📊VisualizationvisualizationResearch in graphic design methods as well as the effective use of visualization methods and techniques to explore and interpret high-dimensional data. Our group has worked in urban and sports computing, as well as in the areas of biochemistry, cybersecurity, development, healthcare, climate science, and human rights. ## Key Research Directions - **Visual Analytics**: Interactive systems for exploring complex datasets - **Urban Visualization**: Tools for understanding city-scale phenomena - **Sports Analytics**: Visualization for esports and traditional sports - **Explainable AI**: Visual methods for understanding machine learning models ## Notable Projects - **Urban Pulse**: Capturing the rhythm of cities through temporal pattern visualization - **Shadow Accrual Maps**: City-scale shadow accumulation over time - **Urban Rhapsody**: Large-scale exploration of urban soundscapes - **PipelineProfiler**: Visual exploration of AutoML pipelines - **VisFlow**: Web-based dataflow framework for visual data exploration
🧠ImagingimagingDevelopment of novel imaging analysis methodologies related to segmentation, shape analysis, and image statistics. Our work is multidisciplinary, primarily collaborating with experts in clinical imaging on research in areas including schizophrenia, autism, and multiple sclerosis. ## Key Research Directions - **Medical Image Segmentation**: Tools and methods for accurate anatomical structure identification - **Shape Analysis**: Statistical analysis of 2D and 3D anatomical shapes - **Longitudinal Analysis**: Tracking changes in brain structure over time - **Neuroimaging**: Analysis of brain development and neurological conditions ## Notable Projects & Software - **ITK-SNAP**: Interactive medical image segmentation tool for 3D images - **Deformetrica**: Statistical analysis of 2D and 3D shape data - **3D Slicer**: Open source platform for medical image informatics - **DTIPrep**: Automatic pipeline for DWI/DTI quality control - **SlicerSALT**: Shape Analysis Toolbox ## Clinical Applications Our imaging research has direct applications in understanding: - Autism spectrum disorder in infants - Schizophrenia - Multiple sclerosis - Age-related macular degeneration - Brain development in early childhood
💻Data AnalysisCreation of novel data management and analysis methods applied to urban, scientific, web, and public health domains. Our group has particular expertise in provenance and computational reproducibility and personal-data informatics. ## Key Research Directions - **Computational Reproducibility**: Tools and methods for reproducible research - **Urban Data Analytics**: Analysis of city-scale spatio-temporal data - **Web Data Discovery**: Finding and organizing information on the web - **AutoML**: Automated machine learning pipeline synthesis - **Responsible Data Science**: Fairness, accountability, and transparency ## Notable Projects - **ReproZip**: Computational reproducibility made easy - **D3M (Data-Driven Discovery of Models)**: Automated model discovery systems - **Vizier**: Open-source tool for building and refining data pipelines - **ACHE**: Focused web crawler for domain-specific content - **Domain Discovery Tool**: Interactive system for web domain exploration ## Research Themes ### Reproducibility Making scientific experiments reproducible and accessible through tools like ReproZip and ReproServer. ### Urban Computing Analyzing urban data including taxi trips, noise pollution (SONYC), and city infrastructure. ### Data Discovery Methods for finding, organizing, and understanding data from diverse sources including the web. ### Automated Machine Learning The D3M program develops systems that enable domain experts to create empirical models without data science expertise.
