Abstract:Abstract Background SPECT/CT integrates functional nuclear medicine imaging with anatomic CT characteristics. Aortic diameter provides important information when increased, indicating patients at higher risk of aortic dissection or rupture, and is routinely evaluated on CT scans but rarely on CT attenuation correction (CTAC) maps from SPECT/CT. Purpose We aimed to develop a deep learning (DL) method for fully automatic aortic diameter measurements on CTAC in patients undergoing SPECT/CT myocardial perfusion imaging (MPI). Methods Figure 1 shows the study design. We included CTAC from patients undergoing SPECT/CT MPI from 2 sites participating in the REFINE SPECT registry. An open-source multi-structure CT segmentation was used to segment the aorta, pulmonary artery (PA), and 3 thoracic vertebrae (Th4-Th6). The diameter of the ascending (Asc) and descending (Desc) aorta was measured automatically at each slice (on average 7 slices) at the level of the PA bifurcation (Th4-Th6). The median value of all obtained measurements of the diameter of the Asc and Desc aorta was used in further analysis. The abnormal diameter of the Asc aorta was defined as an aortic diameter >40 mm whereas the abnormal diameter of the Desc aorta >30 mm. Abnormal myocardial perfusion was defined as total perfusion deficit (TPD) >5%. Results In total 4,589 patients were included, of whom 2,596 (56.6%) were male, and the median age was 66.0 (interquartile range [IQR] 57.0-74.0). During the median 3.6 years (2.0-5.0) follow-up, 352 (7.7%) patients died. The median Asc aorta diameter was 36.4 mm (33.6-39.4), whereas the median Desc aorta diameter was 30.8 mm (28.6-33.3). Patients with abnormal Asc and Desc aortic diameter were at higher risk of death compared to subjects with normal aortic diameter (Asc aorta - unadjusted hazard ratio [HR] 1.37, 95% CI 1.08-1.74, p=0.01; Desc aorta – unadjusted HR 1.9, 95% CI 1.51-2.40, p<0.001) (Figure 2). Abnormal TPD was present in 2,302 (50.2%) patients. Patients with normal TPD and abnormal Asc aortic diameter were at higher risk of death compared to patients with normal TPD and normal Asc aorta diameter (unadjusted HR 1.26, 95% CI 0.83-1.90, p=0.27). Subjects with normal TPD and abnormal Desc aorta diameter had an elevated risk of death compared to patients with abnormal TPD and normal Desc aorta diameter (unadjusted HR 2.10, 95% CI 1.43-3.08, p<0.001). Patients with abnormal TPD and abnormal Asc aortic diameter were at higher risk of death compared to patients with abnormal TPD and normal Asc aorta diameter (unadjusted HR 1.35, 95% CI 1.00-1.81, p=0.04). Moreover, subjects with abnormal TPD and abnormal Desc aorta diameter had an elevated risk of death compared to patients with abnormal TPD and normal Desc aorta diameter (unadjusted HR 1.57, 95% CI 1.18-2.10, p<0.001). Conclusion DL-based aortic diameter measurements on CTAC in patients undergoing SPECT/CT MPI can help identify patients at higher risk of death due to abnormal aortic diameter.

Appropriate sizing of the frozen elephant trunk: How to predict proximal descending aortic diameter prior to dissection?

New predictor of aortic enlargement in uncomplicated type B aortic dissection based on elliptic Fourier analysis

Prevention of Aortic Dissection Suggests a Diameter Shift to a Lower Aortic Size Threshold for Intervention

Simple Death Risk Models to Predict In-hospital Outcomes in Acute Aortic Dissection in Emergency Department

Predicting the risk for acute type B aortic dissection in hypertensive patients using anatomic variables

Risk prediction of preoperative acute ischemic stroke in acute type A aortic dissection

Clinical Usefulness of Echocardiographic Measurement of Proximal Aortic Diameter in Early Differentiation Between Type A Acute Aortic Dissection and ST-Segment-Elevation Myocardial Infarction

Analysis of Ascending Aortic Diameter and Long-Term Prognosis in Patients with Ascending Aortic Dissection.

Aortic elongation part II: the risk of acute type A aortic dissection

Computational fluid dynamics-based prediction of aortic aneurysm rupture in a patient with chronic aortic dissection

Ascending aortic volume: A feasible indicator for ascending aortic aneurysm elective surgery?

The aortic paradox: a nationwide analysis of 523,994 individual echocardiograms exploring aortic dissection

Volumetric Measurements Improve the Accuracy of Aortic Remodeling Prediction in Aortic Dissection

WITHDRAWN: Machine Learning-Based Prediction of Distal Aortic Enlargement after Thoracic Endovascular Aortic Repair of Stanford Type B Aortic Dissection: A Multicenter, Retrospective Cohort Study

Values of aortic dissection detection risk score combined with ascending aorta diameter >40 mm for the early identification of type A acute aortic dissection.

Identification of High-Risk Patients for Development of Type B Aortic Dissection Based on Novel Morphological Parameters.

Fully automated aortic diameter measurements on SPECT/CT attenuation maps in patients undergoing myocardial perfusion imaging

Development and Evaluation of an Early Death Risk Prediction Model after Acute Type A Aortic Dissection

The aortic paradox: a nationwide analysis of 523 994 individual echocardiograms exploring fatal aortic dissection

Prediction of Distal Aortic Enlargement after Proximal Repair of Aortic Dissection Using Machine Learning.

Biomechanical stress analysis of Type-A aortic dissection at pre-dissection, post-dissection, and post-repair states