A Survey of Methods for 3D Histology Reconstruction
Histology is concerned with the various methods of microscopic examination of a thin tissue section (Culling, 2013), most commonly sampled from a specimen post mortem or from a biopsy. Cutting through a specimen reveals its internal topography and staining the sections permits the observation of complex differentiated structures. Then, the digitisation of histological sections (referred to as digital pathology) makes high-resolution microscope sections available for image computing and machine learning algorithms. (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 21, 2018 Category: Radiology Authors: Jonas Pichat, Juan Eugenio Iglesias, Tarek Yousry, S ébastien Ourselin, Marc Modat Source Type: research
Multiscale deep neural network based analysis of FDG-PET images for the early diagnosis of Alzheimer ’s disease
Alzheimer ’s disease (AD) accounts for about 50–75% of all types of dementia, affecting 1 out of 9 people aged above 65 years (Kawas, 2003; Alzheimer’s and Association, 2011). It is characterized by progressive cognitive decline such as deficits in memory, attention and executive function. Although curr ently there are no treatments for curing AD, there are several promising pharmacologic compounds in advanced stages of development, and it is expected that a breakthrough in treatment will be achieved very soon. (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 20, 2018 Category: Radiology Authors: Donghuan Lu, Karteek Popuri, Gavin Weiguang Ding, Rakesh Balachandar, Mirza Faisal Beg, for the Alzheimer ’s Disease Neuroimaging Initiative Source Type: research
The decomposition of deformation: New metrics to enhance shape analysis in medical imaging.
Medical image analysis experienced, in the last two decades, a significant boost in both shape recognition from medical images and analysis of related geometric data. In particular, 2D and 3D Speckle Tracking Echocardiography (2D and 3D STE), CT-scan and Magnetic Resonance (MR) represented the most used technologies, albeit not unique, in order to extract anatomical shapes from medical images to be analyzed and evaluated in terms of form, function and, ultimately, physiology. Shapes are identified by points (=landmarks) in 2D/3D space. (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 20, 2018 Category: Radiology Authors: Valerio Varano, Paolo Piras, Stefano Gabriele, Luciano Teresi, Paola Nardinocchi, Ian L. Dryden, Concetta Torromeo, Paolo E. Puddu Source Type: research
Fully-automated alignment of 3D fetal brain ultrasound to a canonical reference space using multi-task learning
Fetal neurosonography has improved significantly in the last few decades. It is emerging as a clinically useful imaging technology for assessing brain development and detecting cerebral abnormalities in the womb, which has applications in settings where expensive magnetic resonance imaging (MRI) is unavailable or not well-suited. Regardless of imaging modality, fetal brain localization and geometric alignment are the primordial steps for neuroimage analysis. This analysis relies on (i) initial localization of the brain, (ii) removal of extracranial and maternal tissues, and (iii) alignment of the region of interest to a re...
Source: Medical Image Analysis - February 20, 2018 Category: Radiology Authors: Ana I.L. Namburete, Weidi Xie, Mohammad Yaqub, Andrew Zisserman, J. Alison Noble Source Type: research
Multiscale deep neural network based analysis of FDG-PET images for the early diagnosis of Alzheimer ’s disease
Alzheimer ’s disease (AD) accounts for about 50–75% of all types of dementia, affecting 1 out of 9 people aged above 65 years (Kawas, 2003; Alzheimer’s and Association, 2011). It is characterized by progressive cognitive decline such as deficits in memory, attention and executive function. Although curr ently there are no treatments for curing AD, there are several promising pharmacologic compounds in advanced stages of development, and it is expected that a breakthrough in treatment will be achieved very soon. (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 20, 2018 Category: Radiology Authors: Donghuan Lu, Karteek Popuri, Gavin Weiguang Ding, Rakesh Balachandar, Mirza Faisal Beg, for the Alzheimer ’s Disease Neuroimaging Initiative Source Type: research
Multiscale Deep Neural Networks based analysis of FDG-PET images for the Early Diagnosis of Alzheimer ’s Disease
Alzheimer ’s disease (AD) accounts for about 50-75% of all types of dementia, affecting 1 out of 9 people aged above 65 years (Kawas, 2003; Alzheimer’s and Association, 2011). It is characterized by progressive cognitive decline such as deficits in memory, attention and executive function. Although curren tly there are no treatments for curing AD, there are several promising pharmacologic compounds in advanced stages of development, and it is expected that a breakthrough in treatment will be achieved very soon. (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 20, 2018 Category: Radiology Authors: Donghuan Lu, Karteek Popuri, Gavin Weiguang Ding, Rakesh Balachandar, Mirza Faisal Beg, Alzheimer ’s Disease Neuroimaging Initiative Source Type: research
The Decomposition of Deformation: new metrics to enhance shape analysis in medical imaging.
