The National Science Foundation (NSF) under its Computer and Information Science and Engineering (CISE) Federal Grant Program (CFDA 47.070) awarded a $169,982 Project Grant to the University of Houston System (UHS) for the collaborative research project "Towards Robust and Human-Aligned Deep Learning for Medical-Sensor Time Series" from September 1, 2024 to August 31, 2027.
The project aims to develop robust deep learning techniques to analyze medical sensor time series data, addressing the issue of spurious correlations between data features and medical labels that can undermine the generalization of deep neural networks. The research will focus on 1) identifying input confounders in time series data, 2) designing knowledge-editing strategies to mitigate spurious correlations, and 3) investigating the approaches in medical applications such as Parkinson's disease monitoring and fall detection for the elderly. The project plans to produce open-source tools that can benefit a range of sensor-based medical monitoring and diagnosis tasks. No subawards are planned for this grant.
Generated 4/1/25, 3:56 AM