Elucidating coronary atherosclerosis using CT imaging and histology

• Applicant: Professor Dr. Marc Dewey
• Subject Area: Radiology ; Medical Physics, Biomedical Technology
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 567897117

Background and Objectives: Accurately identifying patients at risk for myocardial infarction remains a challenge in coronary atherosclerosis evaluation. Computed tomography (CT) is the preferred non-invasive imaging method for coronary atherosclerosis assessment, but has limitations in characterizing the microstructure of coronary plaques due to spatial resolution constraints. Histology remains the gold standard, yet histology-based coronary CT plaque software has only been validated with carotid histology, which differs in pathology from coronary atherosclerosis. This project aims to integrate histology-based validation into coronary CT plaque characterization, for improved accuracy beyond solely density-based assessment. The key objectives are to establish an ex-vivo coronary plaque database of 50 patient hearts with CT imaging co-registered with histology and micro-CT to subsequently increase the resolution of CT for the quantification of coronary plaques using the information derived from histology. Methods and Work Programme: We will employ a multi-platform ex-vivo imaging approach, using photon-counting detector CT (PCD-CT), standard CT (three different platforms), micro-CT, and histology to create a comprehensive imaging database of 50 patient hearts. This is expected to yield 1,500,000 slices from CT, 2,000,000 slices from micro-CT, and 7,500 histology cuts. The database will provide a reference standard for coronary plaque characterization to be co-registered with CT, and will be utilized to develop a novel framework to perform coronary plaque segmentation enhanced by histological morphology and classification. The proposed method consists of three work packages: 1) Automated plaque segmentation and deformable co-registration of coronary plaques in CT with micro-CT and histology; 2) Up-sampling of PCD-CT to the resolution of micro-CT to overcome spatial resolution constraints of CT; 3) Coronary plaque quantification in super-resolution CT based on histology to go beyond density-based quantification. The final method will be able to operate solely on information from clinical CT platforms. Diagnostic performance will be evaluated based on the accuracy of detecting thin-cap fibroatheromas in coronary CT with a statistical power >90% to confirm an area under the receiver operating characteristic curve of at least 0.80 as the diagnostic accuracy measure. Anticipated Gain of Knowledge: Elucidating coronary atherosclerosis using CT imaging and histology has the potential to enhance the accurate detection of patients at risk through non-invasive coronary CT by providing histology-level information. This will not only help to improve our understanding of coronary atherosclerosis, but could also support more personalized cardiovascular prevention strategies and allow for better monitoring of the effectiveness of anti-atherosclerotic treatments based on histology informed plaque quantification in coronary CT.

• DFG Programme: Research Grants