Medical image analysis experienced, in the last two decades, a significant boost in both shape recognition from medical images and analysis of related geometric data. In particular, 2D and 3D Speckle Tracking Echocardiography (2D and 3D STE), CT-scan and Magnetic Resonance (MR) represented the most used technologies, albeit not unique, in order to extract anatomical shapes from medical images to be analyzed and evaluated in terms of form, function and, ultimately, physiology. Shapes are identified by points (=landmarks) in 2D/3D space. (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 20, 2018 Category: Radiology Authors: Valerio Varano, Paolo Piras, Stefano Gabriele, Luciano Teresi, Paola Nardinocchi, Ian L. Dryden, Concetta Torromeo, Paolo E. Puddu Source Type: research
Fully-Automated Alignment of 3D Fetal Brain Ultrasound to a Canonical Reference Space using Multi-Task Learning
Fetal neurosonography has improved significantly in the last few decades. It is emerging as a clinically useful imaging technology for assessing brain development and detecting cerebral abnormalities in the womb, which has applications in settings where expensive magnetic resonance imaging (MRI) is unavailable or not well-suited. Regardless of imaging modality, fetal brain localization and geometric alignment are the primordial steps for neuroimage analysis. This analysis relies on (i) initial localization of the brain, (ii) removal of extracranial and maternal tissues, and (iii) alignment of the region of interest to a re...
Source: Medical Image Analysis - February 20, 2018 Category: Radiology Authors: Ana I.L. Namburete, Weidi Xie, Mohammad Yaqub, Andrew Zisserman, J. Alison Noble Source Type: research
Intrasubject multimodal groupwise registration with the conditional template entropy
Biomedical image registration is the process of spatially aligning medical images, allowing for an accurate and quantitative comparison. An increasing number of image analysis tasks calls for the alignment of multiple (more than two) images. Examples include the joint analysis of tissue properties using multi-parametric MRI (Huizinga et al., 2016; Wells et al., 2015), spatio-temporal motion estimation from dynamic sequences (Metz et al., 2011; Vandemeulebroucke et al., 2011), atlas construction (Fletcher et al., 2009; Joshi et al., 2004; Wu et al., 2011) and population analyses (Geng et al., 2009). (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 17, 2018 Category: Radiology Authors: Mathias Polfliet, Stefan Klein, Wyke Huizinga, Margarethus M. Paulides, Wiro J. Niessen, Jef Vandemeulebroucke Source Type: research
Intrasubject Multimodal Groupwise Registration with the Conditional Template Entropy
Biomedical image registration is the process of spatially aligning medical images, allowing for an accurate and quantitative comparison. An increasing number of image analysis tasks calls for the alignment of multiple (more than two) images. Examples include the joint analysis of tissue properties using multi-parametric MRI (Huizinga et al., 2016; Wells et al., 2015), spatio-temporal motion estimation from dynamic sequences (Metz et al., 2011; Vandemeulebroucke et al., 2011), atlas construction (Fletcher et al., 2009; Joshi et al., 2004; Wu et al., 2011) and population analyses (Geng et al., 2009). (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 17, 2018 Category: Radiology Authors: Mathias Polfliet, Stefan Klein, Wyke Huizinga, Margarethus M. Paulides, Wiro J. Niessen, Jef Vandemeulebroucke Source Type: research
Estimating fiber orientation distribution from diffusion MRI with spherical needlets
