scholarly journals Comparison of Two-Dimensional- and Three-Dimensional-Based U-Net Architectures for Brain Tissue Classification in One-Dimensional Brain CT

2022 ◽  
Vol 15 ◽  
Author(s):  
Meera Srikrishna ◽  
Rolf A. Heckemann ◽  
Joana B. Pereira ◽  
Giovanni Volpe ◽  
Anna Zettergren ◽  
...  
Keyword(s):  
Deep Learning ◽  
Brain Tissue ◽  
Three Dimensional ◽  
Two Dimensional ◽  
3D Segmentation ◽  
Tissue Classification ◽  
Tissue Segmentation ◽  
One Dimensional ◽  
Ct Brain ◽  
2D And 3D

Brain tissue segmentation plays a crucial role in feature extraction, volumetric quantification, and morphometric analysis of brain scans. For the assessment of brain structure and integrity, CT is a non-invasive, cheaper, faster, and more widely available modality than MRI. However, the clinical application of CT is mostly limited to the visual assessment of brain integrity and exclusion of copathologies. We have previously developed two-dimensional (2D) deep learning-based segmentation networks that successfully classified brain tissue in head CT. Recently, deep learning-based MRI segmentation models successfully use patch-based three-dimensional (3D) segmentation networks. In this study, we aimed to develop patch-based 3D segmentation networks for CT brain tissue classification. Furthermore, we aimed to compare the performance of 2D- and 3D-based segmentation networks to perform brain tissue classification in anisotropic CT scans. For this purpose, we developed 2D and 3D U-Net-based deep learning models that were trained and validated on MR-derived segmentations from scans of 744 participants of the Gothenburg H70 Cohort with both CT and T1-weighted MRI scans acquired timely close to each other. Segmentation performance of both 2D and 3D models was evaluated on 234 unseen datasets using measures of distance, spatial similarity, and tissue volume. Single-task slice-wise processed 2D U-Nets performed better than multitask patch-based 3D U-Nets in CT brain tissue classification. These findings provide support to the use of 2D U-Nets to segment brain tissue in one-dimensional (1D) CT. This could increase the application of CT to detect brain abnormalities in clinical settings.

Combined Effects of 2D and 3D Inhomogeneities on the Dynamic Susceptibility of Superlattices

2010 ◽  
Vol 168-169 ◽  
pp. 97-100
Author(s):  
V.A. Ignatchenko ◽  
D.S. Tsikalov
Keyword(s):  
Density Of States ◽  
Three Dimensional ◽  
Wave Number ◽  
Dynamic Susceptibility ◽  
Combined Effects ◽  
Two Dimensional ◽  
Simultaneous Presence ◽  
One Dimensional ◽  
The One ◽  
2D And 3D

The dynamic susceptibility and the one-dimensional density of states (DOS) of an initially sinusoidal superlattice (SL) with simultaneous presence of two-dimensional (2D) phase inhomogeneities that simulate the deformations of the interfaces between the SL’s layers and three-dimensional (3D) amplitude inhomogeneities of the layer material of the SL were investigated. An analytical expression for the averaged Green’s function of the sinusoidal SL with 2D phase inhomogeneities was obtained in the Bourret approximation. It was shown that the effect of increasing asymmetry of heights of the dynamic susceptibility peaks at the edge of the Brillouin zone of the SL, which was found in [6] at increasing the rms fluctuations of 2D inhomogeneities, also takes place at increasing the correlation wave number of such inhomogeneities. It was also shown that the increase of the rms fluctuations of 3D amplitude inhomogeneities in the superlattice with 2D phase inhomogeneities leads to the suppression of the asymmetry effect and to the decrease of the depth of the DOS gap.

scholarly journals From Auto-encoders to Capsule Networks: A Survey

2021 ◽  
Vol 229 ◽  
pp. 01003
Author(s):  
Omaima El Alaoui-Elfels ◽  
Taoufiq Gadi
Keyword(s):  
Neural Networks ◽  
Image Processing ◽  
Deep Learning ◽  
Learning Algorithm ◽  
Three Dimensional ◽  
Future Research ◽  
Two Dimensional ◽  
One Dimensional ◽  
New Generation ◽  
Open Issues

