The National Science Foundation (NSF) awarded a syndicate of eight Southeast universities — with Georgia Tech as the lead — a $15 million grant to support the development of a regional innovation ecosystem that addresses underrepresentation and increases entrepreneurship and technology-oriented workforce development. 

The NSF Innovation Corps (I-Corps) Southeast Hub is a five-year project based on the I-Corps model, which assists academics in moving their research from the lab to the market. 

Led by Georgia Tech’s Office of Commercialization and Enterprise Innovation Institute, the NSF I-Corps Southeast Hub encompasses four states — Georgia, Florida, South Carolina, and Alabama. 

Its member schools include:

• Clemson University 
• Morehouse College 
• University of Alabama 
• University of Central Florida 
• University of Florida 
• University of Miami 
• University of South Florida 

In January 2025, when the NSF I-Corps Southeast Hub officially launches, the consortium of schools will expand to include the University of Puerto Rico. Additionally, through Morehouse College’s activation, Spelman College and the Morehouse School of Medicine will also participate in supporting the project. 

With a combined economic output of more than $3.2 trillion, the NSF I-Corps Southeast Hub region represents more than 11% of the entire U.S. economy. As a region, those states and Puerto Rico have a larger economic output than France, Italy, or Canada. 

“This is a great opportunity for us to engage in regional collaboration to drive innovation across the Southeast to strengthen our regional economy and that of Puerto Rico,” said the Enterprise Innovation Institute’s Nakia Melecio, director of the NSF I-Corps Southeast Hub. As director, Melecio will oversee strategic management, data collection, and overall operations​. 

Additionally, Melecio serves as a national faculty instructor for the NSF I-Corps program. 

“This also allows us to collectively tackle some of the common challenges all four of our states face, especially when it comes to being intentionally inclusive in reaching out to communities that historically haven’t always been invited to participate,” he said. 

That means bringing solutions to market that not only solve problems but are intentional about including researchers from Black and Hispanic-serving institutions, Melecio said. 

Keith McGreggor, director of Georgia Tech’s VentureLab, is the faculty lead charged with designing the curriculum and instruction for the NSF I-Corps Southeast Hub’s partners. 

McGreggor has extensive I-Corps experience. In 2012, Georgia Tech was among the first institutions in the country selected to teach the I-Corps curriculum, which aims to further research commercialization. McGreggor served as the lead instructor for I-Corps-related efforts and led training efforts across the Southeast, as well as for teams in Puerto Rico, Mexico, and the Republic of Ireland. 

Raghupathy “Siva” Sivakumar, Georgia Tech’s vice president of Commercialization and chief commercialization officer, is the project’s principal investigator. 

The NSF I-Corps Southeast Hub is one of three announced by the NSF. The others are in the Northwest and New England regions, led by the University of California, Berkeley, and the Massachusetts Institute of Technology, respectively. The three I-Corps Hubs are part of the NSF’s planned expansion of its National Innovation Network, which now includes 128 colleges and universities across 48 states. 

As designed, the NSF I-Corps Southeast Hub will leverage its partner institutions’ strengths to break down barriers to researchers’ pace of lab-to-market commercialization. 

"Our Hub member institutions have successfully commercialized transformative technologies across critical sectors, including advanced manufacturing, renewable energy, cybersecurity, and biomedical fields,” said Sivakumar. “We aim to achieve two key objectives: first, to establish and expand a scalable model that effectively translates research into viable commercial ventures; and second, to address pressing societal needs.

"This includes not only delivering innovative solutions but also cultivating a diverse pipeline of researchers and innovators, thereby enhancing interest in STEM fields — science, technology, engineering, and mathematics.”

U.S. Rep. Nikema Williams, D-Atlanta, is a proponent of the Hub’s STEM component. 

“As a biology major-turned-congresswoman, I know firsthand that STEM education and research open doors far beyond the lab or classroom.,” Williams said. “This National Science Foundation grant means Georgia Tech will be leading the way in equipping researchers and grad students to turn their discoveries into real-world impact — as innovators, entrepreneurs, and business leaders. 

“I’m especially excited about the partnership with Morehouse College and other minority-serving institutions through this Hub, expanding pathways to innovation and entrepreneurship for historically marginalized communities and creating one more tool to close the racial wealth gap.” 

That STEM aspect, coupled with supporting the growth of a regional ecosystem, will speed commercialization, increase higher education-industry collaborations, and boost the network of diverse entrepreneurs and startup founders, said David Bridges, vice president of the Enterprise Innovation Institute. 