Diffusion-weighted MRI (D-MRI) (Le Bihan et al., 2001; Mori, 2007) has become a widely used, non-invasive tool for clinical and experimental neuroscience due to its capability of characterizing tissue microstructure in vivo by measuring the diffusion displacement of water molecules. Specifically, high angular resolution diffusion imaging (HARDI) enables extraction of accurate and detailed information about fiber tract directions through measurements made along a large number of gradient directions (Tuch et al., 2002). (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 7, 2018 Category: Radiology Authors: Hao Yan, Owen Carmichael, Debashis Paul, Jie Peng, for the Alzheimer ’s Disease Neuroimaging Initiative Source Type: research
Estimating fiber orientation distribution from diffusion MRI with spherical needlets
Diffusion-weighted MRI (D-MRI) (Le Bihan et al., 2001; Mori, 2007) has become a widely used, non-invasive tool for clinical and experimental neuroscience due to its capability of characterizing tissue microstructure in vivo by measuring the diffusion displacement of water molecules. Specifically, high angular resolution diffusion imaging (HARDI) enables extraction of accurate and detailed information about fiber tract directions through measurements made along a large number of gradient directions (Tuch et al., 2002). (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 7, 2018 Category: Radiology Authors: Hao Yan, Owen Carmichael, Debashis Paul, Jie Peng, Alzheimer ’s Disease Neuroimaging Initiative Source Type: research
DeepMitosis: Mitosis detection via deep detection, verification and segmentation networks
According to the Nottingham Grading System (Elston and Ellis, 1991), there are three important morphological features on Hematoxylin and Eosin (H&E) stained slides for breast cancer grading. They are mitotic count, tubule formation, and nuclear pleomorphism. Among them, mitotic count is the most important biomarker. Pathologists usually search for mitosis on high-power fields (HPFs) manually. It is a time-consuming and tedious task due to the large number of HPFs in a single whole slide and the high variation in the appearance of mitotic cells. (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 4, 2018 Category: Radiology Authors: Chao Li, Xinggang Wang, Wenyu Liu, Longin Jan Latecki Source Type: research
Universal ventricular coordinates: A generic framework for describing position within the heart and transferring data
The ventricles of mammalian hearts share many common characteristics. These include a biventricular geometry, Purkinje system (PS), and helical myocardial fiber orientation (Streeter et al., 1969). They also express electrical heterogeneity with respect to transmural (Lou et al., 2011; Sabir et al., 2007), apicobasal (Janse et al., 2012), and left-right gradients (Pandit et al., 2011; Volders et al., 1999). (Source: Medical Image Analysis)
Source: Medical Image Analysis - February 4, 2018 Category: Radiology Authors: Jason Bayer, Anton J. Prassl, Ali Pashaei, Juan F. Gomez, Antonio Frontera, Aurel Neic, Gernot Plank, Edward J. Vigmond Source Type: research
DeepMitosis: Mitosis Detection via Deep Detection, Verication and Segmentation Networks
Mitotic count is a critical predictor of tumor aggressiveness in the breast cancer diagnosis. Nowadays mitosis counting is mainly performed by patholo-gists manually, which is extremely arduous and time-consuming. In this paper, we propose an accurate method for detecting the mitotic cells from histopatho-logical slides using a novel multi-stage deep learning framework. Our method consists of a deep segmentation network for generating mitosis region when on-ly a weak label is given (i.e., only the centroid pixel of mitosis is annotated), an elaborately designed deep detection network for localizing mitosis by using context...
Source: Medical Image Analysis - February 4, 2018 Category: Radiology Authors: Chao Li, Xinggang Wang, Wenyu Liu, Longin Jan Latecki Source Type: research
