Grant Details

Grants Office Grantwriting service fee is currently unavailable for this grant
Get more information on grantwriting

Authority

Summary

History of Funding

None is available.

Additional Information

Generating these transformations will require fundamental research and development of new tools, workflows and methods across many dimensions; some of the themes are highlighted below. These themes should be seen as examples and not exhaustive.

• Fairness and Trustworthiness: Advancing fairness and trustworthiness in modeling in AI/ML is a highly interdisciplinary endeavor. Real world considerations go beyond the analytics and can inform new directions for computational science to better realize the benefits of algorithmic and data fairness and trustworthiness. The complexity of biomedical and health systems requires deeper understanding of causality in AI/ML models; new ways of integrating social and economic data to optimize health, such as disease heterogeneity, disease prevention, resilience, and treatment response, while systematically accounting for a broad range of uncertainties; and new insights into human-AI systems for clinical decision support. In general, this thrust supports the conduct of fundamental computational research into theories, techniques, and methodologies that go well beyond today's capabilities and are motivated by challenges and requirements in biomedical applications.
• Transformative Analytics in Biomedical and Behavioral Research: As biomedical and behavioral research continues to generate large amount of multi-level and multi-scale data (e.g., clinical, imaging, personal, social, contextual, environmental, and organizational data), challenges remain. New development in areas such as artificial intelligence and machine learning (AI/ML), natural language technologies (NLT), mathematics and statistics and/or quantum information science (QIS) also bring opportunities to address important biomedical and behavioral problems. This theme will support efforts to push forward the current frontline of AI/ML and advanced analytics for biomedical and behavioral research including:
  • novel data reduction methods;
  • new robust knowledge representations, visualizations, reasoning algorithms, optimization, modeling and inference methods to support development of innovative models for the study of health and disease;
  • new computational approaches with provable mathematical guarantees for fusion and analysis of complex behavioral, biomedical and image data to improve inference accuracy, especially in scenarios of noisy and limited data records;
  • novel explainable and interpretable AI/ML model development;
  • advanced data management systems with the capability to deal with security, privacy and provenance issues;
  • novel data systems to build a connected and modernized biomedical data ecosystem;
  • development of novel technologies to extract information from unstructured text data such as clinical notes, radiology and pathology reports;
  • development of novel simulation and modeling methods to aid in the design and evaluation of new assessments, treatments and medical devices; and
  • novel QIS approaches to unique challenges in biomedical and behavioral research.
• Next Generation Multimodal and Reconfigurable Sensing Systems: This theme addresses the need for new multimodal and reconfigurable sensing systems/platforms and analytics to generate predictive and personalized models of health. The next generation of sensor systems for smart health must have just-in-time monitoring of biomarkers from multiple sensor modalities (e.g., electrochemical, electromagnetic, mechanical, optical, acoustic, etc.) interfaced with different platforms (e.g., mobile, wearable, and implantable). Existing sensor systems generally operate either in discrete formats or with limited inter-connectivity, and are limited in accuracy, selectivity, reliability and data throughput.
• Cyber-Physical Systems: Development and adoption of automation has lagged in the biomedical and public health communities. Cyber-physical systems (CPS) are controlled systems built from, and dependent upon, the seamless integration of computation and physical components. These data-rich systems enable new and higher degrees of automation and autonomy. 
• Robotics: This theme addresses the need for novel robotics to provide support and/or automation to enhance health, lengthen lifespan and reduce illness, enhance social connectedness and reduce disabilities. The theme encourages research on robotic systems that exhibit significant levels of both computational capability and physical complexity. Robots are defined as intelligence embodied in an engineered construct, with the ability to process information, sense, plan, and move within or substantially alter its working environment.
• Biomedical Image interpretation. This theme's goal is to determine how characteristics of human pattern recognition, visual search, perceptual learning, attentional biases, etc. can inform and improve image interpretation. This theme would include using and developing presentation modalities (e.g., pathologists reading optical slides through a microscope vs. digital whole-slide imagery) and identifying the sources of inter- and intra-observer variability. The theme encourages development of models of how multi-modal contextual information (e.g., integrating patient history, omics, etc. with imaging data) changes the perception of complex images. It also supports new methods to exploit experts' implicit knowledge to improve perceptual decision making (e.g., via rapid gist extraction, context-guided search, etc.).

Contacts