
Research Projects
Major Ongoing Research Projects
Our laboratory conducts a wide range of research from mathematical modeling of biosignals to machine learning recognition and practical system development.
Biosignal Modeling and Analysis
Mathematical modeling and analysis of biosignals using signal processing and probability theory.

Stochastic Generative Model and Signal Analysis of Surface EMG
Development of novel EMG signal models considering uncertainty in variance

Stochastic Generative Models of EEG and Brain State Analysis
Quantifying brain states through probabilistic models that capture EEG non-Gaussianity

Sequential Bayesian Learning of Biosignal Patterns
Development of efficient adaptive learning algorithms for changing signal characteristics
Machine Learning / Deep Learning Recognition
Building intelligent systems using pattern recognition, time series analysis, and deep learning.

Robust EMG Pattern Recognition Against Individual Differences
Motion classification algorithms that adapt to inter-subject and inter-trial variability

Recognition of Unmeasured Motions via Synthetic EMG Generation
Synthesizing EMG patterns with generative models to classify combined motions

Detection of Cancer Therapy-Related Cardiac Dysfunction Using ECG
Construction of early risk assessment systems using machine learning

EEG-Based Epileptic Seizure Detection and Cross-Patient Generalization
Seizure detection algorithms that require no per-patient calibration
Applied Systems and Interface Development
Implementing research results as practical systems and interfaces for real-world use in medicine and industry.

Robot Prosthetic Hand Control Mimicking Human Motor Characteristics
Development of biomimetic control systems achieving natural movements

Infant Developmental Assessment Using Video and Eye Gaze Measurement
Supporting developmental and ASD risk assessment by quantifying spontaneous movements and gaze

Quantifying Rehabilitation Assessment with Deep Learning
Automated estimation of FIM motor item scores from simple exercises

Dynamic Plaque Evaluation Based on Ultrasound Video Images
Development of high-precision evaluation systems for arteriosclerosis risk

Operation Pattern Analysis and Operator State Assessment for Construction Machinery
Multi-scale latent-space analysis of hydraulic excavator operation using generative models
Research Approach
Our laboratory promotes research through both theoretical and practical approaches. We conduct comprehensive research and development from mathematical fundamental research to practical system development and social implementation. Through collaboration with domestic and international research institutions and companies, we aim to create innovative research results while incorporating the latest technological trends.