“This multi-university, regional approach is a successful model because it has been proven that bringing a diversity of stakeholders together leads to unique solutions to very difficult problems,” he said. “And while the Southeast faces different challenges that vary from state to state and Puerto Rico has its own needs, they call for a more comprehensive approach to solving them. Adopting a region-oriented focus allows us to understand what these needs are, customize tailored solutions, and keep not just our hub but our nation economically competitive.” 

Researchers have long known about the relationship between cannabis use and mental health. But how that practice has affected prescriptions for drugs to treat mental health disorders has been less clear, until now.

A new study from Georgia Tech’s School of Public Policy, recently published in JAMA Network Open, shows that commercially insured patients living in states with legal cannabis sales filled fewer prescriptions for benzodiazepine-class anti-anxiety drugs, but turned in scripts for antipsychotic and antidepressants at rates higher than residents of states without legal cannabis access.

In one way, the news could be good: benzodiazepines are commonly misused, with sometimes fatal results. But the increase in antipsychotic and antidepressant prescriptions is uncharted territory, said Ashley Bradford, the lead researcher on the study and an assistant professor in the School of Public Policy.

“Does this reflect a social benefit with fewer people feeling anxious, or a social harm with fewer people treating their anxiety effectively and more people experiencing psychosis and depression?” Bradford said. “We can’t say. What we can say is that physicians and patients seem to be responding to cannabis access in clinically meaningful ways.”

The researchers analyzed prescription data from more than 10 million commercially insured patients and five classes of psychotropic drugs – benzodiazepines, antidepressants, antipsychotics, barbiturates, and sleep medications. They then used a synthetic control method to compare prescription fill rates in states with medical and recreational cannabis laws to those without.

 They found that in states where medical cannabis laws were in place, the prescription fill rate for benzodiazepines fell by 12.4% compared to states that did not allow any form of legal marijuana. Legal recreational marijuana caused a bigger drop: 15.2%.

However, in states with medical cannabis laws, the antidepressant prescriptions fill rate increased by 3.8% while fill rates for antipsychotics rose by 2.5%. Recreational cannabis availability resulted in an 8.8% increase in the antidepressant prescription fill rate, according to the study.

The impact of legal cannabis on barbiturates and sleeping medications was insignificant.

"This study suggests that cannabis laws may be significantly associated with the population-level use of prescription drugs to treat mental health disorders, although the associations vary by drug class and state,” the authors wrote in the paper. "Our results suggest that additional research is needed to assess whether changes in dispensing of (mental health drugs) are associated with differences in health care outcomes."

Previous studies focused primarily on the impact of medical and, to a lesser extent, recreational laws on prescription dispensing in the Medicaid and Medicare populations. This work reveals that commercially insured patients seem to respond to legal cannabis access in similar ways to those on Medicare and Medicaid.

The study also demonstrates the impact of different state laws, Bradford said. She said that the results suggest that researchers could identify which aspects of cannabis policies lead to socially optimal outcomes and help policymakers in each state tailor their laws to the outcomes they most care about.

“It’s important to remember that these results don’t tell us anything about the mental health outcomes of people who may be using cannabis instead of anxiety medications, or why prescriptions for these other drugs are increasing,” she said. “So, there’s room for a lot of future research here.”. 

The study, published Sept. 5, 2024, in JAMA Network Open, is available at https://doi.org/10.1001/jamanetworkopen.2024.32021.

Within Antarctic ice, the IceCube Neutrino Observatory is recording rare astronomical phenomena. Constructed in the harsh conditions of the South Pole, it is the first detector of its kind. But now, a sister project is underway — one located over 2,600 meters beneath the surface of the Pacific Ocean.

Called the Pacific Ocean Neutrino Experiment (P-ONE), it will be built off the coast of Washington State in the Cascadia Basin with global collaboration including Georgia Tech’s Ignacio Taboada. 

Taboada, who is the current spokesperson of the IceCube collaboration and a professor in the School of Physics, has been awarded over $1.5 million in funding through a Major Research Instrumentation grant from the National Science Foundation (NSF) to build P-ONE’s sensor trigger system, which will record and identify sources of light as they are captured by the telescope’s sensors.

“This is a multi-institute collaboration,” Taboada shares. Co-PI’s include Naoko Kurahashi Neilson of Drexel University, Nathan Whitehorn and Tyce DeYoung of Michigan State University, and Alexandra Rahlin of the University of Chicago.

2,600 meters under the sea

Taboada says the team was drawn to the underwater location, despite the associated building challenges because “the characteristics of the seawater mean that we could identify more individual sources better than IceCube can, if we can build a detector of the same size.”

Capturing astrophysical particles is a balance of finding the right medium for the sensors: the medium’s density contributes to how many particles are captured. 

While an open-air observatory would be possible, Taboada explains that “air is about 1,000 times less dense, so it means that we would get 1,000 times fewer neutrinos interacting in the detector — and neutrino detections are very, very rare.” Using a medium like ice or seawater maximizes the possibility of capturing these particles.