Convolutional Neural Networks are a very powerful Deep Learning structure used in image processing, object classification and segmentation. They are very robust in extracting features from data and largely used in several domains. Nonetheless, they require a large number of training datasets and relations between features get lost in the Max-pooling step, which can lead to a wrong classification. Capsule Networks(CapsNets) were introduced to overcome these limitations by extracting features and their pose using capsules instead of neurons. This technique shows an impressive performance in one-dimensional, two-dimensional and three-dimensional datasets as well as in sparse datasets. In this paper, we present an initial understanding of CapsNets, their concept, structure and learning algorithm. We introduce the progress made by CapsNets from their introduction in 2011 until 2020. We compare different CapsNets series architectures to demonstrate strengths and challenges. Finally, we quote different implementations of Capsule Networks and show their robustness in a variety of domains. This survey provides the state-of-theartof Capsule Networks and allows other researchers to get a clear view of this new field. Besides, we discuss the open issues and the promising directions of future research, which may lead to a new generation of CapsNets.

scholarly journals Homogeneous 2D and 3D alignment of cardiomyocyte in dilated cardiomyopathy revealed by intravital heart imaging

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Kiyoshi Masuyama ◽  
Tomoaki Higo ◽  
Jong-Kook Lee ◽  
Ryohei Matsuura ◽  
Ian Jones ◽  
...  
Keyword(s):  
Heart Failure ◽  
Dilated Cardiomyopathy ◽  
Cardiac Dysfunction ◽  
Three Dimensional ◽  
Single Layer ◽  
Specific Pattern ◽  
Two Dimensional ◽  
Heart Imaging ◽  
Novel Method ◽  
2D And 3D

AbstractIn contrast to hypertrophic cardiomyopathy, there has been reported no specific pattern of cardiomyocyte array in dilated cardiomyopathy (DCM), partially because lack of alignment assessment in a three-dimensional (3D) manner. Here we have established a novel method to evaluate cardiomyocyte alignment in 3D using intravital heart imaging and demonstrated homogeneous alignment in DCM mice. Whilst cardiomyocytes of control mice changed their alignment by every layer in 3D and position twistedly even in a single layer, termed myocyte twist, cardiomyocytes of DCM mice aligned homogeneously both in two-dimensional (2D) and in 3D and lost myocyte twist. Manipulation of cultured cardiomyocyte toward homogeneously aligned increased their contractility, suggesting that homogeneous alignment in DCM mice is due to a sort of alignment remodelling as a way to compensate cardiac dysfunction. Our findings provide the first intravital evidence of cardiomyocyte alignment and will bring new insights into understanding the mechanism of heart failure.

scholarly journals Do 3D Face Images Capture Cues of Strength, Weight, and Height Better than 2D Face Images do?

2021 ◽  
Vol 7 (3) ◽  
pp. 209-219
Author(s):  
Iris J Holzleitner ◽  
Alex L Jones ◽  
Kieran J O’Shea ◽  
Rachel Cassar ◽  
Vanessa Fasolt ◽  
...  
Keyword(s):  
Three Dimensional ◽  
Physical Characteristics ◽  
Two Dimensional ◽  
3D Images ◽  
3D Face ◽  
Face Images ◽  
Similar Accuracy ◽  
2D And 3D ◽  
2D Images ◽  
Better Than

Abstract Objectives A large literature exists investigating the extent to which physical characteristics (e.g., strength, weight, and height) can be accurately assessed from face images. While most of these studies have employed two-dimensional (2D) face images as stimuli, some recent studies have used three-dimensional (3D) face images because they may contain cues not visible in 2D face images. As equipment required for 3D face images is considerably more expensive than that required for 2D face images, we here investigated how perceptual ratings of physical characteristics from 2D and 3D face images compare. Methods We tested whether 3D face images capture cues of strength, weight, and height better than 2D face images do by directly comparing the accuracy of strength, weight, and height ratings of 182 2D and 3D face images taken simultaneously. Strength, height and weight were rated by 66, 59 and 52 raters respectively, who viewed both 2D and 3D images. Results In line with previous studies, we found that weight and height can be judged somewhat accurately from faces; contrary to previous research, we found that people were relatively inaccurate at assessing strength. We found no evidence that physical characteristics could be judged more accurately from 3D than 2D images. Conclusion Our results suggest physical characteristics are perceived with similar accuracy from 2D and 3D face images. They also suggest that the substantial costs associated with collecting 3D face scans may not be justified for research on the accuracy of facial judgments of physical characteristics.

Two-dimensional-Wilkins and real-time transesophageal three-dimensional scoring systems: Which one is preferred to mitral valve morphological assessment?

