Frank (Sicong) Chen is a postdoctoral researcher in the HumanX Cluster at
Dartmouth College, where he works on human-centered wearable and mobile
systems that sense, interpret, and respond to physiological and behavioral
states in real time. His research examines how everyday devices can move
beyond passive monitoring to become intelligent systems that detect risk,
uncertainty, and vulnerability, and intervene in ways that meaningfully
support human well-being.
His work integrates wearable sensing, behavioral biometrics, and multimodal
machine learning to study signals related to movement, interaction patterns,
and biosignals. By modeling how these signals evolve over time, he develops
closed-loop systems that can adapt their behavior based on user state,
confidence, and context. This perspective underlies his research on
continuous authentication, stress and mental health monitoring, and
adaptive intervention systems.
Frank earned his Ph.D. in Computer and Information Science and Engineering
from Syracuse University, where his doctoral research focused on robust
continuous authentication and real-world behavioral biometric systems. By
bridging biometric security and digital health, his work aims to enable
technologies that are both reliable and empathetic—systems that recognize
who users are, infer how they are doing, and provide timely, personalized
support.
Education
Ph.D., Computer and Information Science and Engineering, Syracuse
University
Advisor: Prof. Vir V. Phoha
M.S. in Computer Sciecne, Syracuse University
B.S. in Mathematircs and Applid Mathematics, Tianjin University
Wearable and Mobile Sensing for Human State Inference
Designing non-invasive wearable and mobile sensing systems to infer
cognitive, emotional, and behavioral states in everyday settings.
Integrating multimodal sensing and machine learning to enable
continuous monitoring, early risk detection, and context-aware support
across health, well-being, and daily human–technology interaction.
Systems
Closed-Loop Intelligent Interventions
Building end-to-end systems that connect real-time physiological and
behavioral sensing with adaptive decision-making, enabling
interventions to be delivered based on user state, context, and model
confidence.
Security
Multimodal Continuous Authentication on Mobile and Wearable Devices
Developing continuous authentication frameworks that leverage
multimodal behavioral and contextual signals on mobile and wearable
devices to achieve robustness to real-world variability,
uncertainty-aware modeling, and secure deployment beyond single-session
or static authentication.
Publications
SSPRA: A robust approach to continuous authentication amidst real-world adversarial challenges
IEEE Transactions on Biometrics, Behavior, and Identity Science, 2024
Frank Sicong Chen; Jingyu Xin; Vir V. Phoha
DSTER: A dual-stream transformer-based emotion recognition model through keystrokes dynamics
IEEE International Joint Conference on Biometrics (IJCB), 2024
Best-Reviewed Paper
Frank Sicong Chen; Shruti Rao; Brijesh Tiwari; Vir V. Phoha
Gaitpoint: A gait recognition network based on point cloud analysis
IEEE International Conference on Image Processing (ICIP), 2022
Jiajing Chen; Huantao Ren; Frank Sicong Chen; Senem Velipasalar;
Vir V. Phoha
An AI-Based Gait Authentication Framework Leveraging Swing Phase Dynamics
ACM Digital Threats: Research and Practice (DTRAP), 2026 (in press)
Frank Sicong Chen; Amith K. Belman; Pronab Mohanty
Beyond normality: Rethinking behavioral biometric data
AI-Enabled Forensic Investigations in Digital Sciences. Springer, 2025 (in press)
Amith K. Belman; Frank Sicong Chen; Vir V. Phoha; Pronab Mohanty
Formalizing PQRST complex in accelerometer-based gait cycle for
authentication
An integrated software–hardware platform that supports living brain
organoids as biological computing substrates, enabling sensing,
interaction, and learning from real-world environmental signals.
A robust multimodal continuous authentication framework that models
uncertainty and transitional states to enable secure, real-world
deployment on mobile and wearable devices.
A wearable sensing and machine learning framework for continuous
stress detection under real-world conditions with incomplete and noisy
multimodal data.
Honors & Awards
Best-Reviewed Paper Selection and Journal Invitation, IEEE IJCB, 2024.
Program Committee Member, IEEE Transactions on Computational Social Systems
Reviewer, ACM Digital Threats: Research and Practice
Conference Program Committees
Program Committee Member, IEEE FG 2026
Program Committee Member, AAAI 2026
Program Committee Member, IEEE CogMI 2025
Conference Reviewing
Reviewer, CAIP 2025
Reviewer, InterID Workshop (FG 2025)
Teaching Experience
Graduate and undergraduate instruction in machine learning for security,
biometrics, software implementation, and discrete mathematics. Delivered
independent lectures, developed instructional modules, and mentored
student research projects, including projects leading to peer-reviewed
publications. Emphasis on systems thinking and applied AI.
For a complete list of courses and responsibilities, see CV.