Ice and seawater also present unique challenges. “The ice in Antarctica is extremely transparent,” Taboada explains. This means that when a photon enters the ice, it can travel a very long distance within that ice. “But it doesn't travel in a straight line,” he says. Instead, the particle ricochets and scatters, deviating from its original path.

This makes it more difficult to determine exactly where the particle has come from — a key aspect for astronomical observations. “In comparison, light entering seawater scatters much less," Taboada says. “It always travels in a straight line.” Because of this, neutrino directions are determined more precisely in seawater than in ice.

Tracing the cosmos

Key to capturing these particles is the trigger system that Taboada will build with this new funding. That component  will collect data around interesting events, which are seen as light to the system.  

But there are many sources of light in the ocean that aren’t from astronomical phenomena. “It's not something that can be trivially predicted,” says Taboada. “It's a very complicated situation and you have to adapt the trigger to various amounts of background light.”

For example, there’s bioluminescence to consider.

Some sources, like fish or small organisms, can move around independently, while others, like bioluminescent plankton, might instead react to turbulence. The trigger system will need to identify and filter out all of these sources.

“Seawater also has a lot of potassium,” Taboada adds. “One of the isotopes of potassium is radioactive, and the optical sensors can catch light from that.”

Once the trigger system recognizes and captures the event, the data is sent to the mainland, where computers will leverage machine and deep learning to determine exactly what the sensor has captured.

“It's a process of gathering and analyzing interesting data,” Taboada says, similar to looking into a night sky and differentiating shooting stars, constellations, satellites, and planes.

From sea to space

Because P-ONE is one of the first projects of its kind, the research team plans to initially install six or seven lines of instrumentation across the seafloor. “That is rather small,” says Taboada, “but it will demonstrate how to build the instrument and how to operate it.”

“P-ONE has the eventual objective of being similar to IceCube in size,” he adds. “But it will be a northern hemisphere detector (meaning it can ‘see’ different parts of the sky than IceCube), and should have significantly better angular resolution and sensitivity.” And while P-ONE’s location will provide views that IceCube can’t, the effort also has the potential to provide a new perspective of the ocean floor.

The system will continuously monitor the deep ocean at an unprecedented scale, capturing data about environmental conditions and biological processes, key information for oceanographers and marine biologists — all while furthering the field of neutrino astrophysics.

 

Funding: NSF

P-ONE is a collaboration between the following organizations: 

Ocean Networks Canada; University of Victoria; University of Alberta; Department of Physics, Queen's University; Department of Physics, Simon Fraser University; TRIUMF; Department of Physics, Technical University of Munich; Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen Centre for Astroparticle Physics; Collaborative Research Centre 1258 (SFB1258) at TUM funded by the Deutsche Forschungsgemeinschaft (DFG); European Southern Observatory; Institut für Kernphysik, Goethe Universität Frankfurt; GSI Helmholtzzentrum für Schwerionenforschung; Max Planck Institute for Physics; Institute of Nuclear Physics, Polish Academy of Science; University College London; Department of Physics and Astronomy, Michigan State University; Georgia Institute of Technology; Drexel University; University of Chicago

Imagine being a passenger in a self-driving car as the vehicle starts veering off the road. It’s not a faulty sensor causing the dangerous situation — it’s a cyberattack. Hackers can access the deep learning (DL) neural networks at the heart of the vehicle’s computer system, compromising the safety of its passengers, as well as other drivers and pedestrians. 

Stopping such cyberattacks requires understanding them first, but this can be challenging. Finding a computing system’s exact deep neural network has many roadblocks. They are often proprietary and, therefore, inaccessible to investigators without considerable legal intervention. Another common problem is that they are updated frequently, making it difficult for investigating researchers to access the most current network iteration. But a new tool from Georgia Tech could unlock the mysterious malware on myriad neural networks in everything from self-driving cars to the IMDB entertainment database. AI Psychiatry (AiP) is a postmortem cybersecurity forensic tool that uses artificial intelligence to recover the exact models a compromised machine runs on and discover where the fatal error occurred.

“We trust self-driving cars with our lives and ChatGPT with our careers, but when those systems fail, how are we going to investigate them?” said Brendan Saltaformaggio, an associate professor with joint appointments in the School of  Cybersecurity and Privacy and the School of Electrical and Computer Engineering (ECE).

AiP can recover the original DL model on both the local network’s memory and the graphics processing unit that trains the network. It can accomplish this without any specific knowledge of the model’s framework, platform, or version. Instead, it recreates the model using what Saltaformaggio refers to as “clues,” or common components in all neural networks. These include weights, biases, shapes, and layers from the model’s memory image — a frozen set of the bits and bytes operating when the model is running normally. The memory image is crucial because it enables AiP to compare it with the model post-attack.

