Skip to content

Artificial intelligence tool may enhance usability of medical images - The Source - Washington University in St. Louis

Artificial intelligence tool may enhance usability of medical images - The Source - Washington University in St. Louis
A new artificial intelligence image processing tool developed by researchers at Washington University in St. Louis helps clear up the view of cardiac defects.

Table of Contents

Biomedical Engineer Develops AI Tool to Enhance Usability of Medical Images

Abhinav Jha, an assistant professor at the McKelvey School of Engineering and of radiology at the School of Medicine at Washington University in St. Louis, has emphasized the need for artificial intelligence (AI) tools in medical image processing to be evaluated based on clinical tasks rather than visual appeal. In a study published in IEEE Transactions on Radiation and Plasma Medical Sciences, Jha and his team developed a tool called DEMIST, which has the potential to improve performance on clinical tasks in the context of denoising myocardial perfusion imaging (MPI) single-photon emission computed tomography (SPECT) images. This tool could benefit patients by reducing the dose of radioactive tracer and the time required for imaging, while still preserving the ability to visualize cardiac defects. The new tool leverages a deep-learning framework to selectively clean low-count MPI SPECT images while preserving features that influence detection tasks. For more information, visit the McKelvey School of Engineering website.

Source

Latest