385 research outputs found
Author Correction: High-dimensional detection of imaging response to treatment in multiple sclerosis (npj Digital Medicine, (2019), 2, 1, (49), 10.1038/s41746-019-0127-8)
No full textThe original version of the published Article listed the incorrect affiliations for Jorge Cardoso, Carole H. Sudre, and Sebastien Ourselin. These authors’ affiliations are now correctly noted as School of Biomedical Engineering & Imaging Sciences, King’s College London, London, UK. Additionally, a statement from the Acknowledgments section was omitted. The Acknowledgments have been updated to include the following: “We are grateful to the MS Society for its support of the Multiple Sclerosis Research Centre at the Queen Square Institute of Neurology where this work was completed.” The HTML and PDF versions of the Article have been corrected
Author Correction: High-dimensional detection of imaging response to treatment in multiple sclerosis (npj Digital Medicine, (2019), 2, 1, (49), 10.1038/s41746-019-0127-8)
No full textThe original version of the published Article listed the incorrect affiliations for Jorge Cardoso, Carole H. Sudre, and Sebastien Ourselin. These authors’ affiliations are now correctly noted as School of Biomedical Engineering & Imaging Sciences, King’s College London, London, UK. Additionally, a statement from the Acknowledgments section was omitted. The Acknowledgments have been updated to include the following: “We are grateful to the MS Society for its support of the Multiple Sclerosis Research Centre at the Queen Square Institute of Neurology where this work was completed.” The HTML and PDF versions of the Article have been corrected
MSJ852093_supplemental_material – Supplemental material for Periventricular magnetisation transfer ratio abnormalities in multiple sclerosis improve after alemtuzumab
No full textSupplemental material, MSJ852093_supplemental_material for Periventricular magnetisation transfer ratio abnormalities in multiple sclerosis improve after alemtuzumab by J William L Brown, Ferran Prados Carrasco, Arman Eshaghi, Carole H Sudre, Tom Button, Matteo Pardini, Rebecca S Samson, Sebastien Ourselin, Claudia AM Gandini Wheeler-Kingshott, Joanne L Jones, Alasdair J Coles and Declan T Chard in Multiple Sclerosis Journal</p
MSJ865739_supplemental_material – Supplemental material for Single-subject structural cortical networks in clinically isolated syndrome
No full textSupplemental material, MSJ865739_supplemental_material for Single-subject structural cortical networks in clinically isolated syndrome by Sara Collorone, Ferran Prados, Marloes HJ Hagens, Carmen Tur, Baris Kanber, Carole H Sudre, Carsten Lukas, Claudio Gasperini, Celia Oreja-Guevara, Micaela Andelova, Olga Ciccarelli, Mike P Wattjes, Sebastian Ourselin, Daniel R Altmann, Betty M Tijms, Frederik Barkhof and Ahmed T Toosy in Multiple Sclerosis Journal</p
MSJ841810_supplemental_table_2 – Supplemental material for Magnetisation transfer ratio abnormalities in primary and secondary progressive multiple sclerosis
No full textSupplemental material, MSJ841810_supplemental_table_2 for Magnetisation transfer ratio abnormalities in primary and secondary progressive multiple sclerosis by James William L Brown, Azmain Chowdhury, Baris Kanber, Ferran Prados Carrasco, Arman Eshaghi, Carole H Sudre, Matteo Pardini, Rebecca S Samson, Steven HP van de Pavert, Claudia Gandini Wheeler-Kingshott and Declan T Chard in Multiple Sclerosis Journal</p
MSJ841810_supplemental_table_1 – Supplemental material for Magnetisation transfer ratio abnormalities in primary and secondary progressive multiple sclerosis
No full textSupplemental material, MSJ841810_supplemental_table_1 for Magnetisation transfer ratio abnormalities in primary and secondary progressive multiple sclerosis by James William L Brown, Azmain Chowdhury, Baris Kanber, Ferran Prados Carrasco, Arman Eshaghi, Carole H Sudre, Matteo Pardini, Rebecca S Samson, Steven HP van de Pavert, Claudia Gandini Wheeler-Kingshott and Declan T Chard in Multiple Sclerosis Journal</p
Improving Error Detection in Deep Learning Based Radiotherapy Autocontouring Using Bayesian Uncertainty
