Dr. Bob D. de Vos Profile

Dr. Bob D. de Vos

Assistant Professor at Amsterdam UMC

SPIE Involvement:

Author

Publications (17)

SPIE Journal Paper | 31 May 2022

Generative models for reproducible coronary calcium scoring

Sanne G. Van Velzen, Bob de Vos, Julia M. Noothout, Helena Verkooijen, Max Viergever, Ivana Išgum

JMI, Vol. 9, Issue 05, 052406, (May 2022) https://doi.org/10.1117/12.10.1117/1.JMI.9.5.052406

KEYWORDS: Calcium, Heart, Image segmentation, Computed tomography, Arteries, Gallium nitride, Tissues, Chest, Chemical vapor deposition, Visualization

Read Abstract +

Proceedings Article | 4 April 2022 Presentation + Paper

Generative adversarial network for coronary artery plaque synthesis in coronary CT angiography

Zhiwei Zhai, Yining Wang, Bob de Vos, Julia M.H. Noothout, Nils Planken, Ivana Isgum

Proceedings Volume 12032, 120321R (2022) https://doi.org/10.1117/12.2605905

KEYWORDS: Realistic image synthesis, Machine learning, Arteries, Computed tomography, Angiography, Image segmentation, Image analysis

Read Abstract +

Detection and quantification of atherosclerotic plaque in the coronary arteries is important in cardiovascular risk analysis. Atherosclerotic plaque can be visualized using coronary computed tomography angiography (CCTA). Manual identification and segmentation of plaque is a complex task that requires high level of expertise. Hence, several automatic approaches have been designed. Automatic methods employing deep learning have shown to outperform conventional approaches, but they are hampered by the requirement on the availability of large and diverse training data. To address this, we designed a method that synthesizes calcified and non-calcified atherosclerotic plaque in the coronary arteries. First, we generate plaque geometry using conventional image analysis approach that varies the radius, length and angle of a plaque and we use this to generate a crude inpainting of a plaque on a target artery. Thereafter, we employ a conditional generative adversarial network (GAN) to synthesize the plaque texture in CCTA. The generator is trained to generate fake images with realistic appearance. The discriminator is trained to distinguish the synthesized fake and the real images. The data set for training and evaluation of the plaque synthesis contained CCTA scans of 102 patients (50 training, 52 testing) with manually annotated calcified and non-calcified plaque. To evaluate performance of the synthesis method, we compared CCTA patches with real and synthesized plaque. The evaluation resulted in mean (standard deviation) structural similarity index of 0.99 (0.01), peak signal noise ratio of 73.99 (5.52) and mean absolute error of 5.56 (3.23) HU. To evaluate whether synthesized data enables plaque segmentation, an additional set of CCTA scans of 92 patients without visible plaque was collected. In these scans, plaque was synthesized using the developed approach, containing in total 615 calcified and 544 non-calcified plaque lesions. The synthesized data was used to train a 3D UNet for segmenting calcified and non-calcified plaque lesions. Automatic segmentation which was trained with real data only resulted in Dice coefficients of 0.68 and 0.35 for calcified and non-calcified plaque, respectively. This was significantly improved by pretraining the network with synthetic data and refining it with real data, which resulted in a Dice coefficients of 0.70 (p=0.03) and 0.36 (p=0.02) for calcified and non-calcified plaque, respectively. The results demonstrate that training with CCTA scans with automatically synthesized calcified and non-calcified plaque improves the performance of plaque segmentation.

Proceedings Article | 15 February 2021 Presentation + Paper

Automatic segmentation of the olfactory bulbs in MRI

Julia M. H. Noothout, Elbrich M. Postma, Sanne Boesveldt, Bob D. de Vos, Paul A. M. Smeets, Ivana Išgum

Proceedings Volume 11596, 115961J (2021) https://doi.org/10.1117/12.2580354

KEYWORDS: Magnetic resonance imaging, Pathology, Nervous system, Convolutional neural networks, Alzheimer's disease

Read Abstract +

Proceedings Article | 15 February 2021 Presentation + Paper

Unsupervised super-resolution: creating high-resolution medical images from low-resolution anisotropic examples

Jörg Sander, Bob de Vos, Ivana Išgum

Proceedings Volume 11596, 115960E (2021) https://doi.org/10.1117/12.2580412

KEYWORDS: Super resolution, Medical imaging, Image resolution, 3D image processing, Spatial resolution, 3D acquisition

Read Abstract +

Proceedings Article | 10 March 2020 Presentation + Paper

Mutual information for unsupervised deep learning image registration

Bob de Vos, Bas H. M. van der Velden, Jörg Sander, Kenneth G. A. Gilhuijs, Marius Staring, Ivana Išgum

Proceedings Volume 11313, 113130R (2020) https://doi.org/10.1117/12.2549729

KEYWORDS: Image registration, Magnetic resonance imaging, Breast, Medical imaging, Image enhancement, Infrared imaging, Image restoration, Image contrast enhancement, Cardiovascular magnetic resonance imaging, Quantitative analysis

Read Abstract +

Proceedings Article | 10 March 2020 Presentation + Paper

Coronary artery calcium scoring: Can we do better?

