Calcifications in the aortic wall, heart valves and coronary arteries are strong and independent predictors of cardiovascular disease (CVD) including myocardial infarction (MI), sudden cardiac death, and stroke. The Dutch lung cancer screening (NELSON) trial offers a possibility to investigate presence and risk of CVD in an asymptomatic high risk population.
In this project, we are developing automatic algorithms to measure existing and novel imaging biomarkers related to CVD in CT images from the NELSON trial. These biomarkers will enrich the phenotype information, enabling genetic studies and focusing on the identification of genes related to CVD risk. Finally, the combined information from imaging and genetics offers the possibility to investigate whether imaging markers and genetic analysis provide equivalent or complementary evidence with respect to CVD risk stratification.
Chest CT scan acquired in the Dutch-Belgian lung cancer screening trial. Calcifications in the coronary arteries and in the aorta that are visible in this image (high intensity areas), allow determination of the cardiovascular risks of this subject.
Papers in international journals
|1.||ConvNet-based localization of anatomical structures in 3D medical images. IEEE Transactions on Medical Imaging, 2017, vol. 36, nr. 7, pp. 1470-1481.|
|2.||Multiethnic Exome-Wide Association Study of Subclinical Atherosclerosis. Circulation. Cardiovascular genetics, 2016, vol. 9, nr. 6, pp. 511-520.|
|3.||Predominance of nonatherosclerotic internal elastic lamina calcification in the intracranial internal carotid artery. Stroke, 2016, vol. 47, nr. 1, pp. 221-3.|
Papers in conference proceedings
|1.||Automatic segmentation of thoracic aorta segments in low-dose chest CT. In: SPIE Medical Imaging, in press, 2018.|
|2.||End-to-end unsupervised deformable image registration with a convolutional neural network. In: Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: Third International Workshop, DLMIA 2017ML-CDS 2017, Held in Conjunction with MICCAI 2017, Quebec City, QC, Canada, September 14, Proceedings , 2017, pp. 204–212.|
|3.||Nonrigid image registration using multi-scale 3D convolutional neural networks. In: Medical Image Computing and Computer Assisted Intervention − MICCAI 2017: 20th International Conference, Quebec City, QC, Canada, September 11-13, 2017, Proceedings, Part 1, 2017, vol. 10433, pp. 232–239.|
|4.||Classification of coronary artery calcifications according to motion artifacts in chest CT using a convolutional neural network. In: SPIE Medical Imaging, 2017.|
|5.||Automatic Slice Identification in 3D Medical Images with a ConvNet Regressor. In: Deep Learning and Data Labeling for Medical Applications: First International Workshop, LABELS 2016, and Second International Workshop, DLMIA 2016, MICCAI 2016, Athens, Greece, 2016, pp. 161–169.|
|6.||2D image classification for 3D anatomy localization; employing deep convolutional neural networks. In: SPIE Medical Imaging, 2016, vol. 9784, pp. 97841Y-1-97841Y-7.|
|7.||Genome-Wide Association Study of Coronary and Aortic Calcification in Lung Cancer Screening CT. In: SPIE Medical Imaging, 2016, vol. 9784, pp. 97841L-1-97841L-6.|
|8.||Automatic machine learning based prediction of cardiovascular events in lung cancer screening data. In: SPIE Medical Imaging, 2015, vol. 9414, pp. 94140D.|
|1.||Direct coronary artery calcium scoring in low-dose chest CT using deep learning analysis. Radiological Society of North America, 103rd Annual Meeting, 2017.|
|2.||Increasing the Interscan Reproducibility of Coronary Calcium Scoring by Partial Volume Correction in Low-Dose non-ECG Synchronized CT: Phantom Study. Radiological Society of North America, 2015.|