Ultrasound guidance in minimally invasive robotic procedures
Minimally invasive surgery (MIS) is usually associated with a long learning curve due to technical limitations faced by surgeons in dealing with limited field of view and complete lack of depth perception (Vitiello et al., 2013) (Elgezua et al., 2013). Introduction of robotic technology to MIS has been proposed to bring accuracy and precision and the early clinical adoption of robots in surgery dates back to 1980s. By 1994 voice controlled camera holding robots were demonstrated to work in symbiosis with surgeons performing complex laparoscopic surgery (Sackier& Wang, 1994). (Source: Medical Image Analysis)
Source: Medical Image Analysis - January 10, 2019 Category: Radiology Authors: Maria Antico, Fumio Sasazawa, Liao Wu, Anjali Jaiprakash, Jonathan Roberts, Ross Crawford, Ajay K. Pandey, Davide Fontanarosa Source Type: research
Learning to detect chest radiographs containing pulmonary lesions using visual attention networks
Lung cancer is the most common cancer worldwide and the second most common cancer in Europe and the USA (Ferlay et al., 2013; American Cancer Society, 1999). Due to delays in diagnosis, it is typically discovered at an advanced stage with a very low survival rate (Cancer Research UK, 2014). The chest radiograph is the most commonly performed radiological investigation in the initial assessment of suspected lun g cancer because it is inexpensive and delivers a low radiation dose. On a chest radiograph, a nodule is defined as a rounded opacity ≤ 3cm, which can be well- or poorly marginated. (Source: Medical Image Analysis)
Source: Medical Image Analysis - January 9, 2019 Category: Radiology Authors: Emanuele Pesce, Samuel Withey, Petros-Pavlos Ypsilantis, Robert Bakewell, Vicky Goh, Giovanni Montana Source Type: research
Repetitive Motion Compensation for Real Time Intraoperative Video Processing
Optical imaging is a modality of choice for surgery assistance: it is inexpensive, it has high temporal and spatial resolution and, as opposed to magnetic resonance or computational tomography, it implies only limited constraints on the surgical material and room. This is why even simple devices such as true color camera are often used for assistance or monitoring during surgical intervention or radiotherapy for a variety of body parts: brain (Pichette et al., 2016), heart (Richa et al., 2011) or abdomen (Spinczyk et al., 2014) for example. (Source: Medical Image Analysis)
Source: Medical Image Analysis - January 3, 2019 Category: Radiology Authors: Micha ël Sdika, Laure Alston, David Rousseau, Jacques Guyotat, Laurent Mahieu-Williame, Bruno Montcel Source Type: research
MILD-Net: Minimal Information Loss Dilated Network for Gland Instance Segmentation in Colon Histology Images
Colorectal cancer is the third most commonly occurring cancer in men and the second most commonly occurring cancer in women, where approximately 95% of all colorectal cancers are adenocarcinomas (Fleming et al., 2012). Colorectal adenocarcinoma develops in the lining of the colon or rectum, which makes up the large intestine and is characterised by glandular formation. Histological examination of the glands, most frequently with the Hematoxylin& Eosin (H&E) stain, is routine practice for assessing the differentiation of the cancer within colorectal adenocarcinoma. (Source: Medical Image Analysis)
Source: Medical Image Analysis - December 20, 2018 Category: Radiology Authors: Simon Graham, Hao Chen, Jevgenij Gamper, Qi Dou, Pheng-Ann Heng, David Snead, Yee Wah Tsang, Nasir Rajpoot Source Type: research
Automated diagnosis of breast ultrasonography images using deep neural networks
Breast cancer is the most common cancer among women worldwide. There has been a general uptrend in the morbidity of breast cancer since the 1990s. (Fitzmaurice et al., 2015) According to the World Health Organization, breast cancer is responsible for over 500,000 deaths each year and 1.7 million new cases are diagnosed every year. Breast cancer is the cancer with the highest incidence for women in 161 countries and the most common cause for cancer deaths in women in 98 countries. Breast cancer can be cured if diagnosed and treated early. (Source: Medical Image Analysis)
Source: Medical Image Analysis - December 20, 2018 Category: Radiology Authors: Xiaofeng Qi, Lei Zhang, Yao Chen, Yong Pi, Yi Chen, Qing Lv, Zhang Yi Source Type: research
Towards Cross-modal Organ Translation and Segmentation: A Cycle- and Shape-Consistent Generative Adversarial Network
In current clinical practice, multiple imaging modalities may be available for disease diagnosis and surgical planning (Cao et al., 2017; Chen et al., 2017). For a specific patient group, a certain imaging modality might be more popular than others. Due to the proliferation of multiple imaging modalities, there is a strong clinical need to develop a cross-modality image transfer analysis system to assist clinical treatme nt. For example, synthesized computed tomography (CT) data can provide X-ray attenuation map for radiation therapy planning (Burgos et al., 2015). (Source: Medical Image Analysis)
Source: Medical Image Analysis - December 19, 2018 Category: Radiology Authors: Jinzheng Cai, Zizhao Zhang, Lei Cui, Yefeng Zheng, Lin Yang Source Type: research
Robust motion correction for cardiac T1 and ECV mapping using a T1 relaxation model approach
Accurate measurement of myocardial T1 and extra cellular volume (ECV) using cardiac MRI is highly relevant for the diagnosis of diffuse myocardial diseases such as diffuse fibrosis, amyloidosis and Anderson Fabry disease (Schelbert and Messroghli, 2016; Haaf et al., 2017). Compared to the conventional late gadolinium enhanced (LGE) images where diagnosis is based on the subjective assessment of relative contrast differences, T1 and ECV mapping allow quantitative characterization of the myocardium. (Source: Medical Image Analysis)
