A14Science focuses on developing AI/ML models to extract novel insights in scientific and healthcare applications. Targeted application domains include polymer recycling, molecular dynamics simulations, healthcare-patient matching, and Transmission Electron Microscopy (TEM) diffraction.
Projects under this research area focus on applying state of the art Artificial Intelligence/Machine Learning models to extract novel insights or scale computations for scientific applications.
Predictive Framework to Indicate the Age of Plastics For Proper Recycling
This project is in collaboration with Prof. Metin Karayilan and Prof. Divita Mathur from the Department of Chemistry. The objective of this project is to develop ML driven framework for indicating the age of plastics using spectral data to facilitate their classification prior to the recycling process.
Funding Source: MDS-Rely
Funding Source: MDS-Rely
Usability and Validation Study of RE-Assist: an Artificially Intelligent Care Coordination Technology
RE-Assist is an AI based personalized care coordination tool. In this collaborative project with Prof. Ronald Hickman (CWRU – School of Nursing), Dr. Carla Harwell (University Hospitals), and the owner of RE-Assist Ashley Barrow, our objective is to assess the usability of the tool and evaluate its effectiveness by employing various performance evaluation metrics.
Funding Source: Clinical and Translational Science Collaborative (CTSC)
Funding Source: Clinical and Translational Science Collaborative (CTSC)
Automated Indexing of TEM Diffraction Patterns Using Machine Learning
Indexing Transmission Electron Microscopy (TEM) diffraction patterns is a critical step in materials characterization. Despite the manually intensive indexing process, work related to Machine Learning (ML) in its space is sparse. The objective of this work is to develop novel vision based models to classify the diffraction patterns to identify the material properties and structure.
Upscaling Particle Simulations via Machine Learning in the Loop
This is a collaborative project with Prof. Abhinendra Singh of the Department of Macromolecular Science and Engineering. State-of-the-art particulate simulation frameworks are limited in their capability to simulate on the order of thousands of particles and are not scalable to simulate real-world industrial or natural scale systems that require simulating on the order of millions of particles. The objective of this project is to achieve 3-4 orders of magnitude improvement scalability in these simulation frameworks by integrating Machine Learning in the loop to bypass numerical simulations.
