Led by principal investigator and radiologist Carlo N. De Cecco MD, PhD, the Translational Laboratory for Cardiothoracic Imaging and Artificial Intelligence develops, evaluates, and implements novel cardiothoracic imaging techniques, analysis methods such as radiomics, and artificial intelligence applications. The goal: improving the accuracy of imaging and imaging interpretation and optimizing radiological workflow, all to benefit patient care.
Research Spotlight
With our international colleagues, visiting scholar Dr. Cherry Kim, and Dr. Dong Hyun Yang from Asan Medical Center in South Korea, we are working on the "Quantitative Analysis of Chest X-Ray for Diagnosing Cardiomediastinal Abnormality (ADC) Study" initiative. This multinational, multi-center, retrospective, observational study aims to create a normal reference standard for age and gender of the cardiovascular border indices automatically assessed by AI. We then aim to use this standard to calculate the z-score for each cardiovascular border index, and measure the clinical utility of the cardiovascular border indices across various test datasets. Results are anticipated soon, as is further collaboration!
Partners: Siemens Healthineers
Collaboration is key for artificial intelligence development, validation, and implementation. That's why we're working with Siemens Healthineers to develop a convolutional neural network algorithm for automatic thoracic aorta sizing. we then performance test in a heterogeneous population with multivendor CT datasets. Learn more.
Canon Young Investigator Award
Dr. Alexander Razavi is a finalist for the Canon Young Investigator Award presented by the Society of Cardiovascular Computed Tomography (SCCT). He will present his abstract "Discordance Between Coronary Artery Calcium Area and Density Predicts Long-term Atherosclerotic Cardiovascular Disease Risk" on Saturday, July 16 during the SCCT Annual Meeting.
Talks and Presentations
We maintain a busy schedule of invited talks and presentations at local, national, and interntational conferences and peer institutions. Here's where we'll be next:
- Artificial Intelligence and radiomics in cardiac imaging. Invited Lecturer. American Roentgen Ray Scholars (ARRS) Annual Conference. New Orleans, LA, USA. May 5, 2021.
- The new doctor-patient-imager relationship: Where does AI fit? Invited Lecturer. Society of Cardiac Computed Tomography (SCCT) Annual Conference, Las Vegas, NV, USA, July 7, 2022.
- Artifical intelligences applied in cardiac CT. Invited Keynote Lecturer. 20th Asian Oceanian Congress of Radiology & 78th Korean Congress of Radiology, Seoul, South Korea, Sept. 24, 2022.
- Artificial intelligences applied to the coronaries. Invited Lecturer. Italian Society of Medical Radiology (SIRM) Annual Meeting. Rome, Italy, Oct. 9, 2022.
- Predictive imaging: An AI imaging pipeline for CV risk. Invited Lecturer. American Heart Association (AHA) annual conference. Chicago, IL. USA. Nov. 7, 2022.
- Dynamic myocardial CT perfusion: Technical consideration and clinical evidence. Invited Lecturer. Comprehensive Cardiac CT: CTP and CT-FR Workshop. Society of Cardiac Computed Tomography (SCCT) Online Course, Nov. 14, 2022.
Talks and presentations completed in second half, 2021:
- Radiologic Society of North America, November 29-December 2
- Korean Society of Thoracic Radiology, October 23
- Czech Congress of Radiology, October 12
- Society of Advanced Body Imaging 2021 Annual Meeting, Boston, MA, October 8-12
- North American Society of Cardiac Imaging 2021 Annual Meeting, September 17-21
- Innovation in Patient-Centered Multimodality Cardiac Imaging, Milan, Italy, September 17
- Society of Cardiac Computed Tomography AI and Machine Learning in Cardiovascular CT Symposium, September 17
- Society of Cardiac Computed Tomography CTP and CTFFR Workshop, September 11
- Italian Society of Radiology Cardiac Section Meeting, Turin, Italy, September 8-10
- Society of Cardiac Computed Tomography 2021 Annual Meeting, July 16-17
- CADECI 2021 - Congreso Anual de Cardiologia Intervencionista, Guadalajara, Mexico, July 16-19
Recent Publications
Can Radiomics Help in the Identification of Vulnerable Coronary Plaque? Carlo N. DeCecco and Marly van Assen. Radiology. February 2023. https://doi.org/10.1148/radiol.223342
Artificial Intelligence in Cardiac Imaging: Where We Are and What We Want. Marly van Assen, Alexander C. Razavi, Seamus Whelton, Carlo N. DeCecco. European Heart Journal. February 2023. 10.1093/eurheartj/ehac700
Discordance Between Coronary Artery Calcium Area and Density Predicts Long-Term Atherosclerotic Cardiovasculary Disease Risk. Alexander C. Razavi, Marly van Assen, Carlo N. DeCecco, Zeina A. Dardari, Daniel S. Berman, Matthew J. Budoff, Michael D. Miedema, Khurram Nasir, Alan Rozanski, John A. Rumberger, Leslee J. Shaw, Laurence S. Sperling, Seamus P. Whelton, Martin Bodtker Mortensen, Michael J. Blaha, Omar, Dzaye. JACC Cardiovascular Imaging. November 2022. 10.1016/j.jcmg.2022.06.007
Evolving Role of Calcium Density in Coronary Artery Calcium Scoring and Atherosclerotic Cardiovascular Disease Risk. Alexander C. Razavi, Arthur S. Agatston, Leslee J. Shaw, Carlo N. DeCecco, Marly van Assen, Laurence S. Sperling, Macio S. Bittencourt, Melissa A. Daubert, Khurram Nasir, Roger S. Blumenthal, Martin Bodtker Mortensen, Seamus P. Whelton, Michael J. Blaha, Omar, Dzaye. JACC Cardiovascular Imaging. September 2022. https://doi.org/10.1016/j.jcmg.2022.02.026
Evaluating the Performance of a Convolutional Neural Network Algorithm for Measuring Thoracic Aortic Diameters in a Heterogeneous Population. Caterina B. Monti, Marly van Assen, Arthur E. Stillman, Scott J. Lee, Philipp Hoelzer, George S. K. Fung, Francesco Secchi, Francesco Sardanelli, and Carlo N. De Cecco. Radiology: Artificial Intelligence 4:2. https://doi.org/10.1148/ryai.210196
Bridging the gap between structured and free-form radiology reporting: A case study on coronary CT angiography. Amara Tariq, Marly van Assen, Carlo N. De Cecco, and Imon Banerjee. ACM Transactions on Computing for Healthcare. January 2022. doi.org/10.1145/3474831. https://dl.acm.org/doi/10.1145/3474831