A5040063476- AI in cancer detection
- Medical Imaging and Analysis
- Surgical Simulation and Training
- Cell Image Analysis Techniques
- Digital Imaging for Blood Diseases
- Augmented Reality Applications
- Dental Radiography and Imaging
- Radiomics and Machine Learning in Medical Imaging
- Medical Imaging Techniques and Applications
- Hepatocellular Carcinoma Treatment and Prognosis
- Colorectal Cancer Surgical Treatments
- Robot Manipulation and Learning
- Advanced Radiotherapy Techniques
- Dental Research and COVID-19
- Advanced MRI Techniques and Applications
- Modular Robots and Swarm Intelligence
- Anatomy and Medical Technology
- Robotics and Sensor-Based Localization
- Oral microbiology and periodontitis research
- Advanced X-ray and CT Imaging
- Micro and Nano Robotics
SINTEF
2018-2024
SINTEF Digital
2020
Norwegian University of Science and Technology
2020
Centre for Automation and Robotics
2017
Universidad Politécnica de Madrid
2017
This article presents a review on trends in modular reconfigurable robots, comparing the evolution of features most significant robots over years and focusing latest designs. These are reconfiguration, docking, degrees freedom, locomotion, control, communications, size, powering. For each feature, some relevant designs presented current design discussed.
Abstract Background Dental caries diagnosis requires the manual inspection of diagnostic bitewing images patient, followed by a visual and probing identified dental pieces with potential lesions. Yet use artificial intelligence, in particular deep-learning, has to aid providing quick informative analysis images. Methods A dataset 13,887 bitewings from HUNT4 Oral Health Study were annotated individually six different experts, used train three object detection deep-learning architectures:...
Deep convolutional neural networks (CNNs) are the current state-of-the-art for digital analysis of histopathological images. The large size whole-slide microscopy images (WSIs) requires advanced memory handling to read, display and process these There several open-source platforms working with WSIs, but few support deployment CNN models. These applications use third-party solutions inference, making them less user-friendly unsuitable high-performance image analysis. To make CNNs feasible on...
This study aims to evaluate the accuracy of point-based registration (PBR) when used for augmented reality (AR) in laparoscopic liver resection surgery.The was conducted three different scenarios which sampling targets PBR decreases: using an assessment phantom with machined divot holes, a patient-specific markers visible computed tomography (CT) scans and vivo, relying on surgeon's anatomical understanding perform annotations. Target error (TRE) fiducial (FRE) were five randomly selected...
Image guided surgery systems aim to support surgeons by providing reliable pre-operative and intra-operative imaging of the patient combined with corresponding tracked instrument location. The image guidance is based on a combination medical images, such as Magnetic Resonance Imaging (MRI), Computed Tomography (CT) Ultrasonography (US), surgical tracking. For this reason, tracking are great importance they determine location orientation equipment used surgeons. Assessment accuracy these...
Test the feasibility of novel Single Landmark image-to-patient registration method for use in operating room future clinical trials. The algorithm is implemented open-source platform CustusX, a computer-aided intervention research dedicated to intraoperative navigation and ultrasound, with an interface laparoscopic ultrasound probes.The compared fiducial landmark on IOUSFAN (Kyoto Kagaku Co., Ltd., Japan) soft tissue abdominal phantom T2 magnetic resonance scans it.The experiments show that...
Purpose: This study aims to explore training strategies improve convolutional neural network-based image-to-image deformable registration for abdominal imaging. Methods: Different strategies, loss functions, and transfer learning schemes were considered. Furthermore, an augmentation layer which generates artificial image pairs on-the-fly was proposed, in addition a that enables dynamic weighting. Results: Guiding using segmentations the step proved beneficial deep-learning-based...
Background: Dental caries diagnosis requires the manual inspection of diagnostic bitewing images patient, followed by a visual and probing identified dental pieces with potential lesions. Yet use artificial intelligence, in particular deep-learning, has to aid providing quick informative analysis images. Methods: A dataset 13,887 bitewings from HUNT4 Oral Health Study were annotated individually six different experts, used train three object detection deep-learning architectures: RetinaNet...
Deep convolutional neural networks (CNNs) are the current state-of-the-art for digital analysis of histopathological images. The large size whole-slide microscopy images (WSIs) requires advanced memory handling to read, display and process these There several open-source platforms working with WSIs, but few support deployment CNN models. These applications use third-party solutions inference, making them less user-friendly unsuitable high-performance image analysis. To make CNNs feasible on...
Purpose: This study aims to explore training strategies improve convolutional neural network-based image-to-image deformable registration for abdominal imaging. Methods: Different strategies, loss functions, and transfer learning schemes were considered. Furthermore, an augmentation layer which generates artificial image pairs on-the-fly was proposed, in addition a that enables dynamic weighting. Results: Guiding using segmentations the step proved beneficial deep-learning-based...