forming 3D reconstructions of organisms to detecting neurons deep within tissue. His work lies at the intersection of optics, biology, signal processing and optimization.

A doctoral graduate of Caltech’s electrical engineering department, Dr. Horstmeyer was most recently an Einstein International Postdoctoral Fellow with the Judkewitz lab at Charitè Medical School in Berlin, where he worked to apply digital optical phase conjugation in neuroscience.  He earned a master of science from the MIT Media Lab and bachelor’s degrees in physics and Japanese from Duke University.

Deep learning algorithms offer a powerful means to automatically analyze the content of medical images. However, many biological samples of interest are difficult to resolve with a standard optical microscope. Here, we use a convolutional neural network (CNN) not only to interpret images, but also to re-design the physical layout of the microscope itself, which helps us optimize what it can capture. To achieve this goal, we merge an optical model of image formation into the pipeline of a CNN, which allows us to simultaneously determine an ideal illumination and lens arrangement to highlight important sample features during image acquisition, along with a set of convolutional weights to process the detected images post-capture. We explore how this paradigm can enable new optical designs for gigapixel-scale imaging of biological specimens.

Prior to arriving at Duke, I was a postdoctoral research associate in the program in Atmospheric and Oceanic Sciences at Princeton University and had a dual appointment as a visiting research scientist at the NOAA Geophysical Fluid Dynamics Laboratory. I obtained my undergraduate degree at U.C. Berkeley where I received a Bachelor of Arts in Atmospheric Sciences and Applied Mathematics. For my graduate studies, I attended Princeton University where I completed a Ph.D. in Hydrology in the department of Civil and Environmental Engineering.

My research harnesses the existing petabytes of global environmental data to improve understanding of the terrestrial water cycle. More specifically, I focus on quantifying and uncovering the role of multi-scale spatial organization over land (i.e., heterogeneity) in the Earth system. To this end, my group's research has three overarching themes: 1) improve the representation of land heterogeneity in Earth system models, 2) harness environmental data to characterize the observed spatial patterns and features over land, and 3) quantify the sensitivity of the hydrologic cycle to spatial heterogeneity. The tools that my group uses include numerical modeling, satellite remote sensing, machine learning, and high performance computing.

I am currently looking for highly motivated Ph.D. and postdocs. If the research themes of my group are of interest to you, please don't hesitate to email me.

The deluge of environmental data from satellites, observations networks, and computer models has triggered an ever-increasing lag between the availability of information and the subsequent analysis and use of these data. The underuse of these exabytes of data is a missed opportunity given their large potential to provide solutions to pressing environmental challenges such as quantifying and predicting the water cycle. In this presentation, I will illustrate how high performance computing and machine learning provide an answer to this challenge by making it possible to harness these exabytes of information. More specifically, I will focus on how analysis of these data is providing a critical breakthrough to understand and quantify the environmental characteristics that drive hydrologic processes across spatial scales—a persistent limitation in predicting the water cycle over the globe. I will discuss how these breakthroughs have the potential in the coming decades to transform water resource management, precision agriculture, flood control, and drought mitigation.

The Musah Lab is interested in understanding how molecular signals and biophysical forces can function either synergistically or independently to guide organ development and physiology, and how these processes can be therapeutically harnessed to treat human disease. Given the escalating medical crisis in nephrology as growing number of patients suffer from kidney disease that can lead to organ failure, the Musah Lab focuses on engineering stem cell fate for applications in human kidney disease, extra-renal complications, and therapeutic development. Dr. Musah’s research interests include stem cell biology and regenerative medicine, molecular and cellular basis of human organ development and disease progression, organ engineering, patient-specific disease models, biomarker identification, therapeutic discovery, tissue and organ transplantation, microphysiological systems including Organ Chips (organs-on-chips) and organoids, matrix biology, mechanotransduction and disease biophysics.

A prime interest in our research is to understand how molecular signals and biophysical forces function either synergistically or independently to guide organ development and physiology, and how these processes can be therapeutically harnessed to treat human disease. Given the escalating medical crisis in nephrology as growing number of patients suffer from kidney disease that lead to organ failure, our research group focus on integrating stem cell biology with engineering principles and synthetic chemistry techniques to uncover the mechanisms of human kidney development and function. We aim to extend these studies to engineer functional in vitro models to facilitate the development of novel therapeutic modalities for human kidney disease.

