Ivana Išgum

ORCID: 0000-0003-1869-5034
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Cardiac Imaging and Diagnostics
  • Advanced X-ray and CT Imaging
  • Medical Imaging Techniques and Applications
  • Radiomics and Machine Learning in Medical Imaging
  • Medical Image Segmentation Techniques
  • Neonatal and fetal brain pathology
  • Coronary Interventions and Diagnostics
  • Advanced MRI Techniques and Applications
  • Fetal and Pediatric Neurological Disorders
  • Lung Cancer Diagnosis and Treatment
  • Cardiac Valve Diseases and Treatments
  • Cerebrovascular and Carotid Artery Diseases
  • Radiation Dose and Imaging
  • AI in cancer detection
  • Advanced Neuroimaging Techniques and Applications
  • Advanced Radiotherapy Techniques
  • Medical Imaging and Analysis
  • Cardiovascular Function and Risk Factors
  • Neonatal Respiratory Health Research
  • Chronic Obstructive Pulmonary Disease (COPD) Research
  • Cardiovascular Disease and Adiposity
  • Acute Ischemic Stroke Management
  • Advanced Image Processing Techniques
  • Infant Development and Preterm Care
  • COVID-19 diagnosis using AI

Amsterdam University Medical Centers
2019-2025

University of Amsterdam
2019-2025

Cancer Center Amsterdam
2025

University Medical Center Utrecht
2013-2023

Amsterdam Neuroscience
2023

Amsterdam UMC Location University of Amsterdam
2021-2022

University Medical Center
2014-2022

Centrum Wiskunde & Informatica
2022

Utrecht University
2010-2021

Heidelberg University
2003-2020

Delineation of the left ventricular cavity, myocardium, and right ventricle from cardiac magnetic resonance images (multi-slice 2-D cine MRI) is a common clinical task to establish diagnosis. The automation corresponding tasks has thus been subject intense research over past decades. In this paper, we introduce "Automatic Cardiac Diagnosis Challenge" dataset (ACDC), largest publicly available fully annotated for purpose MRI (CMR) assessment. contains data 150 multi-equipments CMRI recordings...

10.1109/tmi.2018.2837502 article EN IEEE Transactions on Medical Imaging 2018-05-17

Noise is inherent to low-dose CT acquisition. We propose train a convolutional neural network (CNN) jointly with an adversarial CNN estimate routine-dose images from and hence reduce noise. A generator was trained transform into using voxelwise loss minimization. An discriminator simultaneously distinguish the output of images. The performance this used as for generator. Experiments were performed anthropomorphic phantom containing calcium inserts, well patient non-contrast-enhanced cardiac...

10.1109/tmi.2017.2708987 article EN IEEE Transactions on Medical Imaging 2017-05-26

Automatic segmentation in MR brain images is important for quantitative analysis large-scale studies with acquired at all ages. This paper presents a method the automatic of into number tissue classes using convolutional neural network. To ensure that obtains accurate details as well spatial consistency, network uses multiple patch sizes and convolution kernel to acquire multi-scale information about each voxel. The not dependent on explicit features, but learns recognise classification...

10.1109/tmi.2016.2548501 article EN IEEE Transactions on Medical Imaging 2016-03-30

A novel atlas-based segmentation approach based on the combination of multiple registrations is presented. Multiple atlases are registered to a target image. To obtain target, labels atlas images propagated it. The combined by spatially varying decision fusion weights. These weights derived from local assessment registration success. Furthermore, an selection procedure proposed that equivalent sequential forward statistical pattern recognition theory. method compared three existing...

10.1109/tmi.2008.2011480 article EN IEEE Transactions on Medical Imaging 2009-01-16

Background Although several deep learning (DL) calcium scoring methods have achieved excellent performance for specific CT protocols, their in a range of examination types is unknown. Purpose To evaluate the DL method automatic across wide and to investigate whether can adapt different examinations when representative images are added existing training data set. Materials Methods The study included 7240 participants who underwent various nonenhanced that heart: coronary artery (CAC) CT,...

10.1148/radiol.2020191621 article EN Radiology 2020-02-11

Coronary Artery Calcium Can Predict All-Cause Mortality and Cardiovascular Events on Low-Dose CT Screening for Lung CancerPeter C. Jacobs1 2, Martijn J. A. Gondrie1, Yolanda van der Graaf1, Harry de Koning3, Ivana Isgum4, Bram Ginneken5 Willem P. T. M. Mali2Audio Available | Share

10.2214/ajr.10.5577 article EN American Journal of Roentgenology 2012-02-22

<h3>Background</h3> Emphysema and small airway disease both contribute to chronic obstructive pulmonary (COPD), a characterised by accelerated decline in lung function. The association between the extent of emphysema male current former smokers function was investigated. <h3>Methods</h3> Current heavy participating cancer screening trial were recruited study all underwent CT. Spirometry performed at baseline 3-year follow-up. 15th percentile (Perc15) used assess severity emphysema....

10.1136/thx.2010.145995 article EN Thorax 2011-04-07

Heavy smokers undergoing screening with low-dose chest CT are affected by cardiovascular disease as much lung cancer. Low-dose scans acquired in enable quantification of atherosclerotic calcifications and thus identification subjects at increased risk. This paper presents a method for automatic detection coronary artery, thoracic aorta cardiac valve using two consecutive convolutional neural networks. The first network identifies labels potential according to their anatomical location the...

10.1109/tmi.2017.2769839 article EN IEEE Transactions on Medical Imaging 2017-11-03

Various types of atherosclerotic plaque and varying grades stenosis could lead to different management patients with a coronary artery disease. Therefore, it is crucial detect classify the type plaque, as well determine degree stenosis. This paper includes retrospectively collected clinically obtained CT angiography (CCTA) scans 163 patients. In these, centerlines arteries were extracted used reconstruct multi-planar reformatted (MPR) images for arteries. To define reference standard,...

10.1109/tmi.2018.2883807 article EN IEEE Transactions on Medical Imaging 2018-11-28

Epidemiological studies show that high circulating cystatin C is associated with risk of cardiovascular disease (CVD), independent creatinine-based renal function measurements. It unclear whether this relationship causal, arises from residual confounding, and/or a consequence reverse causation. The aim study was to use Mendelian randomization investigate causally related CVD in the general population. We incorporated participant data 16 prospective cohorts (n = 76,481) 37,126 measures and...

10.1016/j.jacc.2016.05.092 article EN cc-by-nc-nd Journal of the American College of Cardiology 2016-08-01
Coming Soon ...