3D image fusion may help diagnose CAD
Click Here to Manage Email Alerts
A novel method that incorporates information from cardiac CT and cardiac MRI aided cardiologists in diagnosing CAD, according to a study published in Radiology: Cardiothoracic Imaging.
“For the first time, our technique shows all pathologic aspects of CAD assessable by radiologic techniques within one detailed, high-quality, 3D rendering (anatomy, coronary stenoses, their hemodynamic relevance, resulting hypoperfusion, and resulting myocardial scar),” Jochen von Spiczak, MD, MSc, radiologist and computer scientist at the Institute of Diagnostic and Interventional Radiology at University Hospital Zurich, told Healio. “Using this image, coronary stenoses and their hemodynamic relevance can be correlated to myocardial ischemia and scar very easily and with a high degree of certainty. Thereby, the technique allows for an easier and possibly more accurate identification of patients and coronary stenoses that are likely to benefit from revascularization (by either coronary intervention or bypass surgery).”
Suspected or known CAD
Researchers analyzed data from 17 patients (mean age, 54 years; one woman) who underwent cardiac MRI and cardiac CT for suspected or known CAD between May 2009 and January 2016. Software was developed to facilitate multimodal 3D image fusion.
“An accurate diagnosis is essential to find an optimal therapy for each individual patient,” von Spiczak said in an interview. “In today’s clinical routine, cardiac CT and MR images are most often interpreted separately from each other. Correlation of different findings is done accordingly to standard supply territories of the coronary arteries. By combining information on different pathologic aspects of the disease in a single image, complementary image data can be interpreted in a very intuitive and quick but at the same time highly accurate fashion.”
Postprocessing of CT data was performed and included CT fractional flow reserve calculation, segmentation of heart contours and the coronary tree, and color coding of the coronary tree based on CT FFR. Cardiac MRI data were also postprocessed, including co-registration of cardiac MRI data to CT data, left ventricular segmentation in cardiac MRI perfusion and cardiac MRI late gadolinium enhancement and the projection of data from cardiac MRI perfusion and late gadolinium enhancement onto high spatial resolution LV from CT.
Image quality for CT coronary angiography was rated as good to excellent, with a mean score of 2.6 out of 3. This method showed significant stenoses in 41% of patients. Most patients (94%) had CT FFR calculated, which showed pathologic flow in 41% of patients. Of these patients, six had significant stenoses at CT coronary angiography.
Hypoperfusion was identified by cardiac MRI in 47% of patients with a mean ischemic burden of 17%. Myocardial scar was also seen by cardiac MRI late gadolinium enhancement in 18% of patients with a mean scar burden of 7%.
Researchers determined uncertain findings in 47% of patients when CT coronary angiography and cardiac MRI were read via conventional 2D methods. Most divergent findings (75%) were solved when information from CT FFR and 3D image fusion were added.
Potential use
“In the current stage of development, our method is only used for research,” von Spiczak told Healio. “However, it would be possible to integrate the technique in the daily clinical routine if a dedicated software tool was developed for the task at hand in the future. Especially for complicated cases, the combination of CT and MR imaging and 3D image fusion might help to come to a more accurate diagnosis.” – by Darlene Dobkowski
For more information:
Jochen von Spiczak, MD, MSc, can be reached at jochen.vonspiczak@usz.ch.
Disclosures: The study was supported in part by Bayer Healthcare and the Swiss National Science Foundation. Von Spiczak reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.
Collapse
Click Here to Manage Email Alerts