Sanne G. Van velzen, Bob de Vos, Helena Verkooijen, Tim Leiner, Max Viergever, Ivana Išgum

Proceedings Volume 11313, 113130G (2020) https://doi.org/10.1117/12.2549557

KEYWORDS: Calcium, Computed tomography, Arteries, Chest, Medical imaging, Radiology

Read Abstract +

Coronary artery calcification (CAC) is a strong and independent predictor of cardiovascular disease (CVD) that can be quantified in CT scans showing the heart. CAC lesions are defined as lesions in the coronary arteries with image intensity above 130 HU. The use of a threshold may lead to under- or over-estimation of the amount of CAC and, hence, to incorrect cardiovascular categorization of patients. This is especially pronounced in CT scans without ECG-synchronization where lesions are more subject to cardiac motion and partial volume effects. To address this, we propose a method for quantification of CAC without a threshold. A set of 373 cardiac and 1181 chest CT scans was included to develop the method and a set of 21 scan-rescan pairs (42 scans) was included for final evaluation. Assuming that the attenuation of CAC is superimposed on the attenuation of the artery, we aimed to separate the CAC from the coronary arteries by employing a CycleGAN to generate a synthetic image without CAC from an image containing CAC and vice versa. By subtracting the synthetic image without CAC from the image with CAC, a CAC map is created. The CAC-map can subsequently be used to identify and quantify CAC. The ground truth, i.e. the true amount of CAC, can not be established, therefore, in this work the results generated by the method are compared with clinical calcium scoring in terms of reproducibility. The average relative difference between the calcium scores in scan-rescan pairs of scans was 50% with the proposed method and 86% for the conventional method. Moreover, the correlation between CAC pseudo masses in scan-rescan pairs was 0.92 with the proposed method and 0.89 with conventional calcium scoring. Our proposed method is able to identify and quantify CAC lesions in CT scans without using an intensity level thresholding. This might allow for more reproducible quantification of CAC in CT scans made without ECG synchronization, and, therefore, it might allow more accurate CVD risk prediction.

Proceedings Article | 15 March 2019 Presentation + Paper

Towards increased trustworthiness of deep learning segmentation methods on cardiac MRI

Jörg Sander, Bob de Vos, Jelmer Wolterink, Ivana Išgum

Proceedings Volume 10949, 1094919 (2019) https://doi.org/10.1117/12.2511699

KEYWORDS: Image segmentation, Calibration, Magnetic resonance imaging, Cardiovascular magnetic resonance imaging, Medical imaging, Reliability, Uncertainty analysis, Human-computer interaction

Read Abstract +

Proceedings Article | 13 March 2019 Presentation + Paper

Response monitoring of breast cancer on DCE-MRI using convolutional neural network-generated seed points and constrained volume growing

Bas H. van der Velden, Bob de Vos, Claudette Loo, Hugo Kuijf, Ivana Išgum, Kenneth G. Gilhuijs

Proceedings Volume 10950, 109500D (2019) https://doi.org/10.1117/12.2508358

KEYWORDS: Tumors, Magnetic resonance imaging, Breast cancer, Breast, Image segmentation, Neural networks, Blood vessels, Cancer, Surgery, Medical imaging

Read Abstract +

SPIE Journal Paper | 11 December 2018

Impact of automatically detected motion artifacts on coronary calcium scoring in chest computed tomography

Jurica Šprem, Bob de Vos, Nikolas Lessmann, Pim de Jong, Max Viergever, Ivana Išgum

JMI, Vol. 5, Issue 04, 044007, (December 2018) https://doi.org/10.1117/12.10.1117/1.JMI.5.4.044007

KEYWORDS: Computed tomography, Chest, Chemical vapor deposition, Motion detection, Calcium, Medical imaging, Medical research, Arteries, Lung, Heart