Dr. Maria Gorlatova's research is focused on reaching the next level of adaptive intelligent behavior in Internet of Things systems and applications. This research involves the development of architectures, algorithms, and protocols for emerging pervasive systems. It crosses traditional discipline boundaries and requires thinking across multiple layers of system and protocol stacks.

Dr. Gorlatova earned her Ph.D. in Electrical Engineering from Columbia University, and her M.Sc. and B.Sc. (Summa Cum Laude) degrees in Electrical Engineering from University of Ottawa, Canada. She has several years of industry experience, where she had been affiliated with Telcordia Technologies, IBM, and D. E. Shaw Research. She came to Duke from Princeton University, where she held the positions of an Associate Research Scholar in the Electrical Engineering Department and an Associate Director of the Princeton EDGE Lab.

Dr. Gorlatova is a recipient of the Google Anita Borg USA Fellowship, Canadian Graduate Scholar CGS NSERC Fellowships, the Columbia University Presidential Fellowship, and the Columbia University Jury Award for Outstanding Achievement in Communications. She is a co-recipient of the ACM SenSys Best Student Demonstration Award, the IEEE Communications Society Young Author Best Paper Award, and the IEEE Communications Society Award for Advances in Communications.

Recent advances in edge computing, the movement of computing resources away from the datacenters and closer to the end users, are enabling the next level of responsiveness, adaptiveness, and intelligence in mobile systems and Internet of Things deployments. This talk describes our ongoing efforts in bringing intelligence to the edge, and taking advantage of it.

Professor Tatiana Segura received her BS degree in Bioengineering from the University of California Berkeley and her doctorate in Chemical Engineering from Northwestern University. Her graduate work in designing and understanding non-viral gene delivery from hydrogel scaffolds was supervised by Prof. Lonnie Shea. She pursued post-doctoral training at the Swiss Federal Institute of Technology, Lausanne under the guidance of Prof. Jeffrey Hubbell, where her focus was self-assembled polymer systems for gene and drug delivery. Professor Segura's Laboratory studies the use of materials for minimally invasive in situ tissue repair.   On this topic, she has published over 60 peered reviewed publications. She has been recognized with the Outstanding Young Investigator Award from the American Society of Gene and Cell Therapy, the American Heart Association National Scientist Development Grant, and the CAREER award from National Science Foundation.  She was Elected to the College of Fellows at the American Institute for Medical and Biological Engineers (AIMBE) in 2017. She spent the first 11 years of her career at UCLA department of Chemical and Biomolecular Engineering and has recently relocated to Duke University, where she holds appointments in Biomedical Engineering, Neurology and Dermatology.

The Segura laboratory focuses on the design of materials to promote endogenous tissue repair. In this approach, stem cells are not delivered to the injured or deceased tissue, rather a material is delivered, which induces endogenous stem cells to repair the lost tissue. In other words, their approach uses materials as drugs. Since no stem cells are delivered, the material must converse with the damaged tissue and encourage local stem cells to regenerate the injured or lost tissue rather than scar. The materials are often made from naturally occurring polymers and are processed to mimic the repairing environment not the normal environment. The Segura laboratory has found that injectable porous materials that can take the shape of the wound are in general better able to repair tissue than non-porous injectable materials. Thus, understanding the influence of microporosity (%void space and interconnectivity), the type of porosity (periodic structure vs. random structure), the curvature of the individual pores (convex versus concave), the biochemical composition of the pores (bioactive or inert), and the feature size (larger than a cell or smaller than a cell) on tissue regeneration is of key importance. These features are studied in silico with mathematical models of local void geometry, in vitro with model cells to understand the cell material interaction and how materials can drive the interaction, and in vivo with animal models disease states, understanding how materials can influence endogenous regenerative processes. Using these approaches, the Segura lab has demonstrated that even the brain can repair itself, an organ often thought not be regenerative.

Dr. Brinson's research interests include the study of advanced material systems and developing new methods to characterize and to model material behavior. Her lab’s research objective is to characterize and model advanced materials systems, at scales spanning the range of molecular interactions, micromechanical and macroscopic behavior.

