A5022946341- AI in cancer detection
- Image Retrieval and Classification Techniques
- Global Cancer Incidence and Screening
- Digital Radiography and Breast Imaging
- Medical Image Segmentation Techniques
- Radiomics and Machine Learning in Medical Imaging
- Osteoarthritis Treatment and Mechanisms
- Advanced Vision and Imaging
- Cell Image Analysis Techniques
- Medical Imaging and Analysis
- Advanced Neuroimaging Techniques and Applications
- Rheumatoid Arthritis Research and Therapies
- Morphological variations and asymmetry
- Advanced Image and Video Retrieval Techniques
- Bone and Joint Diseases
- Glaucoma and retinal disorders
- Lower Extremity Biomechanics and Pathologies
- Cancer-related molecular mechanisms research
- Image Processing Techniques and Applications
- Gaze Tracking and Assistive Technology
- Dental Radiography and Imaging
- Cancer Risks and Factors
- COVID-19 diagnosis using AI
- Tactile and Sensory Interactions
- Spine and Intervertebral Disc Pathology
University of Copenhagen
2003-2024
Radboud University Nijmegen
2023
Gentofte Hospital
2023
Radboud University Medical Center
2023
Quality of Life Research Center
2012-2013
ScienceScope (United Kingdom)
2012
Nordic Bioscience (Denmark)
2010-2011
University College London
2007-2009
IT University of Copenhagen
2001-2006
Mammographic risk scoring has commonly been automated by extracting a set of handcrafted features from mammograms, and relating the responses directly or indirectly to breast cancer risk. We present method that learns feature hierarchy unlabeled data. When learned are used as input simple classifier, two different tasks can be addressed: i) density segmentation, ii) mammographic texture. The proposed model at multiple scales. To control models capacity novel sparsity regularizer is...
Background Developments in artificial intelligence (AI) systems to assist radiologists reading mammograms could improve breast cancer screening efficiency. Purpose To investigate whether an AI system detect normal, moderate-risk, and suspicious a sample safely reduce radiologist workload evaluate across Breast Imaging Reporting Data System (BI-RADS) densities. Materials Methods This retrospective simulation study analyzed mammographic examination data consecutively collected from January...
Background Retrospective studies have suggested that using artificial intelligence (AI) may decrease the workload of radiologists while preserving mammography screening performance. Purpose To compare and performance for two cohorts women who underwent before after AI system implementation. Materials Methods This retrospective study included 50–69-year-old biennial in Capital Region Denmark. Before implementation (October 1, 2020, to November 17, 2021), all screenings involved double...
This paper presents a brain T1-weighted structural magnetic resonance imaging (MRI) biomarker that combines several individual MRI biomarkers (cortical thickness measurements, volumetric hippocampal shape, and texture). The method was developed, trained, evaluated using two publicly available reference datasets: standardized dataset from the Alzheimer's Disease Neuroimaging Initiative (ADNI) arm of Australian Imaging Biomarkers Lifestyle flagship study ageing (AIBL). In addition, by...
Clinical studies including thousands of magnetic resonance imaging (MRI) scans offer potential for pathogenesis research in osteoarthritis. However, comprehensive quantification all bone, cartilage, and meniscus compartments is challenging. We propose a segmentation framework fully automatic knee MRI. The combines multiatlas rigid registration with voxel classification was trained on manual segmentations varying configurations bones, cartilages, menisci. validation included high- low-field...
Abstract Background Screening mammography works better in fatty than dense breast tissue. Computerized assessment of parenchymal texture is a non-subjective method to obtain refined description tissue, potentially valuable addition density scoring for the identification women need supplementary imaging. We studied sensitivity screening by combination radiologist-assessed Breast Imaging Reporting and Data System (BI-RADS) score computer-assessed marker, resemblance (MTR), population-based...
Abstract Patients with severe COVID-19 have overwhelmed healthcare systems worldwide. We hypothesized that machine learning (ML) models could be used to predict risks at different stages of management and thereby provide insights into drivers prognostic markers disease progression death. From a cohort approx. 2.6 million citizens in Denmark, SARS-CoV-2 PCR tests were performed on subjects suspected for disease; 3944 cases had least one positive test subjected further analysis. from the...