Read Abstract +

Proceedings Article | 2 March 2018 Paper

Automatic segmentation of thoracic aorta segments in low-dose chest CT

Julia M. H. Noothout, Bob de Vos, Jelmer Wolterink, Ivana Išgum

Proceedings Volume 10574, 105741S (2018) https://doi.org/10.1117/12.2293114

KEYWORDS: Aorta, Image segmentation, Computed tomography, Image analysis, Convolutional neural networks

Read Abstract +

Proceedings Article | 24 February 2017 Presentation + Paper

Classification of coronary artery calcifications according to motion artifacts in chest CT using a convolutional neural network

Jurica Šprem, Bob de Vos, Pim de Jong, Max Viergever, Ivana Išgum

Proceedings Volume 10133, 101330R (2017) https://doi.org/10.1117/12.2253669

KEYWORDS: Convolutional neural networks, Computed tomography, Chest, Neural networks, Arteries, Lung, Scanners, Calcium, Lung cancer, Motion detection, Electrocardiography, System identification, Image resolution, Motion analysis

Read Abstract +

Proceedings Article | 24 March 2016 Paper

Deep convolutional neural networks for automatic coronary calcium scoring in a screening study with low-dose chest CT

Nikolas Lessmann, Ivana Išgum, Arnaud A. Setio, Bob de Vos, Francesco Ciompi, Pim de Jong, Matthjis Oudkerk, Willem P. Th. Mali, Max Viergever, Bram van Ginneken

Proceedings Volume 9785, 978511 (2016) https://doi.org/10.1117/12.2216978

KEYWORDS: Arteries, Neural networks, Convolutional neural networks, Calcium, Chest, Computed tomography, Lung cancer, Heart, Convolution, Chemical vapor deposition

Read Abstract +

Proceedings Article | 21 March 2016 Paper

Genome-wide association study of coronary and aortic calcification in lung cancer screening CT

Bob de Vos, Jessica van Setten, Pim de Jong, Willem Mali, Matthijs Oudkerk, Max Viergever, Ivana Išgum

Proceedings Volume 9784, 97841L (2016) https://doi.org/10.1117/12.2217024

KEYWORDS: Genetics, Lung cancer, Computed tomography, Arteries, Calcium, Bone, Computer aided diagnosis and therapy, Chest, 3D image processing, Computer aided design

Read Abstract +

Proceedings Article | 21 March 2016 Paper

Automatic detection of cardiovascular risk in CT attenuation correction maps in Rb-82 PET/CTs

Ivana Išgum, Bob de Vos, Jelmer Wolterink, Damini Dey, Daniel Berman, Mathieu Rubeaux, Tim Leiner, Piotr Slomka

Proceedings Volume 9784, 978405 (2016) https://doi.org/10.1117/12.2216992

KEYWORDS: Calcium, Algorithm development, Computed tomography, Arteries, Chest, Positron emission tomography

Read Abstract +

Proceedings Article | 21 March 2016 Paper

2D image classification for 3D anatomy localization: employing deep convolutional neural networks

Bob de Vos, Jelmer Wolterink, Pim de Jong, Max Viergever, Ivana Išgum

Proceedings Volume 9784, 97841Y (2016) https://doi.org/10.1117/12.2216971

KEYWORDS: Convolutional neural networks, Image classification, 3D image processing, Neural networks, Chest, Computed tomography, Machine learning, Heart, Image analysis, Medical imaging, Receivers

Read Abstract +

Proceedings Article | 21 March 2016 Paper

Supervised novelty detection in brain tissue classification with an application to white matter hyperintensities

Hugo Kuijf, Pim Moeskops, Bob de Vos, Willem Bouvy, Jeroen de Bresser, Geert Jan Biessels, Max Viergever, Koen Vincken

Proceedings Volume 9784, 978421 (2016) https://doi.org/10.1117/12.2216358

KEYWORDS: Brain, Tissues, Magnetic resonance imaging, Neuroimaging, Pathology, Image segmentation, Probability theory, Machine learning, Pattern recognition, Detection and tracking algorithms, Data centers, Medical imaging, Brain mapping

Read Abstract +

Proceedings Article | 20 March 2015 Paper

Automatic machine learning based prediction of cardiovascular events in lung cancer screening data

Bob de Vos, Pim de Jong, Jelmer Wolterink, Rozemarijn Vliegenthart, Geoffrey Wielingen, Max Viergever, Ivana Išgum

Proceedings Volume 9414, 94140D (2015) https://doi.org/10.1117/12.2082242

KEYWORDS: Calcium, Lung cancer, Image analysis, Computed tomography, Feature extraction, Machine learning, Chest, Algorithm development, Pattern recognition, Scanners

Read Abstract +

Showing 5 of 17 publications

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years