Positive publication bias has complicated the assessment of AI tools, according to a new multisociety scientific statement published January 28 in Radiology.
To better assess the current landscape and challenges of AI applications in cardiac CT and MRI specifically, the paper aims to bridge the gap between burgeoning research developments and limited clinical applications, according to lead author Domenico Mastrodicasa, MD, from the University of Washington in Seattle, and colleagues.
A wide variety of AI applications are at different maturity levels, ranging from early research development to full clinical implementation, according to the authors.
For various types of AI tools in cardiac imaging, the paper considers clinical relevance and analyzes current challenges at each stage in the cardiac imaging workflow. The paper also provides a technology readiness level (TRL) framework designed to facilitate informed decision-making and communication for optimized resource allocation so that technologies with the highest potential for clinical impact can be prioritized.
The RSNA and representatives from the European Society of Cardiovascular Radiology (ESCR), European Society of Medical Imaging Informatics (EuSoMII), North American Society for Cardiovascular Imaging (NASCI), Society of Cardiovascular Computed Tomography (SCCT), Society for Cardiovascular Magnetic Resonance (SCMR), and the Society for Imaging Informatics in Medicine (SIIM) have endorsed the document.
Read the full paper here.