2021 ◽  
pp. 021849232110304
Author(s):  
Mehrnoush Toufan ◽  
Zahra Jabbary ◽  
Naser Khezerlou aghdam
Keyword(s):  
Mitral Valve ◽  
Transthoracic Echocardiography ◽  
Scoring System ◽  
Three Dimensional ◽  
Scoring Systems ◽  
Two Dimensional ◽  
Mitral Valve Area ◽  
2D And 3D ◽  
Morphological Assessment ◽  
Valve Area

Background To quantify valvular morphological assessment, some two-dimensional (2D) and three-dimensional (3D) scoring systems have been developed to target the patients for balloon mitral valvuloplasty; however, each scoring system has some potential limitations. To achieve the best scoring system with the most features and the least restrictions, it is necessary to check the degree of overlap of these systems. Also the factors related to the accuracy of these systems should be studied. We aimed to determine the correlation between the 2D Wilkins and real-time transesophageal three-dimensional (RT3D-TEE) scoring systems. Methods This cross-sectional study was performed on 156 patients with moderate to severe mitral stenosis who were candidates for percutaneous balloon valvuloplasty. To morphologic assessment of mitral valve, patients were examined by 2D-transthoracic echocardiography and RT3D-TEE techniques on the same day. Results A strong association was found between total Wilkins and total RT3D-TEE scores (r = 0.809, p < 0.001). The mean mitral valve area assessed by the 2D and 3D was 1.07 ± 0.25 and 1.03 ± 0.26, respectively, indicating a mean difference of 0.037 cm2 (p = 0.001). We found a strong correlation between the values of mitral valve area assessed by 2D and 3D techniques (r = 0.846, p < 0.001). Conclusion There is a high correlation between the two scoring systems in terms of evaluating dominant morphological features. Partially, mitral valve area overestimation in the 2D-transthoracic echocardiography and its inability to assess commissural involvement as well as its dependence on patient age were exceptions in this study.

Transformations of two-dimensional layered zinc phosphates to three-dimensional and one-dimensional structures

2002 ◽  
Vol 12 (4) ◽  
pp. 1044-1052 ◽  
Author(s):  
Amitava Choudhury ◽  
S. Neeraj ◽  
Srinivasan Natarajan ◽  
C. N. R. Rao
Keyword(s):  
Three Dimensional ◽  
Two Dimensional ◽  
One Dimensional ◽  
Zinc Phosphates

Mine Impact Burial Prediction From One to Three Dimensions

2008 ◽  
Vol 62 (1) ◽  
Author(s):  
Peter C. Chu
Keyword(s):  
Three Dimensional ◽  
Time History ◽  
Three Dimensions ◽  
Vertical Position ◽  
Burial Depth ◽  
Prediction Errors ◽  
Two Dimensional ◽  
Dimensional Model ◽  
One Dimensional ◽  
Dimensional Models

The Navy’s mine impact burial prediction model creates a time history of a cylindrical or a noncylindrical mine as it falls through air, water, and sediment. The output of the model is the predicted mine trajectory in air and water columns, burial depth/orientation in sediment, as well as height, area, and volume protruding. Model inputs consist of parameters of environment, mine characteristics, and initial release. This paper reviews near three decades’ effort on model development from one to three dimensions: (1) one-dimensional models predict the vertical position of the mine’s center of mass (COM) with the assumption of constant falling angle, (2) two-dimensional models predict the COM position in the (x,z) plane and the rotation around the y-axis, and (3) three-dimensional models predict the COM position in the (x,y,z) space and the rotation around the x-, y-, and z-axes. These models are verified using the data collected from mine impact burial experiments. The one-dimensional model only solves one momentum equation (in the z-direction). It cannot predict the mine trajectory and burial depth well. The two-dimensional model restricts the mine motion in the (x,z) plane (which requires motionless for the environmental fluids) and uses incorrect drag coefficients and inaccurate sediment dynamics. The prediction errors are large in the mine trajectory and burial depth prediction (six to ten times larger than the observed depth in sand bottom of the Monterey Bay). The three-dimensional model predicts the trajectory and burial depth relatively well for cylindrical, near-cylindrical mines, and operational mines such as Manta and Rockan mines.

Theory of ultrasonic attenuation in one and two dimensional metals

1976 ◽  
Vol 54 (14) ◽  
pp. 1454-1460 ◽  
Author(s):  
T. Tiedje ◽  
R. R. Haering
Keyword(s):  
Ultrasonic Attenuation ◽  
Three Dimensional ◽  
Electronic Systems ◽  
Two Dimensional ◽  
Temperature Dependent ◽  
One Dimensional ◽  
The Difference

The theory of ultrasonic attenuation in metals is extended so that it applies to quasi one and two dimensional electronic systems. It is shown that the attenuation in such systems differs significantly from the well-known results for three dimensional systems. The difference is particularly marked for one dimensional systems, for which the attenuation is shown to be strongly temperature dependent.

Sign in / Sign up

Export Citation Format

Share Document