No full textBayesian Neural Nets (BNN) are increasingly used for robust organ auto-contouring. Uncertainty heatmaps extracted from BNNs have been shown to correspond to inaccurate regions. To help speed up the mandatory quality assessment (QA) of contours in radiotherapy, these heatmaps could be used as stimuli to direct visual attention of clinicians to potential inaccuracies. In practice, this is non-trivial to achieve since many accurate regions also exhibit uncertainty. To influence the output uncertainty of a BNN, we propose a modified accuracy-versus-uncertainty (AvU) metric as an additional objective during model training that penalizes both accurate regions exhibiting uncertainty as well as inaccurate regions exhibiting certainty. For evaluation, we use an uncertainty-ROC curve that can help differentiate between Bayesian models by comparing the probability of uncertainty in inaccurate versus accurate regions. We train and evaluate a FlipOut BNN model on the MICCAI2015 Head and Neck Segmentation challenge dataset and on the DeepMind-TCIA dataset, and observed an increase in the AUC of uncertainty-ROC curves by 5.6% and 5.9%, respectively, when using the AvU objective. The AvU objective primarily reduced false positives regions (uncertain and accurate), drawing less visual attention to these regions, thereby potentially improving the speed of error detection.Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Computer Graphics and VisualisationPattern Recognition and Bioinformatic
Biomarkers of Migraine and Cluster Headache: Differences and Similarities
No full textObjective: This study was undertaken to identify magnetic resonance imaging (MRI) biomarkers that differentiate migraine from cluster headache patients and imaging features that are shared. Methods: Clinical, functional, and structural MRI data were obtained from 20 migraineurs, 20 cluster headache patients, and 15 healthy controls. Support vector machine algorithms and a stepwise removal process were used to discriminate headache patients from controls, and subgroups of patients. Regional between-group differences and association between imaging features and patients' clinical characteristics were also investigated. Results: The accuracy for classifying headache patients from controls was 80%. The classification accuracy for discrimination between migraine and controls was 89%, and for cluster headache and controls it was 98%. For distinguishing cluster headache from migraine patients, the MRI classifier yielded an accuracy of 78%, whereas MRI–clinical combined classification model achieved an accuracy of 99%. Bilateral hypothalamic and periaqueductal gray (PAG) functional networks were the most important MRI features in classifying migraine and cluster headache patients from controls. The left thalamic network was the most discriminative MRI feature in classifying migraine from cluster headache patients. Compared to migraine, cluster headache patients showed decreased functional interaction between the left thalamus and cortical areas mediating interoception and sensory integration. The presence of restlessness was the most important clinical feature in discriminating the two groups of patients. Interpretation: Functional biomarkers, including the hypothalamic and PAG networks, are shared by migraine and cluster headache patients. The thalamocortical pathway may be the neural substrate that differentiates migraine from cluster headache attacks with their distinct clinical features. ANN NEUROL 2023
Evaluation of Uncertainty-Aware Multi-software Ensembles for Hippocampal Segmentation
No full textAccurate hippocampal segmentation can be a useful tool for diagnosing and monitoring neurological conditions such as Alzheimer’s disease and epilepsy. While numerous automated segmentation methods exist, their clinical adoption remains limited. Reliable uncertainty assessment can enhance trust and facilitate clinical translation. This study evaluates five heterogeneous hippocampal segmentation methods InnerEye, ASHS, FastSurfer, HippoSeg, and FreeSurfer—across two dementia datasets and one epilepsy dataset. The sub-ensemble containing InnerEye, FastSurfer, and HippoSeg emerged as both accurate and efficient, highlighting the feasibility of balancing computational cost and performance. Additionally, ensemble-derived uncertainty quantification with sample variance, mutual information, and predictive entropy is shown to reduce inaccurate segmentations by flagging low-confidence cases, potentially providing a mechanism for automatically escalating ambiguous cases for expert assessment
- …
