December 2024
Friends,
As we complete another trip around the sun and winter settles in with frosted windows and shorter days, I’d like to celebrate the connections that bind us together. Our cozy corner of Vanderbilt is a fantastic place to wonder and explore. Over the past year, we’ve shared more than 3,000 hours together in person and over 200 hours virtually, strengthening our bonds and advancing our shared goals.* We’re nurturing new connections through VALIANT Reach and celebrating a first year together with research awards, pilot grants, new initiatives, and fellowships—all in 2024! Our community’s productivity continues to soar, with remarkable growth in both publications and citations. We have been here to support our community with dozens of letters of support and multiple training awards.
As we welcome the new year, we’re energized by the prospect of bringing people together in innovative and meaningful ways. For our AI Fellows, we’re launching an AI Salon Series to discover cross-silo connections while sparking vibrant discussions. Meanwhile, our AI Scholars VALIANT Attempt will foster social, scientific, and collaborative engagement for scholars across our community. Our first data releases from our strategic projects will be ready soon in 2025. (I’m looking forward to seeing a neuroanatomy in a new way!) We are also starting a series of deep dives on subject areas (starting just after we return in the new year for a deep dive into diffusion MRI), and we will support pop-up efforts to explore Hot Topics in AI with an un-journal club.
To close, I would like to share a puzzle that brings me joy every year. The 2024 winter solstice is Saturday, December 21, at 3:20 am, when the sun will rise at 6:54 am and set at 4:53 pm for a daylight time of 9 hours 41 minutes and 53 seconds. Interestingly, while the 2024 winter solstice marks the shortest day of the year, the latest sunrise doesn’t occur until January 2, 2024, at 6:58 am—by which time we’ll already have 9 hours, 46 minutes, and 8 seconds of daylight. How does that happen?
Bridging AI and Community
For Elias Levy, becoming a Research Technician at VALIANT has been the culmination of years of curiosity, exploration, and a passion for problem-solving. A graduate of Vanderbilt’s Computer Science program with a minor in Engineering Management, Elias first discovered his love for technology on his high school robotics team. Although his initial role on the team was in management, the coding aspect piqued his interest, leading him to explore computer science and apply to Vanderbilt with ambitious plans for a triple major in computer science, math, and physics. While those plans evolved over time, his dedication to innovation never wavered.
Elias first connected with VALIANT through the MASI Lab during his undergraduate research, where he worked on integrating cutting-edge AI tools into the lab’s workflows (“ChatGP-Me?“). His projects included exploring Retrieval-Augmented Generation (RAG) models, which enhance large language model (LLM) capabilities by verifying responses against external data sources. “RAG models bridge the gap between traditional AI and practical, specialized solutions,” he explains. Through these efforts, Elias helped streamline research processes and demonstrated the real-world potential of AI in academic settings.
Beyond his technical skills, Elias emphasizes the collaborative spirit of VALIANT as a key factor in his growth. “AI isn’t just about building models; it’s about the community,” he shares. “Being able to discuss ideas, share challenges, and learn from one another is invaluable.” His work has extended to partnerships across Vanderbilt, including conversations with the Stevenson Library about how AI might support their efforts. This cross-disciplinary collaboration has opened doors to new possibilities, fueling Elias’s excitement for the future of AI.
Elias’s contributions go beyond the technical realm. His knack for making advanced concepts accessible has inspired those around him, including researchers new to AI. By combining his technical expertise with a desire to drive impactful change, Elias continues to expand the horizons of what AI can achieve. “I’ve been surprised by how much interest there is in specialized AI tools,” he reflects. “It’s rewarding to see people recognize the potential for these technologies in their own fields.”
As he looks ahead, Elias is unwavering in his commitment to advancing AI’s role in society. Whether through academic research or industry applications, he hopes to remain at the forefront of this rapidly evolving field. “AI has the power to change the world for the better,” he says. “I’m grateful to be part of that journey, especially here at Vanderbilt, where innovation thrives.”
Alumni Lookout
After finishing my PhD, I completed a 2-year postdoctoral fellowship with Vanderbilt’s Data Science Institute. My postdoc research focused on developing deep learning beamforming techniques to improve B-mode ultrasound image quality. After completing my postdoc, I joined Philips in Cambridge, MA as an Ultrasound Systems Engineer. At Philips, I worked on probe integration and image optimization for their hand-held Lumify platform. At the end of 2023, I joined Rivanna, a company based out of Charlottesville, VA.
At Rivanna, we’re developing point-of-care ultrasound systems for targeted clinical applications that aim to improve procedure and diagnostic accuracy while also minimizing workflow and physician training challenges. AI has become increasingly more impactful towards achieving these goals in real-time on the products we’re developing. Specifically, compared to analytical alternatives, AI-driven approaches have greatly improved adaptability and efficiency, allowing for broader patient impact and more streamlined workflows for physicians.
Something I value from my time at Vanderbilt is learning how to foster interdisciplinary collaborations, which has been an immensely important skill to have while working in industry.
VALIANT Ventures
- Transforming medical imaging with Transformers? A comparative review of key properties, current progresses, and future perspectives
- Multiplexed 3D atlas of state transitions and immune interaction in colorectal cancer
- Competencies for the Use of Artificial Intelligence-Based Tools by Health Care Professionals
- Wassterstein Embedding for Graph Learning
- Advances and prospects for the Human BioMolecular Atlas Program (HuBMAP)
- Deep multimodal fusion of image and non-image data in disease diagnosis and prognosis: a review
- Metrics reloaded: recommendations for image analysis validation
- Telemedicine Impact on the Patient-Provider Relationship in Primary Care During the COVID-19 Pandemic
- CLIP-Driven Universal Model for Organ Segmentation and Tumor Detection
Vision Problems in Mild Traumatic Brain Injury Uncovered With Comprehensive Workup
Bryam Chairs NIH’s ITD study section
KPMP Glue Grant
APSA Travel Award
Chloe Cho was selected for a Vanderbilt MSTP American Physcian Scientist Association travel award.
Happy Holidays!
Today, the AI Scholars are celebrating the end of the fall semester and the beginning of winter break with our annual white elephant & poetry exchange. Scholars, see your e-mail for details.
AI Showcase
SPIE Travel Award
BrainHack 2025
Alchemists’ Corner
- Brain Age Is Not a Significant Predictor of Relapse Risk in Late-Life Depression
- Modeling Marronage: GIS Heuristics of Refuge Affordances in Colonial St. Croix
- “Math is a pain!”: Understanding challenges and needs of the Machine Learning community on Stack Overflow
- 3D Gaze Tracking for Studying Collaborative Interactions in Mixed-Reality Environments
- Addressing ethical issues in healthcare artificial intelligence using a lifecycle-informed process
- ChatGP-Me?
- Precision targeting in the globus pallidus interna: insights from the multicenter, prospective, blinded VA/NINDS CSP 468 study
valiant@vanderbilt.edu | https://vanderbilt.edu/valiant | @VandyValiant | LinkedIN
“Vanderbilt” and the Vanderbilt logo are registered trademarks and service marks of Vanderbilt University.