01
Trustworthy & adaptive AI
Generalization, uncertainty, and decision-making when an agent encounters conditions outside its training experience.
Explore the projectResearch with human purpose
Data Science
Innovations Lab
We study how intelligent systems learn, adapt, and work with people. Faculty and students turn research questions into working prototypes and evidence that can be examined.
Faculty inquiry. Student discovery. Shared progress.SUNY Canton · Interdisciplinary research
About the lab
Research is
a shared practice.
Based at SUNY Canton and directed by Dr. Mehdi Ghayoumi, DSIL brings together faculty, advisors, and students from across computing, engineering, and applied research.
Our mission
Develop intelligent systems that people can understand, trust, and use. We connect scientific rigor with human needs, while creating opportunities for students to grow as researchers.
DefineAsk a clear question.
BuildMake ideas testable.
ValidateExamine the evidence.
ShareCommunicate what matters.
01 / Research directions
Questions that guide our work.
Our work connects machine learning and systems engineering with the settings, constraints, and people that give a research problem meaning.
02
Human–AI collaboration
Selective assistance that supports group learning while keeping instructors in control of when and how AI intervenes.
Explore the project03
Embodied intelligence
Social interaction, expressive behavior, and the relationship between simulated decisions and physical robot actions.
Explore the project04
Multimodal health analytics
Methods that combine speech, text, visual, and behavioral signals with careful attention to privacy and evaluation.
Explore the project05
Security & privacy
Robust detection, explainable decisions, and responsible data practices for cyber defense and context-aware monitoring.
Explore the project06
Accessible interaction
Voice interfaces and digital assistants that account for different abilities, communication needs, and user preferences.
Explore the project02 / Research in development
Research in progress.
Projects across AI-assisted learning, adaptive agents, robotics, health, security, and accessibility.
Collaborative learning
LearnSync AI
Human-centered AI facilitation for small-group learning, with an emphasis on purposeful assistance and instructor oversight.
Simulation & data
SevaPAL Studio
A simulation and data-generation environment for studying AI facilitation and supporting the development of LearnSync.
Meeting assistance
LearnSync QuickJoin
An experimental connector exploring how an AI assistant can participate visibly in Microsoft Teams meetings.
Adaptive agents
AI in Unseen Conditions
Research on how intelligent agents and robots respond to unfamiliar environments, situations, and action outcomes.
Human–robot interaction
Reachy Mini Research
Exploring social interaction, expressive behavior, and assistive roles for small robots in learning and collaborative settings.
Health analytics
AMHAT
The Autonomous Mental Health Assessment Tool investigates privacy-conscious, multimodal approaches to stress screening.
Secure systems
Smart Surveillance Systems
Research into context-aware incident detection, with attention to privacy, explainability, and human oversight.
Inclusive interfaces
Digital Assistants & Accessibility
Intelligent avatars and voice-based interfaces that support diverse communication needs, adaptable assistance, and user agency.
03 / Publications
Research, in detail.
Selected work across collaborative learning, cybersecurity, privacy, and health analytics. Search by title or author, or narrow the collection by year and area.
12 papers
2026
LearnSync Human-Governed AI Facilitation with Minimum Intervention for Small-Group Learning: Architecture and Engineering Feasibility
LLM
Sevapal Studio: Reproducible Simulation and Audit for AI Group Facilitation
LLM
Trustworthy Deep Learning for Cybersecurity: A Structured Review Across Detection, Robustness, Privacy, Explainability, and Deployment
Electronics, 15(11), 2421
AMHAT-DL: Multimodal Deep Learning Pipeline for Privacy-Preserving Stress Screening
AIR-RES/CAC
2025
Rethinking Privacy Laws for Subscriptions: A Consumer Harm Perspective
CSCE
Human Rights in the Shadow of AI: Confronting Bias and Accountability
IEEE UEMCON
AMHAT: Multimodal Pipeline for Privacy-Preserving Stress Screening
CSCI
Detection of Alzheimer's Disease Using Bidirectional LSTM and Attention Mechanisms
Machine Learning and Applications: An International Journal
2024
Extending the Frontiers of Eye Tracking: Early Detection of Alzheimer's Disease Using Bidirectional LSTM and Attention Mechanisms
ACM Transactions on Applied Perception
MAISON: A Model for Effective Hybrid Management of Cybersecurity and Cyber-Trust
IJIT
Early Alzheimer's Detection: Bidirectional LSTM and Attention Mechanisms in Eye Tracking
CSCE
Advancing MAISON: Integrating Deep Learning and Social Dynamics in Cyberbullying Detection and Prevention
APCS
No papers match this search. Try another keyword or clear the filters.
04 / Our people
A community of inquiry.
Students work alongside faculty and advisors to frame questions, build systems, evaluate evidence, and communicate what they learn.
Leadership & advisors
Dr. Mehdi Ghayoumi
Lab Director
Dr. Kambiz Ghazinour
Entrepreneurship Advisor
Prof. Minhua Wang
Scientific Advisor
Dr. Samantha McCarthy
Scientific Advisor
Prof. Tiffany Forsythe
Scientific Advisor
Student researchers
Richard E. Ennist
Ryan Sessman
Cameron Cook
Behnaz Johnson
Anthony Marrero
Michael May
Amadu Fofana
Dhruven Parvatiya
Rodrico Sanchez
Alumni
Andrew Oakes
Data Science Researcher
Eliza Ochoa
Data Collection Assistant
Dena Barmas
Data Science Researcher
Cory Liu
Data Science Researcher
Elena Nye
Data Science Researcher
05 / Funding & support
Resources for discovery.
Research awards and infrastructure support provide equipment, computing resources, and opportunities to develop ideas beyond the lab.
$800Faculty development funding
SUNY Canton · Dual Reachy Mini Research Platform
Equipment support for two Reachy Mini robots within the embodied LearnSync research program.
11 September 2026 · Canino School of Engineering Technology
$50,000Research translation
NSF I-Corps Program
Entrepreneurial discovery and customer research to examine pathways from lab prototypes to sustainable products and services.
$5,000Cloud support
AWS Startups Grant
Cloud credits and technical support for data-intensive experimentation, model development, and secure infrastructure.
$8,500Project funding
NSF AARIPG Grant
Support for the Autonomous Mental Health Assessment Tool and its multimodal, privacy-focused research.
Our network
Sponsors, collaborators & partners
Relationships across research, healthcare, technology, and the community support the lab and its students.
06 / Join the conversation
Bring a question.
Build something meaningful.
We welcome students, research collaborators, community organizations, and industry partners. Tell us what you would like to investigate and how you hope to contribute.
ghayoumi@canton.eduFor prospective student researchers
Share your interests, relevant coursework or experience, and the time you can commit. Curiosity and a willingness to learn are a good place to start.