Source: Medical Image Analysis - December 18, 2018 Category: Radiology Authors: Sofie Tilborghs, Tom Dresselaers, Piet Claus, Guido Claessen, Jan Bogaert, Frederik Maes, Paul Suetens Source Type: research
Micro-Net: A unified model for segmentation of various objects in microscopy images
In automated microscopic image analysis pipelines, segmentation of key structures such as tumours, glands and cells is an important step ((Awan et al., 2017; Yuan et al., 2012; Qaiser et al., 2017)). Recent advances in deep learning have helped to achieve accurate segmentation of these structures. A major strength of deep learning is that the same network architecture can be used to segment various structures across different modalities by retraining and slight tuning of the input parameters ((Shelhamer et al., 2017; Ronneberger et al., 2015)). (Source: Medical Image Analysis)
Source: Medical Image Analysis - December 15, 2018 Category: Radiology Authors: Shan E Ahmed Raza, Linda Cheung, Muhammad Shaban, Simon Graham, David Epstein, Stella Pelengaris, Michael Khan, Nasir M. Rajpoot Source Type: research
A Deep Learning Framework for Unsupervised Affine and Deformable Image Registration
Image registration is the process of aligning two or more images. It is a well-established technique in (semi-)automatic medical image analysis that is used to transfer information between images. Commonly used image registration approaches include intensity-based methods, and feature-based methods that use handcrafted image features (Sotiras et al., 2013; Viergever et al., 2016). Since recently, supervised and unsupervised deep learning techniques have been successfully employed for image registration (Jaderberg et al., 2015; Wu et al., 2016; Miao et al., 2016; Liao et al., 2017; Krebs et al., 2017; Cao et al., 2...
Source: Medical Image Analysis - December 8, 2018 Category: Radiology Authors: Bob D. de Vos, Floris F. Berendsen, Max A. Viergever, Hessam Sokooti, Marius Staring, Ivana I šgum Source Type: research
The Effect of Motion Correction Interpolation on Quantitative T1 Mapping with MRI
Quantitative magnetic resonance imaging (qMRI) aims to measure the biophysical properties of biological tissue, such as the longitudinal and transverse magnetization relaxation times, T1 and T2, respectively. Unlike weighted anatomical MR images used in the clinic, qMRI maps aim to remove instrumental biases, yielding interpretable physical units. This measurement of microstructural changes in a tissue non-invasively is considered critical for clinical diagnosis and brain research (Cercignani et al., 2018). (Source: Medical Image Analysis)
Source: Medical Image Analysis - December 1, 2018 Category: Radiology Authors: Amitay Nachmani, Roey Schurr, Leo Joskowicz, Aviv A. Mezer Source Type: research
Editorial Board
(Source: Medical Image Analysis)
Source: Medical Image Analysis - November 27, 2018 Category: Radiology Source Type: research
Reviewers- an acknowledgement
(Source: Medical Image Analysis)
Source: Medical Image Analysis - November 27, 2018 Category: Radiology Source Type: research
A Graph-Cut Approach for Pulmonary Artery-Vein Segmentation in Noncontrast CT Images
Computed tomography (CT) technology has been improving during the last decades and currently it is possible to obtain near-isotropic, sub-millimetre resolution acquisition of the complete chest in a single breath-hold, avoiding partial volume effects and breathing artefacts. For that reason CT has become the reference modality in pulmonary imaging and the radiological study of complex biological structures such as pulmonary vessel trees (Hsieh, 2009; Sluimer et al., 2006). (Source: Medical Image Analysis)
Source: Medical Image Analysis - November 26, 2018 Category: Radiology Authors: Daniel Jimenez-Carretero, David Bermejo-Pelaez, Pietro Nardelli, Patricia Fraga, Eduardo Fraile, Raul San Jose Estepar, Maria J Ledesma-Carbayo Source Type: research
GAS: a Genetic Atlas Selection Strategy in Multi-Atlas Segmentation Framework
Multi-Atlas based Segmentation (MAS) algorithms have been successfully applied to a wide range of medical image segmentation tasks (Isgum et al., 2009; Cardoso et al., 2013; Aljabar et al., 2009). Their success relies on the introduction of a priori knowledge using a set of pre-segmented images. An atlas consists of a medical image and a corresponding segmented label image. According to the subdivision proposed in Iglesias and Sabunc u (2014), a MAS algorithm can be implemented following four sequential steps: first each atlas image is registered to the target image (registration step), second either all or a subset of...
Source: Medical Image Analysis - November 19, 2018 Category: Radiology Authors: Michela Antonelli, M. Jorge Cardoso, Edward W. Johnston, Mrishta Brizmohun Appayya, Benoit Presles, Marc Modat, Shonit Punwani, Sebastien Ourselin Source Type: research
Automated Segmentation of Knee Bone and Cartilage combining Statistical Shape Knowledge and Convolutional Neural Networks: Data from the Osteoarthritis Initiative
Knee osteoarthritis (OA) is a chronic, degenerative joint disease affecting a significant fraction of the human population (Lawrence et al., 2008). Due to the rising average life expectancy, an increasing obesity, and a prolonged desire for an active lifestyle, research to understand and prevent OA will become even more important. Magnetic Resonance Imaging (MRI) is commonly used to assess knee joint degeneration, especially of th e femoral bone (FB), tibial bone (TB), and the respective femoral and tibial cartilage (FC,TC) (Conaghan et al., 2011). (Source: Medical Image Analysis)
Source: Medical Image Analysis - November 17, 2018 Category: Radiology Authors: Felix Ambellan, Alexander Tack, Moritz Ehlke, Stefan Zachow Source Type: research
