Diagnosis of coronary artery disease in patients with type 2 diabetes mellitus based on computed tomography and pericoronary adipose tissue radiomics: a retrospective cross-sectional study

Fig. 1. Flowchart showing inclusion and exclusion criteria for the study population. CT computed tomography, CTA computed tomography angiography, CAD coronary artery disease, T2DM type 2 diabetes mellitus

Fig. 2. Representative case of imaging parameters measured by noncontrast chest CT and coronary CTA. a Example diagram of CACS measurements, yellow indicates calcification on LAD and green indicates calcification on LCX; b PAT volume measurement, red areas indicate pericardial adipose tissue; c CT-FFR model, CT-FFR values are marked in blue to red in the colour bar; d Curved multiplanar view showing PCAT quantification around the proximal coronary artery within reference diameter from the outer wall of the vessel, PCAT is visualized with adipose tissue HU colour table shown with colour bars; e shows the segment of the proximal coronary artery in a straightened view. CT computed tomography, CTA computed tomography angiography, CACS coronary artery calcium score, PAT pericardial adipose tissue, CT-FFR computed tomography-based fractional flow reserve, PCAT pericoronary adipose tissue

Fig. 3. A flow chart of model development. CTA computed tomography angiography, CT computed tomography, PCAT pericoronary adipose tissue, CACS coronary artery calcium score, PAT pericardial adipose tissue, CT-FFR computed tomography-based fractional flow reserve, PCATA pericoronary adipose tissue attenuation

Fig. 4. Receiver operating characteristic curves of all models in the training a and test b sets. Model 1 clinical factors model, Model 2 clinical factors and imaging indexes model, Model 3 clinical factors and Radscore model, Model 4 combined model

Fig. 5. The nomogram of Model 2. The sum of all factors equals the total points. Model 2 clinical factors and imaging indexes model, LDL low-density lipoprotein, HbA1c glycated haemoglobin, CACS coronary artery calcium score, CT-FFR computed tomography-based fractional flow reserve, DS diameter stenosis, PCATA pericoronary adipose tissue attenuation

Fig. 6. Decision curve analysis for the four models in the training a and test b sets. Model 1 clinical factors model, Model 2 clinical factors and imaging indexes model, Model 3 clinical factors and Radscore model, Model 4 combined model