“These models often refine their information as they go, based on their current environment, so an attack might happen as a result of an attacker poisoning the information a particular model is learning,” said David Oygenblik, an ECE Ph.D. student. “We determined that a memory image would capture all those changes that occur during a runtime.”

Once the model is recovered, AiP can run it on another device, letting investigators test it thoroughly to determine where the flaws lie. AiP has been tested with different versions of both popular machine learning frameworks (TensorFlow and PyTorch) and datasets (CIFAR-10, LISA, and IMDB). It successfully recovered and rehosted 30 models with 100% accuracy. 

“Before our research, you couldn't go to the cyber ‘crime scene’ and find clues because there was no technique available to do that,” Saltaformaggio said. “That's what we are pioneering in the cyber forensics lab right now — techniques to get that evidence out of a crime scene.”

Tools like AiP will allow cyber investigators to see the whole picture immediately. Solving cybercrimes can help prevent future ones, from safeguarding a user’s data to keeping a car on the road. 

AiP is the inaugural winner of GTRI's Graduate Student Fellowship Program. 

Tech AI at Georgia Tech has appointed Tim Brown as interim director of professional education. He is also the new academic program director for AI, a joint appointment by Tech AI and Georgia Tech Professional Education (GTPE). Previously, Brown served as managing director of Georgia Tech’s Supply Chain and Logistics Institute for nearly 10 years, where he focused on program expansion and partnership development.

In his new role, Brown will work closely with the College of Lifetime Learning and Tech AI to develop innovative AI programs. He will identify industry needs and create interdisciplinary academic offerings serving a diverse range of learners, from K-12 students to executives. His initial emphasis will be on mid-career professionals seeking to upskill or reskill, equipping them with the technical skills essential for success in the AI field. He will also enhance existing programs and provide educational opportunities to companies and organizations, addressing current market demands.

Brown has more than 35 years of experience in professional education and supply chain optimization, including roles at IBM, Accenture, Chainalytics, Frito-Lay, and Tropicana. He has worked with executives in various industries, advising on supply chain management and securing $81 million in funding for AI in manufacturing through the Georgia AIM coalition.

In a statement, Brown said, “I look forward to contributing to innovative AI programs at Georgia Tech. Our goal is to create educational opportunities that meet the diverse needs of learners and equip them with the skills necessary to thrive in this evolving field.”

Brown’s leadership underscores Georgia Tech’s commitment to innovation and education in the rapidly changing landscape of AI. He is dedicated to establishing Georgia as a leader in AI and highlighting the resources and capabilities that Georgia Tech offers.

Susan Puryear has been named interim associate vice president for Research Administration at Georgia Tech, effective November 1, 2024. Puryear most recently served as the associate vice chancellor for Research Administration, Integrity, and Development at the University of Massachusetts Lowell, where she led research infrastructure operations. 

Puryear has worked for more than 15 years in both private and public higher-education leadership in research administration, compliance, and development. Prior to the University of Massachusetts Lowell, Puryear oversaw research administration and compliance at Clark University in Worcester, Massachusetts. Her earlier career included consulting work with higher education institutions and other nonprofit and for-profit organizations in the areas of business development, program design and implementation, and fundraising.

Puryear’s extensive academic background includes an undergraduate degree from Yale University, a master’s in education from Harvard University, a J.D. degree from Stanford University, and an Ed.D. in higher education management from the University of Pennsylvania.

Krista Walton, associate vice president for Research Operations and Infrastructure, is chairing the search for a permanent associate vice president for Research Administration. The nationwide search is being led by WittKiefer. 

More details on the search process and timeline are forthcoming.

 

 

 

 

For his work creating new kinds of drug delivery techniques and bringing those technologies to patients, Mark Prausnitz is one of the new members of the National Academy of Medicine (NAM).

The Academy announced his election Oct. 21 alongside 99 others. Membership in NAM is considered one of the highest recognitions in health and medicine, reserved for those who’ve made major contributions to healthcare, medical sciences, and public health. The roster is small: only 2,400 or so individuals have been honored.

“It’s an honor to be elected to the National Academy of Medicine and have the work of our team at Georgia Tech recognized in this way,” said Prausnitz, Regents’ Professor and J. Erskine Love Jr. Chair in the School of Chemical and Biomolecular Engineering.

The Academy cited Prausnitz for innovating microneedle and other advanced drug delivery technologies. He also was honored for translating those methods and devices into clinical trials and products and founding companies to bring the advances to patients. NAM praised Prausnitz for “inspiring students to be creative and impactful engineers.”

Read the full story on the College of Engineering website.