Gaze as a sole input modality must support complex navigation and selection tasks. interaction combines specific eye movements graphic display objects (GDOs). This paper suggests unifying taxonomy of gaze principles. The deals with three types movements: fixations, saccades smooth pursuits GDOs: static, dynamic, or absent. is qualified through related research the first main contribution this paper. second part offers an experimental exploration single stroke gestures (SSGG). findings...
Mammographic density is a well-established risk factor for breast cancer. We investigated the association between three different methods of measuring or parenchymal pattern/texture on digitized film-based mammograms, and examined to what extent textural features independently jointly with can improve ability identify screening women at increased The study included 121 cases 259 age- time matched controls based cohort 14,736 negative mammograms from population-based programme in Denmark 2007...
Texture patterns have been shown to improve breast cancer risk segregation in addition area-based mammographic density. The additional value of texture pattern scores on top volumetric density measures a large screening cohort has never studied. Volumetric and were assessed automatically for the first available digital mammography (DM) examination 51,400 women (50–75 years age) participating Dutch biennial program between 2003 2011. assessment method was developed previous study validated...
Background Recent mammography-based risk models can estimate short-term or long-term breast cancer risk, but whether assessment may improve by combining these has not been evaluated. Purpose To determine improves when a diagnostic artificial intelligence (AI) system for lesion detection and mammographic texture model. Materials Methods This retrospective study included Danish women consecutively screened at mammography from November 2012 to December 2015 who had least 5 years of follow-up...
Breast density has been established as a major risk factor for breast cancer. We have previously demonstrated that mammographic texture resemblance (MTR), recognizing the local patterns of mammogram, is also cancer, independent percent density. examine if these findings generalize to another population. Texture were recorded in digitalized pre-diagnosis (3.7 years) film mammograms nested case–control study within Dutch screening program (S1) comprising 245 cancers and 250 matched controls....
This paper examines gaze gestures and their applicability as a generic selection method for gaze-only controlled interfaces. The explored here is the Single Gaze Gesture (SGG), i.e. consisting of single point-to-point eye movement. Horizontal vertical, long short SGGs were evaluated on two tracking devices (Tobii/QuickGlance (QG)). main findings show that there significant difference in times between SGGs, vertical horizontal selections, well different systems.
A longitudinal study was used to investigate the quantification of osteoarthritis and prediction tibial cartilage loss by analysis tibia trabecular bone from magnetic resonance images knees. The Kellgren Lawrence (KL) grades were determined radiologists levels assessed a segmentation process. Aiming quantify potentially capture structure anatomy, machine learning approach set texture features for training classifier recognize knee with radiographic osteoarthritis. Using cross‐validation,...
The currently recommended double reading of all screening mammography examinations is an economic burden for programs. sensitivity higher women with low breast density than high density. One may therefore ask whether single could replace at least We addressed this question using data from a program where the radiologists coded their readings independently.Data include in Capital Region Denmark 1 November 2012 to 31 December 2013. Outcome was assessed by linkage Danish Pathology Register....
This paper introduces and explains the concept of single stroke gaze gestures. Some preliminary results are presented which indicate potential efficiency this interaction method we show how could be implemented for benefit disabled users generally it integrated with dwell to create a new dimension in controlled interfaces.
We consider the measurement of image structure using linear filters, in particular derivative-of-Gaussian (DtG) which are an important model V1 simple cells and widely used computer vision, whether such measurements can determine local symmetry. show that even a single filter be sensitive to symmetry, sense specific responses rule it out. state prove necessary sufficient, readily computable, criterion for symmetry-sensitivity. use six filters second order DtG family have patterns joint...
Early studies reported a 4- to 6-fold risk of breast cancer between women with extremely dense and fatty breasts. As most early were case-control studies, we took advantage population-based screening program study density incidence in cohort design. In the Capital Region, Denmark, aged 50 69 are invited biennially. Women screened November 2012 December 2017 included, classified by BI-RADS code, version 4, at first screen after recruitment. followed up for incident cancer, including ductal...