Benchmarking Off-Grid Edge AI on Legacy Devices for Sustainable Deployment
This technology presents a sustainable, off-grid edge AI system that leverages legacy devices and renewable energy to enable intelligent computing in resource-limited environments.
Background:
Traditional artificial intelligence infrastructures depend heavily on continuous internet connectivity and advanced, energy-demanding hardware, restricting...
Published: 4/14/2026
|
Inventor(s): Asela Abeya, Nick LeJeune
Keywords(s): artificial intelligence for social good, benchmark, Energy-Aware Inference, Large language models, Legacy-Phone Clients, Modular Edge AI, Off-Grid AI, renewable energy, solar energy, Sustainable Computing, Technologies, wind energy
Category(s): Campus > SUNY Polytechnic Institute, Technology Classifications > Artificial Intelligence, Technology Classifications > Energy Conservation
|
Language-Agnostic Voice Biomarker Platform for Early Detection of Alzheimer’s Disease and Related Dementias (GloVoAD)
This technology enables early, non-invasive detection of Alzheimer’s disease and related dementias using a large-scale multilingual voice dataset and language-agnostic AI models that analyze vocal biomarkers without relying on speech transcription.
BACKGROUND:
Early detection of Alzheimer’s disease and related dementias (ADRD) remains...
Published: 4/16/2026
|
Inventor(s): Wei Bo, Wenyao Xu, Aditya Challamarad, Harry Lin
Keywords(s): Featured, Technologies
Category(s): Campus > University at Buffalo, Technology Classifications > Artificial Intelligence, Technology Classifications > Diagnostics
|
Few Shot Visual Counting
050-9160 Few Shot Visual Counting
A technology utilizing a class-agnostic Few-Shot Visual Counting method leveraging an Exemplar Matching Network
Background:
Visual counting, the automatic enumeration of objects or visual patterns within images and videos, is a critical task with widespread applications in areas like crowd monitoring, inventory...
Published: 4/2/2026
|
Inventor(s): Minh Hoai Nguyen, Viresh Ranjan, FNU Udbhav
Keywords(s):
Category(s): Technology Classifications > Artificial Intelligence, Campus > Stony Brook University
|
Beamforming Device and Architectures for Large Language Model Enabled Personal Monitoring
WisperWatcher is a wearable audio-based system that automatically monitors eating behavior by capturing and analyzing mastication and swallowing sounds. Leveraging beamforming techniques combined with Large Audio-Language Models (LALM), it provides accurate, real-time insights into food intake, meal timing, and user disposition while minimizing user...
Published: 4/16/2026
|
Inventor(s): Alexander Gherardi, Wenyao Xu, Wei Bo, Shuwei Hou
Keywords(s): Featured, Technologies
Category(s): Campus > University at Buffalo, Technology Classifications > Artificial Intelligence, Technology Classifications > Medical Devices
|
Optimizing Deep Neural Networks for EEG-Based Speech Recognition: A Multimodal Approach to Assistive Communication
A deep learning-based multimodal framework that combines EEG and audio signals to decode spoken phrases. By integrating neural activity with speech audio, the system improves recognition for individuals with neurological impairments, such as Parkinson's disease, ALS, or cerebral palsy, and enhances robustness in noisy environments. The framework...
Published: 4/16/2026
|
Inventor(s): Anarghya Das, Puru Soni, Wenyao Xu
Keywords(s): Assistive Technology, Featured, Technologies
Category(s): Campus > University at Buffalo, Technology Classifications > Artificial Intelligence, Technology Classifications > Medical Devices
|
Integrating AI in Back Injury Prevention
This wearable technology integrates sensors into a back brace and shoe inserts to monitor posture and movement. Using AI-driven analysis, it provides real-time feedback via a software application, helping Direct Care Workers prevent the risk of back injuries during lifting and transfer tasks.
Background:
Workplace injuries, particularly musculoskeletal...
Published: 2/25/2026
|
Inventor(s): Yuchi Young
Keywords(s): Technologies
Category(s): Campus > University at Albany, Technology Classifications > Artificial Intelligence, Technology Classifications > Healthcare
|
Improved ReRAM: A New Method for Memory Chip Manufacture
A new method of producing ReRAM chips more suitable for neuromorphic computer systems, AI, and other applications. Background: Chip manufacturers are actively pursuing the development of powerful, next-generation memory chips. One of these emerging technologies is ReRAM (resistive random access memory) which can preserve massive amounts of RAM data...
Published: 3/24/2026
|
Inventor(s): Nathaniel Cady, Karsten Beckmann, Joseph Van Nostrand
Keywords(s): artificial intelligence
Category(s): Technology Classifications > Artificial Intelligence, Technology Classifications > Computers, Technology Classifications > Electronics, Technology Classifications > Nanotechnology, Campus > University at Albany
|
Enhanced Fluorophore Discrimination in Fluorescence Microscopy Using Multi View Spectral Unmixing with Photobleaching Dynamics
This technology allows researchers to more easily distinguish hundreds of fluorescent labels in a single image, useful for a wide variety of biological applications.
Background:
There are hundreds to thousands of different cells or other structures that can be labeled with fluorescent reporters in a biological sample, and then imaged on a fluorescence...
Published: 3/26/2026
|
Inventor(s): Alex Valm, Yunlong Feng, Ruogu Wang
Keywords(s): fluorescence, fluoroscopy, Technologies
Category(s): Campus > University at Albany, Technology Classifications > Biomedical Science and Engineering, Technology Classifications > Bioinformatics, Technology Classifications > Artificial Intelligence, Technology Classifications > Imaging, Technology Classifications > Optics and Photonics, Technology Classifications > Research Tools and Reagents, Technology Classifications > Screens and Assays, Technology Classifications > Healthcare, Technology Classifications > Diagnostics, Technology Classifications > Dental
|
3D CT Reconstruction from Few X-rays
A technology that reconstructs CT images by integrating implicit neural representation and conditional diffusion models
Background:
Computed Tomography (CT) imaging faces a persistent challenge in balancing the need for high-quality, accurate three-dimensional reconstructions with the imperative to minimize patient radiation exposure and scanning...
Published: 4/15/2026
|
Inventor(s): Arie Kaufman, Gaofeng Deng
Keywords(s):
Category(s): Technology Classifications > Artificial Intelligence, Technology Classifications > Computer Software, Technology Classifications > Computers, Technology Classifications > Medical Imaging, Campus > Stony Brook University
|
A Novel Method to Assist the Orthopedic Surgeon in Bone Fracture Surgery
An inexpensive, powerful system that facilitates orthopedic fracture reduction by allowing surgeons to manipulate bone fragments virtually
Background:
Orthopaedic surgeons face significant challenges in accurately aligning broken bones during fracture repair, which is critical for restoring correct limb length, alignment, and rotation. Achieving...
Published: 10/7/2025
|
Inventor(s): Fazel Khan, Imin Kao
Keywords(s):
Category(s): Technology Classifications > Artificial Intelligence, Technology Classifications > Computers, Technology Classifications > Medical Devices, Technology Classifications > Medical Imaging, Campus > Stony Brook University
|