Nathaniel Cady
University at Albany, NY·Engineering
Two very different engineering problems sit side by side here: detecting disease from a drop of patient sample quickly and reliably, and making computer chips that process data where it is stored rather than shuttling it back and forth. Both involve building and validating physical devices, one photonic and one electronic, and both are judged by whether the device can be trusted in practice.
- Fields
Engineering
Biomedical Engineering
Computer Science
- Topics
Lyme disease
fluorescence-based diagnostics
biosensors
- Grants
- 2 known grants, through Aug 2027
What they work on
- Evaluate the commercial translation potential of an enhanced fluorescence-based diagnostic platform for Lyme disease, which detects antibodies and other biomarkers in under 40 minutes, through first-hand investigation of the diagnostics industry ecosystem.
- Establish the platform as a rapid, low-cost alternative to standard two-tiered testing for Lyme disease, including for early-stage infection where current serial laboratory testing is unreliable, while retaining two-tiered testing logic.
- Develop formal methods, mathematical techniques that prove a design meets its specification, to verify neural networks accelerated by analog in-memory computing, where calculations occur inside the memory array.
- Synthesize hybrid analog-digital in-memory computing circuits for neural networks using decision diagrams, neural networks, and reinforcement learning, and verify bidirectional digital in-memory circuits.
- Demonstrate the verified in-memory computing designs through in-field fabrication of working crossbar-based hardware.
More about the research
The first project assesses whether an enhanced fluorescence-based diagnostic platform can reach the market, with Lyme disease as the initial target. Fluorescence detection works by labeling a target molecule so it glows; the platform enhances that glow using photonics, the engineering of light, to detect antibodies and other disease markers at low concentrations. Lyme disease is the most common vector-borne illness in the United States, yet diagnosis remains difficult. The accepted standard, standard two-tiered testing, runs two tests in sequence, often in batches at a central laboratory. That takes time, needs specialists, and is especially unreliable early in infection, and the delay can cause real harm once the infection has spread. Proof-of-principle work positions this low-cost, highly sensitive platform as a rapid alternative, under 40 minutes total, that still retains the logic of two-tiered testing and could be run outside traditional clinical laboratories. The same platform has been demonstrated for COVID-19 and for detecting RNA-protein and DNA-protein interactions, which suggests broader use in biomarker discovery and pharmaceutical research. This I-Corps project uses experiential learning and direct investigation of the industry ecosystem to evaluate commercial potential.
Why it matters
A Lyme disease test that returns a result in under 40 minutes, outside a central laboratory, could reduce the diagnostic delay that leads to serious illness when the infection spreads, and the same platform may extend to other diseases and to biomarker discovery. Separately, provably correct in-memory computing could allow reliable artificial intelligence on error-prone emerging hardware while training neural networks with less electricity.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Sumit Jha
University of Florida, FL·Computer Science
Develop formal verification methods that prove neural networks run correctly when accelerated on analog in-memory computing hardware, which can be more error-prone than standard CMOS chips.
Abhijit Chatterjee
Georgia Institute of Technology, GA·Engineering
Develop post-manufacture test and tuning algorithms for deep neural networks running on analog crossbar arrays, recovering accuracy lost to manufacturing variation in seconds rather than hours and detecting shorts and opens.
Farimah Farahmandi
University of Florida, FL·Engineering·Early-career
Build a comprehensive database of security properties and rules for system-on-chip (SoC) designs, using AI to help collect and organize them for use in pre-fabrication security checks.
The school
University at Albany: 15 labs on PI Pages, $472M a year on research, 144 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.