1 code implementation • 5 Apr 2022 • Lucas Fidon, Michael Aertsen, Florian Kofler, Andrea Bink, Anna L. David, Thomas Deprest, Doaa Emam, Fr/'ed/'eric Guffens, András Jakab, Gregor Kasprian, Patric Kienast, Andrew Melbourne, Bjoern Menze, Nada Mufti, Ivana Pogledic, Daniela Prayer, Marlene Stuempflen, Esther Van Elslander, Sébastien Ourselin, Jan Deprest, Tom Vercauteren
Our method automatically discards the voxel-level labeling predicted by the backbone AI that are likely to violate expert knowledge and relies on a fallback atlas-based segmentation method for those voxels.
Importantly, we explore the inclusion of a transformer in the bottleneck of the U-Net architecture.
Label-set loss functions allow to train deep neural networks with partially segmented images, i. e. segmentations in which some classes may be grouped into super-classes.
Our approach allows a surgeon to build a graph of desired views, from which, once built, views can be manually selected and automatically servoed to irrespective of robot-patient frame transformation changes.
We perform an extensive evaluation of state-of-the-art (SOTA) Deep Neural Networks (DNNs) across multiple compute regimes, finding our method transfers from our camera motion free da Vinci surgery dataset to videos of laparoscopic interventions, outperforming classical homography estimation approaches in both, precision by 41%, and runtime on a CPU by 43%.
1 code implementation • 9 Aug 2021 • Lucas Fidon, Michael Aertsen, Nada Mufti, Thomas Deprest, Doaa Emam, Frédéric Guffens, Ernst Schwartz, Michael Ebner, Daniela Prayer, Gregor Kasprian, Anna L. David, Andrew Melbourne, Sébastien Ourselin, Jan Deprest, Georg Langs, Tom Vercauteren
The performance of deep neural networks typically increases with the number of training images.
2 code implementations • 8 Jul 2021 • Lucas Fidon, Michael Aertsen, Doaa Emam, Nada Mufti, Frédéric Guffens, Thomas Deprest, Philippe Demaerel, Anna L. David, Andrew Melbourne, Sébastien Ourselin, Jan Deprest, Tom Vercauteren
Deep neural networks have increased the accuracy of automatic segmentation, however, their accuracy depends on the availability of a large number of fully segmented images.
We demonstrate that an STCNN trained on a HAR dataset can be used in combination with an RNN to accurately represent arbitrarily long videos of seizures.
1 code implementation • 24 May 2021 • Fernando Pérez-García, Reuben Dorent, Michele Rizzi, Francesco Cardinale, Valerio Frazzini, Vincent Navarro, Caroline Essert, Irène Ollivier, Tom Vercauteren, Rachel Sparks, John S. Duncan, Sébastien Ourselin
We fine-tuned our model on three small annotated datasets from different institutions and on the annotated images in EPISURG, comprising 20, 33, 19 and 133 subjects.
While convolutional neural networks (CNNs) trained by back-propagation have seen unprecedented success at semantic segmentation tasks, they are known to struggle on out-of-distribution data.
Learning temporal patterns from multivariate longitudinal data is challenging especially in cases when data is sporadic, as often seen in, e. g., healthcare applications where the data can suffer from irregularity and asynchronicity as the time between consecutive data points can vary across features and samples, hindering the application of existing deep learning models that are constructed for complete, evenly spaced data with fixed sequence lengths.
One of the main sources of error in multi-atlas segmentation propagation approaches comes from the use of atlas databases that are morphologically dissimilar to the target image.
In this work, we aim at improving the fetal brain segmentation for SRR in Spina Bifida.
Also, we propose a scale attention module implicitly emphasizing the most salient feature maps among multiple scales so that the CNN is adaptive to the size of an object.
no code implementations • 7 Sep 2020 • Luis C. García-Peraza-Herrera, Wenqi Li, Caspar Gruijthuijsen, Alain Devreker, George Attilakos, Jan Deprest, Emmanuel Vander Poorten, Danail Stoyanov, Tom Vercauteren, Sébastien Ourselin
The latter, a combination of deep learning with optical flow tracking, yields an average balanced accuracy of 78. 2% across all the validated datasets.
In addition to EPISURG, we used three public datasets comprising 1813 preoperative MR images for training.
Metal artefact reduction (MAR) techniques aim at removing metal-induced noise from clinical images.
Accurate segmentation of the right ventricle (RV) is a crucial step in the assessment of the ventricular structure and function.
Additionally, we provide a graphical interface within a TorchIO extension in 3D Slicer to visualize the effects of transforms.
1 code implementation • 8 Jan 2020 • Lucas Fidon, Michael Aertsen, Thomas Deprest, Doaa Emam, Frédéric Guffens, Nada Mufti, Esther Van Elslander, Ernst Schwartz, Michael Ebner, Daniela Prayer, Gregor Kasprian, Anna L. David, Andrew Melbourne, Sébastien Ourselin, Jan Deprest, Georg Langs, Tom Vercauteren
In order to improve the robustness of machine learning systems, Distributionally Robust Optimization (DRO) has been proposed as a generalization of Empirical Risk Minimization (ERM).