Data-driven approaches allow data collection, sharing and analysis to elucidate the processing-structure-property relationships and facilitates the fast deployment of advanced materials. The complexity of polymer behavior, including high sensitivity of material microstructure and bulk properties to small changes in processing conditions, chemistry of constituents as well as unknown properties “interphase polymer” near particles, leads to increased need for data approaches. In this work, we present NanoMine, a data-driven web-based platform for polymer nanocomposite materials and demonstrate several characterization and design case studies using machine learning methods on top of this data platform. Nanomine consists a combination of databases, search and visualization functions, data-driven analysis tools and physics-based simulation tools for polymer nanocomposite analysis and design. The NanoMine framework is being populated with curated literature data on composition, processing, structure and properties of nano composite materials to help researchers search, visualize, and compare aspects of nano composites, enabling designer performance materials via a suite of data, analytic and physical computational software tools.

Dr. Dunn's research interests include developing new tools and infrastructure for multi-modal biomedical data integration to drive precision/personalized methods for early detection, intervention, and prevention of disease.

Recent technological advancements make it possible to closely and continuously monitor individuals on multiple scales in real time while also incorporating genetic, environmental, and lifestyle information. We are collecting and using this multi-scale biomedical data to gain a more precise understanding of health and disease at molecular and physiological levels and developing actionable, predictive health models for improving cardiometabolic outcomes. We are simultaneously developing tools for the digital health community, including the Digital Biomarker Discovery Pipeline (DBDP), to facilitate the use of mobile device data in healthcare.

Po-Chun Hsu’s research group aims to develop innovative materials for light and heat management. With the application and desired functions in mind, we design, synthesize, and fabricate the materials and devices with ideal photonic structure, chemical properties, or heat transfer characteristics. Focus areas include smart textiles, photonic fibers, solar desalination, and solid-state cooling.

He received his Ph.D. degree in Materials Science and Engineering from Stanford University in 2016 and B.S. also in Materials Science and Engineering from National Tsing Hua University in 2007. His PhD works involve radiative heating/cooling textiles, electrochromic devices, nanofiber electrospinning, and metal nanowire transparent electrodes. During 2016-2018, he was a postdoctoral researcher in Mechanical Engineering at Stanford University, focusing on electrocaloric cooling and thermal properties study of van der Waals heterostructure materials. Having the training in both materials science and heat transfer and participated in a wide range of projects, Dr. Hsu embraces interdisciplinary, multiscale, and solution-oriented research that can benefit humanity.

Maintaining thermal comfort is one of the most important basic needs for living. Extreme ambient temperature and temperature variation not only hurt work productivity but also can cause health problem directly or indirectly. Space temperature control is so indispensable, that we spend a huge amount of energy only to keep the indoor temperature stable and comfortable — in the US, 12% of total energy is used for space heating and cooling by heaters, air conditioners, and heat pumps. These technologies bring us convenience, but at the cost of greenhouse gas emission and climate change. In this talk, I will introduce the concept of personal thermal management which focuses the thermal environment around the human body rather than the entire building. By re-engineering the textiles we wear using the knowledge of materials science, photonics, and radiative heat transfer, we can achieve better thermal management and other smart textile functionalities to reduce the energy demand for building temperature control.

Litchinitser holds a Ph.D. Electrical Engineering from the Illinois Institute of Technology. Her primary focus is on metamaterials that manipulate the visible portion of the electromagnetic spectrum. Litchinitser began her work with metamaterials as a research scientist at the University of Michigan, and joined the faculty at the University of Buffalo in 2008. Over the next decade, she became one of the leading experts in optical metamaterials. Currently, Litchinitser’s research focuses on topological photonics, which seek to direct light around tight corners using tiny waveguides that prevent photons from scattering. She is Fellow of the American Physical Society (APS), a Fellow of the Optical Society of America (OSA), and a Senior Member of the Institute of Electrical and Electronics Engineers (IEEE).

Structured light and structured matter are two fascinating branches of modern optics that recently started having a significant impact on each other. We will discuss fundamental optical phenomena at the interface of singular and nonlinear optics in engineered optical media and show that the unique optical properties of optical nanostructures open unlimited prospects to “engineer” light itself. These studies could find diverse applications in the next generation of integrated optoelectronic devices for optical communications in both quantum and classical regimes, imaging, nanolithography, and spectroscopy.