MIDAS Lab
Research
We develop computational and artificial intelligence methods for biological, biomedical, scientific, and engineering data. Our research combines algorithm development, data analysis, and interdisciplinary applications.
Research Areas
Select a research area to move directly to its section.
Bioinformatics
Computational and AI-driven methods for analyzing complex biological data, from molecular sequences and biological networks to single-cell and spatial omics.
Explore this areaHealthcare AI
Artificial intelligence and signal analysis methods for extracting clinically meaningful information from biomedical signals and human movement data.
Explore this areaAI for Science & Engineering
AI and data-driven methods for diverse scientific and engineering problems, including time-series modeling, intelligent systems, and interdisciplinary research.
Explore this areaBioinformatics
Computational and AI-driven methods for analyzing complex biological data, from molecular sequences and biological networks to single-cell and spatial omics.
Back to topSingle-Cell & Spatial Omics
We develop computational methods for analyzing single-cell and spatial omics data. Our research includes clustering, imputation, feature selection, cell-type identification, trajectory analysis, gene regulatory network inference, and spatial mapping.
We are particularly interested in robust algorithms that can handle sparsity, noise, heterogeneity, and complex relationships among cells, genes, and spatial locations.
Network Biology
We study biological systems using graphs and networks. Our research includes biological network alignment, network clustering, module discovery, network integration, and analysis of protein-protein interaction and gene regulatory networks.
We develop graph-based algorithms to identify conserved structures, functional modules, and biologically meaningful relationships across heterogeneous biological networks.
Biological Sequence Analysis
We develop algorithms for analyzing DNA, RNA, and protein sequences and structures. Our research interests include sequence alignment, structural comparison, functional prediction, and RNA secondary structure analysis.
We are interested in computational models that combine sequence information, structural information, and biological knowledge.
AI-Driven Drug Discovery
We investigate AI-based approaches for drug-target identification, drug-response prediction, molecular representation, virtual screening, and computational drug discovery.
Our long-term goal is to integrate single-cell, spatial omics, biological networks, molecular structures, and scientific knowledge into interpretable drug discovery pipelines.
Healthcare AI
Artificial intelligence and signal analysis methods for extracting clinically meaningful information from biomedical signals and human movement data.
Back to topVoice & Speech Analysis
We develop machine learning methods for extracting health-related information from voice and speech data.
Our research includes acoustic feature analysis, spectrogram-based deep learning, speech-recognition-derived features, and multimodal models for disease screening and health assessment.
Cardiovascular & Respiratory Signals
We study algorithms for analyzing heart rate, pulse, respiration, and related physiological signals.
Our research interests include signal representation, feature extraction, time-series modeling, anomaly detection, health monitoring, and prediction using wearable or noninvasive sensing data.
Gait & Human Movement Analysis
We develop computer vision and time-series analysis methods for gait, posture, and body movement data.
Our research includes disease screening, severity estimation, subtype classification, movement pattern analysis, and interpretable analysis of relationships among body joints and temporal motion features.
Multimodal Health Data
We investigate methods for integrating biomedical signals, movement data, clinical variables, and other heterogeneous health information.
Our goal is to develop robust, interpretable, and clinically meaningful AI models for disease screening, monitoring, and personalized healthcare.
AI for Science & Engineering
AI and data-driven methods for diverse scientific and engineering problems, including time-series modeling, intelligent systems, and interdisciplinary research.
Back to topTime-Series Modeling
We develop machine learning and deep learning methods for sequential and temporal data.
Our research interests include long-term prediction, sequence-to-sequence learning, representation learning, adaptive normalization, uncertainty-aware modeling, and anomaly detection.
Scientific Data Analysis
We study computational methods for extracting meaningful patterns from complex scientific and engineering data.
This includes feature selection, clustering, prediction, visualization, statistical analysis, and interpretable modeling of structured and unstructured datasets.
Intelligent Systems
We investigate AI methods for modeling, monitoring, and predicting the behavior of engineering and physical systems.
Our research includes data-driven system modeling, sensor-data analysis, predictive maintenance, temperature prediction, and intelligent decision support.
Exploratory & Interdisciplinary Research
We are open to interesting research problems that connect artificial intelligence, data analysis, science, and engineering.
We welcome interdisciplinary collaborations and exploratory projects that apply computational methods to new scientific questions, emerging datasets, and practical engineering challenges.