Due to medical data privacy regulations, it is often infeasible to collect and share patient data in a centralised data lake.
no code implementations • 4 Sep 2019 • Carole H. Sudre, Beatriz Gomez Anson, Silvia Ingala, Chris D. Lane, Daniel Jimenez, Lukas Haider, Thomas Varsavsky, Ryutaro Tanno, Lorna Smith, Sébastien Ourselin, Rolf H. Jäger, M. Jorge Cardoso
Classification and differentiation of small pathological objects may greatly vary among human raters due to differences in training, expertise and their consistency over time.
no code implementations • 16 Aug 2019 • Mauricio Orbes-Arteaga, Thomas Varsavsky, Carole H. Sudre, Zach Eaton-Rosen, Lewis J. Haddow, Lauge Sørensen, Mads Nielsen, Akshay Pai, Sébastien Ourselin, Marc Modat, Parashkev Nachev, M. Jorge Cardoso
Inspired by recent work in semi-supervised learning we introduce a novel method to adapt from one source domain to $n$ target domains (as long as there is paired data covering all domains).
Different M-estimators and logistic functions, including a novel type proposed in this study, called modified Stannard, are evaluated on the data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) for robust modeling of volumetric MRI and PET biomarkers, CSF measurements, as well as cognitive tests.
We propose a new deep learning method for tumour segmentation when dealing with missing imaging modalities.
1 code implementation • 1 Apr 2019 • Hugo J. Kuijf, J. Matthijs Biesbroek, Jeroen de Bresser, Rutger Heinen, Simon Andermatt, Mariana Bento, Matt Berseth, Mikhail Belyaev, M. Jorge Cardoso, Adrià Casamitjana, D. Louis Collins, Mahsa Dadar, Achilleas Georgiou, Mohsen Ghafoorian, Dakai Jin, April Khademi, Jesse Knight, Hongwei Li, Xavier Lladó, Miguel Luna, Qaiser Mahmood, Richard McKinley, Alireza Mehrtash, Sébastien Ourselin, Bo-yong Park, Hyunjin Park, Sang Hyun Park, Simon Pezold, Elodie Puybareau, Leticia Rittner, Carole H. Sudre, Sergi Valverde, Verónica Vilaplana, Roland Wiest, Yongchao Xu, Ziyue Xu, Guodong Zeng, Jian-Guo Zhang, Guoyan Zheng, Christopher Chen, Wiesje van der Flier, Frederik Barkhof, Max A. Viergever, Geert Jan Biessels
Segmentation methods had to be containerized and submitted to the challenge organizers.
Segmenting vascular pathologies such as white matter lesions in Brain magnetic resonance images (MRIs) require acquisition of multiple sequences such as T1-weighted (T1-w) --on which lesions appear hypointense-- and fluid attenuated inversion recovery (FLAIR) sequence --where lesions appear hyperintense--.
no code implementations • 9 Jul 2018 • Yipeng Hu, Marc Modat, Eli Gibson, Wenqi Li, Nooshin Ghavami, Ester Bonmati, Guotai Wang, Steven Bandula, Caroline M. Moore, Mark Emberton, Sébastien Ourselin, J. Alison Noble, Dean C. Barratt, Tom Vercauteren
A median target registration error of 3. 6 mm on landmark centroids and a median Dice of 0. 87 on prostate glands are achieved from cross-validation experiments, in which 108 pairs of multimodal images from 76 patients were tested with high-quality anatomical labels.
Training deep neural networks on large and sparse datasets is still challenging and can require large amounts of computation and memory.
The analysis of vessel morphology and connectivity has an impact on a number of cardiovascular and neurovascular applications by providing patient-specific high-level quantitative features such as spatial location, direction and scale.
Robustness to visual challenges during registration and long-range temporal consistency are proposed, offering first positive results on in vivo data for which standard mosaicking techniques are not applicable.
10 code implementations • 11 Sep 2017 • Eli Gibson, Wenqi Li, Carole Sudre, Lucas Fidon, Dzhoshkun I. Shakir, Guotai Wang, Zach Eaton-Rosen, Robert Gray, Tom Doel, Yipeng Hu, Tom Whyntie, Parashkev Nachev, Marc Modat, Dean C. Barratt, Sébastien Ourselin, M. Jorge Cardoso, Tom Vercauteren
NiftyNet provides a modular deep-learning pipeline for a range of medical imaging applications including segmentation, regression, image generation and representation learning applications.
Deep-learning has proved in recent years to be a powerful tool for image analysis and is now widely used to segment both 2D and 3D medical images.
The purpose of this report is to present the distribution of the number of unique original items in a bootstrap sample clearly and concisely, with a view to enabling other machine learning researchers to understand and control this quantity in existing and future resampling techniques.
1 code implementation • 27 Jan 2014 • Da Ma, Manuel J. Cardoso, Marc Modat, Nick Powell, Jack Wells, Holly Holmes, Frances Wiseman, Victor Tybulewicz, Elizabeth Fisher, Mark F. Lythgoe, Sébastien Ourselin
The segmentation accuracy of the multi-atlas framework was evaluated using publicly available mouse brain atlas databases with pre-segmented manually labelled anatomical structures as the gold standard, and optimised parameters were obtained for the STEPS algorithm in the label fusion to achieve the best segmentation accuracy.