Reuben Dorent
A5061736113- Medical Image Segmentation Techniques
- Radiomics and Machine Learning in Medical Imaging
- Medical Imaging and Analysis
- Meningioma and schwannoma management
- Domain Adaptation and Few-Shot Learning
- Advanced Neural Network Applications
- Brain Tumor Detection and Classification
- Vascular Malformations Diagnosis and Treatment
- Vestibular and auditory disorders
- Glioma Diagnosis and Treatment
- Anatomy and Medical Technology
- Medical Imaging Techniques and Applications
- Cerebral Venous Sinus Thrombosis
- Generative Adversarial Networks and Image Synthesis
- Traumatic Brain Injury and Neurovascular Disturbances
- Surgical Simulation and Training
- Facial Nerve Paralysis Treatment and Research
- 3D Shape Modeling and Analysis
- COVID-19 diagnosis using AI
- Multimodal Machine Learning Applications
- Image and Object Detection Techniques
- AI in cancer detection
- Pleural and Pulmonary Diseases
- Ultrasound in Clinical Applications
- Intracranial Aneurysms: Treatment and Complications
Harvard University
2023-2025
Brigham and Women's Hospital
2023-2025
King's College London
2019-2024
University of Lübeck
2024
Technical University of Munich
2023
St Thomas' Hospital
2020
Automatic segmentation of vestibular schwannomas (VSs) from MRI could significantly improve clinical workflow and assist in patient management. Accurate tumor volumetric measurements provide the best indicators to detect subtle VS growth, but current techniques are labor intensive dedicated software is not readily available within setting. The authors aim develop a novel artificial intelligence (AI) framework be embedded routine for automatic delineation volumetry VS.
Automatic segmentation of vestibular schwannomas (VS) from magnetic resonance imaging (MRI) could significantly improve clinical workflow and assist patient management. We have previously developed a novel artificial intelligence framework based on 2.5D convolutional neural network achieving excellent results equivalent to those achieved by an independent human annotator. Here, we provide the first publicly-available annotated dataset VS releasing data annotations used in our prior work....
Domain Adaptation (DA) has recently raised strong interests in the medical imaging community. While a large variety of DA techniques been proposed for image segmentation, most these have validated either on private datasets or small publicly available datasets. Moreover, mostly addressed single-class problems. To tackle limitations, Cross-Modality (crossMoDA) challenge was organised conjunction with 24th International Conference Medical Image Computing and Computer Assisted Intervention...
Abstract The standard of care for brain tumors is maximal safe surgical resection. Neuronavigation augments the surgeon’s ability to achieve this but loses validity as surgery progresses due shift. Moreover, gliomas are often indistinguishable from surrounding healthy tissue. Intraoperative magnetic resonance imaging (iMRI) and ultrasound (iUS) help visualize tumor iUS faster easier incorporate into workflows offers a lower contrast between tumorous tissues than iMRI. With success...
Accurate assessment of lymph node size in 3D CT scans is crucial for cancer staging, therapeutic management, and monitoring treatment response. Existing state-of-the-art segmentation frameworks medical imaging often rely on fully annotated datasets. However, segmentation, these datasets are typically small due to the extensive time expertise required annotate numerous nodes scans. Weakly-supervised learning, which leverages incomplete or noisy annotations, has recently gained interest...
Brain tissue segmentation from multimodal MRI is a key building block of many neuroimaging analysis pipelines. Established approaches have, however, not been developed to cope with large anatomical changes resulting pathology, such as white matter lesions or tumours, and often fail in these cases. In the meantime, advent deep neural networks (DNNs), brain has matured significantly. However, few existing allow for joint normal lesions. Developing DNN task currently hampered by fact that...
Automatic segmentation of vestibular schwannoma (VS) from routine clinical MRI has potential to improve workflow, facilitate treatment decisions, and assist patient management. Previous work demonstrated reliable automatic performance on datasets standardized images acquired for stereotactic surgery planning. However, diagnostic are generally more diverse pose a larger challenge algorithms, especially when post-operative included. In this work, we show the first time that VS is also possible...
Abstract Purpose Management of vestibular schwannoma (VS) is based on tumour size as observed T1 MRI scans with contrast agent injection. The current clinical practice to measure the diameter in its largest dimension. It has been shown that volumetric measurement more accurate and reliable a VS size. reference approach achieve such volumetry manually segment tumour, which time intensive task. We suggest semi-automated segmentation may be clinically applicable solution this problem it could...
The number of international benchmarking competitions is steadily increasing in various fields machine learning (ML) research and practice. So far, however, little known about the common practice as well bottlenecks faced by community tackling questions posed. To shed light on status quo algorithm development specific field biomedical imaging analysis, we designed an survey that was issued to all participants challenges conducted conjunction with IEEE ISBI 2021 MICCAI conferences (80 total)....
ABSTRACT Automatic segmentation of vestibular schwannomas (VS) from magnetic resonance imaging (MRI) could significantly improve clinical workflow and assist patient management. We have previously developed a novel artificial intelligence framework based on 2.5D convolutional neural network achieving excellent results equivalent to those achieved by an independent human annotator. Here, we provide the first publicly-available annotated dataset VS releasing data annotations used in our prior...
The Koos grading scale is a frequently used classification system for vestibular schwannoma (VS) that accounts extrameatal tumor dimension and compression of the brain stem. We propose an artificial intelligence (AI) pipeline to fully automate segmentation VS from MRI improve clinical workflow facilitate patient management.We method does not only rely on available images but also automatically generated segmentations. Artificial neural networks were trained tested based manual segmentations...
Geodesic and Euclidean distance transforms have been widely used in a number of applications where from set reference points is computed.Methods recent years shown effectiveness applying the transform to interactively annotate 3D medical imaging data (Criminisi et al., 2008;Wang 2018).The enables providing segmentation labels, i.e., voxel-wise for different objects interests.Despite existing methods efficient computation on GPU CPU devices 2008(Criminisi , 2009;;Toivanen, 1996;Weber 2008),...
Abstract Objectives The shape is commonly used to describe the objects. State-of-the-art algorithms in medical imaging are predominantly diverging from computer vision, where voxel grids, meshes, point clouds, and implicit surface models used. This seen growing popularity of ShapeNet (51,300 models) Princeton ModelNet (127,915 models). However, a large collection anatomical shapes (e.g., bones, organs, vessels) 3D surgical instruments missing. Methods We present MedShapeNet translate...
Prior to the deep learning era, shape was commonly used describe objects. Nowadays, state-of-the-art (SOTA) algorithms in medical imaging are predominantly diverging from computer vision, where voxel grids, meshes, point clouds, and implicit surface models used. This is seen numerous shape-related publications premier vision conferences as well growing popularity of ShapeNet (about 51,300 models) Princeton ModelNet (127,915 models). For domain, we present a large collection anatomical shapes...
Surgical instrument segmentation is recognised as a key enabler to provide advanced surgical assistance and improve computer assisted interventions. In this work, we propose SegMatch, semi supervised learning method reduce the need for expensive annotation laparoscopic robotic images. SegMatch builds on FixMatch, widespread classification pipeline combining consistency regularization pseudo labelling, adapts it purpose of segmentation. our proposed unlabelled images are weakly augmented fed...