Our paper on ``Federated learning of predictive models from federated Electronic Health Records''...
Dean’s Catalyst Awards Energize Early-Stage Projects
By Mark Dwortzan
College of Engineering research projects in nanofluidics, integrated circuit design, data-driven healthcare management and Terahertz radiation generation are set to take off, thanks to the Dean’s Catalyst Awards (DCA) grant program. This year four research teams will each receive up to $50,000 in DCA funding to develop novel techniques to investigate these topics.
Established by Dean Kenneth R. Lutchen in 2007 and organized by a faculty committee, the annual Dean’s Catalyst Awards program encourages early-stage, innovative, interdisciplinary projects that could spark new advances in a variety of engineering fields. By providing each project with seed funding, the awards give full-time faculty the opportunity to generate initial proof-of-concept results that could help secure external funding.
This year’s DCA-winning projects promise to improve the nation’s healthcare, defense, information and communications systems.
Leveraging their respective skills in computational modeling of nanomaterials and fabrication/testing of nano-electromechanical systems, Assistant Professor Harold Park and Associate Professor Kamil Ekinci (both ME, MSE) will study the ability of graphene, the world’s only known 2D material, to increase fluid flow through nanoscale channels. By utilizing the hydrophobic properties of the world’s thinnest known material (graphene is composed of a single layer of carbon atoms) to significantly enhance the fluid flow rate in confined, nanoscale channels, the researchers aim to improve the performance of nanofluidic devices and lab-on-a-chip technologies. Potential application areas include biological and chemical sensing, homeland security and healthcare.
Combining low-power circuit and micro-architecture design techniques with ideas from coding and information theory, ECE Assistant Professors Ajay Joshi and Bobak Nazer plan to develop a new approach to digital VLSI (very large scale integrated circuit) design that tolerates errors at the device level while maintaining global reliability at the architectural level. Rather than paying the ever-increasing energy and area costs required to keep all devices well-behaved, the researchers intend to allow individual transistors to be “noisy.” The goal is to develop an architecture that exploits feedback and redundancy to maintain nearly the same end-to-end reliability and performance as today’s designs while consuming much less energy.
Professor Ioannis Paschalidis (ECE, SE), Daniel Newman, BUMC assistant professor of medicine and Boston Medical Center (BMC) chief medical information officer, and Shiby Thomas, BMC director of enterprise analytics, will pursue a comprehensive and systematic approach to intelligently processing Electronic Health Records (EHRs) and directing physician attention to preventing serious medical conditions. Using algorithms that assess patients for disease risk and trigger physician actions based on their risk classification, and wireless body sensors that dispatch medical information to the clinic in near real-time, the researchers’ proposed system could significantly reduce costs and improve efficiencies in the U.S. health care system.
Despite major application opportunities in medical and chemical imaging and security screening, there are few sources that emit electromagnetic radiation in the Terahertz (THz) range (0.3-10 trillion Hz). Combining their respective expertise in graphene fabrication and THz radiation studies, Professors Anna Swan and Roberto Paiella (both ECE, MSE) propose an entirely novel way of generating THz radiation that, unlike current state-of-the-art methods, is not limited by intrinsic material properties. They plan to develop a “tabletop cyclotron source” of THz light that exploits the excellent electron transport properties of graphene to produce radiation in the Terahertz range.
“The Dean’s Catalyst Award is an ideal platform for exploring novel ideas that are promising, but need more probing to understand their true potential,” said Joshi. “We are grateful for this opportunity to refine our ideas and develop preliminary results before applying for external funding.”