A5057583165- Medical Image Segmentation Techniques
- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Medical Imaging Techniques and Applications
- Advanced MRI Techniques and Applications
- Advanced Neural Network Applications
- Medical Imaging and Analysis
- Advanced Neuroimaging Techniques and Applications
- MRI in cancer diagnosis
- Cell Image Analysis Techniques
- Generative Adversarial Networks and Image Synthesis
- Advanced X-ray and CT Imaging
- Retinal Imaging and Analysis
- Digital Imaging for Blood Diseases
- COVID-19 diagnosis using AI
- Cardiac Valve Diseases and Treatments
- Lung Cancer Diagnosis and Treatment
- Glaucoma and retinal disorders
- Cardiac Imaging and Diagnostics
- Image and Object Detection Techniques
- Cutaneous Melanoma Detection and Management
- Image Retrieval and Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Retinal Diseases and Treatments
- Corneal surgery and disorders
Eindhoven University of Technology
2016-2025
Utrecht University
2012-2025
University Medical Center Utrecht
2016-2025
Heidelberg University
2001-2023
University Hospital Heidelberg
2001-2023
Philips (Germany)
2023
Philips (Netherlands)
2023
Fontys University of Applied Sciences
2021
IT University of Copenhagen
2021
Rigshospitalet
2021
Medical image registration is an important task in medical processing. It refers to the process of aligning data sets, possibly from different modalities (e.g., magnetic resonance and computed tomography), time points follow-up scans), and/or subjects (in case population studies). A large number methods for are described literature. Unfortunately, there not one method that works all applications. We have therefore developed elastix, a publicly available computer program intensity-based...
Mutual information has developed into an accurate measure for rigid and affine monomodality multimodality image registration. The robustness of the is questionable, however. A possible reason this absence spatial in measure. present paper proposes to include by combining mutual with a term based on gradient images be registered. not only seeks align locations high magnitude, but also aims similar orientation gradients at these locations. Results both standard as well normalized are presented...
EMPIRE10 (Evaluation of Methods for Pulmonary Image REgistration 2010) is a public platform fair and meaningful comparison registration algorithms which are applied to database intrapatient thoracic CT image pairs. Evaluation nonrigid techniques nontrivial task. This compounded by the fact that researchers typically test only on their own data, varies widely. For this reason, reliable assessment different has been virtually impossible in past. In work we present results launch phase...
A popular technique for nonrigid registration of medical images is based on the maximization their mutual information, in combination with a deformation field parameterized by cubic B-splines. The coordinate mapping that relates two found using an iterative optimization procedure. This work compares performance eight methods: gradient descent (with different step size selection algorithms), quasi-Newton, nonlinear conjugate gradient, Kiefer-Wolfowitz, simultaneous perturbation,...
An automatic method for delineating the prostate (including seminal vesicles) in three‐dimensional magnetic resonance scans is presented. The based on nonrigid registration of a set prelabeled atlas images. Each image nonrigidly registered with target patient image. Subsequently, deformed label images are fused to yield single segmentation proposed evaluated 50 clinical scans, which were manually segmented by three experts. Dice similarity coefficient (DSC) used quantify overlap between and...
We present a stochastic gradient descent optimisation method for image registration with adaptive step size prediction. The is based on the theoretical work by Plakhov and Cruz (J. Math. Sci. 120(1):964–973, 2004). Our main methodological contribution derivation of an image-driven mechanism to select proper values most important free parameters method. selection employs general characteristics cost functions that commonly occur in intensity-based registration. Also, convergence conditions...
This paper presents a robust and fully automatic filter-based approach for retinal vessel segmentation. We propose new filters based on 3D rotating frames in so-called orientation scores, which are functions the Lie-group domain of positions orientations [Formula: see text]. By means wavelet-type transform, 2D image is lifted to score, where elongated structures disentangled into their corresponding planes. In text], vessels enhanced by multi-scale second-order Gaussian derivatives...
The introduction of fast digital slide scanners that provide whole images has led to a revival interest in image analysis applications pathology. Segmentation cells and nuclei is an important first step towards automatic digitized microscopy images. We therefore developed automated segmentation method works with hematoxylin eosin (H&E) stained breast cancer histopathology images, which represent regions slides. procedure can be divided into four main steps: 1) pre-processing color unmixing...
In a multi-atlas based segmentation procedure, propagated atlas segmentations must be combined in label fusion process. Some current methods deal with this problem by using selection to construct an set either prior or after registration. Other estimate the performance of and use as weight This paper proposes selective iterative method for level estimation (SIMPLE), which combines both strategies procedure. subsequent iterations refines estimated selected atlases. For dataset 100 MR images...
Accurate segmentation of tubular, network-like structures, such as vessels, neurons, or roads, is relevant to many fields research. For the topology their most important characteristic; particularly preserving connectedness: in case vascular networks, missing a connected vessel entirely alters blood-flow dynamics. We introduce novel similarity measure termed centerlineDice (short clDice), which calculated on inter-section masks and (morphological) skeleta. theoretically prove that clDice...
Accurate segmentation of infant brain magnetic resonance (MR) images into white matter (WM), gray (GM), and cerebrospinal fluid is an indispensable foundation for early studying growth patterns morphological changes in neurodevelopmental disorders. Nevertheless, the isointense phase (approximately 6-9 months age), due to inherent myelination maturation process, WM GM exhibit similar levels intensity both T1-weighted T2-weighted MR images, making tissue very challenging. Although many efforts...
Medical images that are to be registered for clinical application often contain both structures deform and ones remain rigid. Nonrigid registration algorithms do not model properties of different tissue types may result in deformations rigid structures. In this article a local rigidity penalty term is proposed which included the function order penalize deformation objects. This can used any representation field capable modelling locally transformations. By using B‐spline field, fast...
Deformable image registration can be time consuming and often needs extensive parameterization to perform well on a specific application. We present deformable method based 3-D convolutional neural network, together with framework for training such network. The network directly learns transformations between pairs of images. is trained synthetic random which are applied small set representative images the desired Training, therefore, does not require manually annotated ground truth...