The Center for Multimodal Big Data Science and Healthcare builds tools that turn complex, messy scientific data — images, volumes, spectra, sensor streams, and more — into real discovery. Our flagship platform, BisQue, brings this data together so researchers across a wide range of scientific domains can organize, analyze, and share their findings at scale. BisQue Ultra extends this further with domain-tuned AI reasoning trained on scientific data, enabling deeper, more context-aware analysis.
BisQue keeps track of rich, flexible metadata alongside your data, lets you plug in custom analysis modules that run on compute clusters, and preserves full provenance so every result can be traced back to the workflow that produced it — all accessible from any browser and scalable from a single dataset to petabytes of multimodal scientific data.
The interdisciplinary research projects built on top of BisQue span four main areas: marine science, plant biology, materials science, and healthcare.
Bio-Image Informatics
EM data and cellular segmentation
Materials Science & Structural Analysis
DREAM.3D and structural analysis
Healthcare
Patient monitoring and analytics
Marine Sciences
Underwater image segmentation and classification
BisQue Deep Learning Cyberinfrastructure
A feature in Research Outreach introduces the BisQue ecosystem — a cloud-based platform for sharing data and analysis methods and improving the reproducibility of computer-vision techniques across scientific disciplines, from materials and marine science to medical imaging.
Reproducible computer vision: cross-disciplinary and scalable image informatics →
BisQue Deep Learning Workshop, July 9–10, 2026
The center hosted a BisQue Deep Learning (BDL) Workshop at the University of California, Santa Barbara — an open, collaborative gathering of researchers working on deep learning, computer vision, continual learning, and uncertainty estimation for scientific image analysis.
The two-day workshop took place at UCSB from July 9–10, 2026, starting at 9:00 AM on the first day and adjourning by 4:30 PM on the second.
The workshop brought together computer vision researchers, ecologists, and materials scientists to strengthen the essential elements of automated computer-vision workflows in BisQue, with a primary focus on animal detection and monitoring. The main objective was to work through the full pipeline — from getting data into the platform, through continual learning and uncertainty estimation, to acting on model results — and to identify concrete next steps for the platform's capabilities, usability, and supporting training materials.
The first day focused on the BisQue platform and models: updates and demos of the latest capabilities including BisQue Ultra, data upload and the import/export of annotations, model selection and the model library, broader applications in remote and environmental sensing, and model evaluation, re-training, and fine-tuning. The second day addressed uncertainty estimation — how it is integrated into BisQue and used to guide annotation and model selection — along with LLM integration through BisQue Ultra, and an end-to-end deep dive into a complete animal-monitoring workflow.
Participants came from UC Santa Barbara, the Smithsonian's National Zoo & Conservation Biology Institute, UC Riverside, Stanford, Ohio State, the University of Arizona, and Google.
BDL Workshop, February 12, 2025
The center hosted a BisQue Deep Learning (BDL) Workshop at the University of California, Santa Barbara — a collaborative gathering of researchers working on deep learning, computer vision, continual learning, and uncertainty estimation for scientific image analysis.
The full-day workshop took place at UCSB in the Engineering Sciences Building (ESB 1001) on February 12, 2025, with registration at 8:30 AM and a 9:00 AM start.
This workshop brought together researchers on the theme of Multimodal Data Integration and Analytics using BisQue. Collaborators shared their current research and demonstrated how BisQue is enabling large-scale deep learning and computer vision workflows on their own scientific datasets, spanning applications from ecology to materials science. The day also featured focused discussions on the long-term sustainability of BisQue, community adoption of open-source platforms, and plans for evolving the codebase for the future.
LIMPID/BisQue + IDEAS Joint Workshop, February 3–5, 2020
The center hosted the second LIMPID/BisQue workshop on Scalable Image Informatics. This was a joint meeting with the workshop on Applications of Machine Learning and Scalable Image Informatics to Materials Discovery, held as part of the NSF-sponsored IDEAS program on Data-Driven Approaches to Materials Discovery, at the University of California, Santa Barbara. Participation was by invitation only, with over 80 registered participants.
The three-day joint workshop took place at UCSB from February 3–5, 2020. The workshop started at 8:00 AM on Monday, February 3rd and adjourned by 5:00 PM on Wednesday, February 5th.
This workshop brought together researchers from academia and national laboratories who are developing or have adopted new experimental methods and measurement tools that provide very large and rich data sets, with early adopters of machine learning as well as leaders in the machine learning community. The main objective of the LIMPID/BisQue workshop was to identify key requirements for scalability and sustainability of software infrastructures to support multimodal data analysis and machine learning in diverse application areas such as life sciences, marine science, materials and health sciences. A key objective of the IDEAS workshop was to connect researchers in the data sciences field with domain researchers to foster a dialogue and establish the state-of-the-art in the application of machine learning to materials discovery related disciplines, and to identify (i) the potential impact of, and opportunities for, data sciences in materials, with an emphasis on integrating experimental and computational materials; and (ii) the grand challenges at the intersection of materials discovery and machine learning.
Monday focused on sustainable software infrastructure for multimodal informatics and machine learning for image applications. Tuesday addressed data-driven approaches for advanced materials applications, effectively educating a workforce of data-driven innovators, and high-throughput approaches in computational mechanics. Wednesday focused on accelerating data-intensive research with a range of applications, and AI and ML for high-dimensional and multi-scale datasets.
Organizers: Samantha Daly, Mechanical Engineering, UCSB; B.S. Manjunath, Electrical and Computer Engineering, UCSB; Tresa Pollock, Materials, UCSB.
