scholarly journals Machine Learning Application for Rupture Risk Assessment in Small-Sized Intracranial Aneurysm

2019 ◽  
Vol 8 (5) ◽  
pp. 683 ◽  
Author(s):  
Heung Cheol Kim ◽  
Jong Kook Rhim ◽  
Jun Hyong Ahn ◽  
Jeong Jin Park ◽  
Jong Un Moon ◽  
...  
Keyword(s):  
Diagnostic Accuracy ◽  
Detection System ◽  
Three Dimensional ◽  
Region Of Interest ◽  
Large Data ◽  
Aneurysm Rupture ◽  
Computer Assisted ◽  
Rupture Risk ◽  
Operating Characteristics ◽  
Data Set

The assessment of rupture probability is crucial to identifying at risk intracranial aneurysms (IA) in patients harboring multiple aneurysms. We aimed to develop a computer-assisted detection system for small-sized aneurysm ruptures using a convolutional neural network (CNN) based on images of three-dimensional digital subtraction angiography. A retrospective data set, including 368 patients, was used as a training cohort for the CNN using the TensorFlow platform. Aneurysm images in six directions were obtained from each patient and the region-of-interest in each image was extracted. The resulting CNN was prospectively tested in 272 patients and the sensitivity, specificity, overall accuracy, and receiver operating characteristics (ROC) were compared to a human evaluator. Our system showed a sensitivity of 78.76% (95% CI: 72.30%–84.30%), a specificity of 72.15% (95% CI: 60.93%–81.65%), and an overall diagnostic accuracy of 76.84% (95% CI: 71.36%–81.72%) in aneurysm rupture predictions. The area under the ROC (AUROC) in the CNN was 0.755 (95% CI: 0.699%–0.805%), better than that obtained from a human evaluator (AUROC: 0.537; p < 0.001). The CNN-based prediction system was feasible to assess rupture risk in small-sized aneurysms with diagnostic accuracy superior to human evaluators. Additional studies based on a large data set are necessary to enhance diagnostic accuracy and to facilitate clinical application.

scholarly journals Solvent Accessibility of Residues Undergoing Pathogenic Variations in Humans: From Protein Structures to Protein Sequences

2021 ◽  
Vol 7 ◽  
Author(s):  
Castrense Savojardo ◽  
Matteo Manfredi ◽  
Pier Luigi Martelli ◽  
Rita Casadio
Keyword(s):  
Solvent Accessibility ◽  
Protein Structures ◽  
Three Dimensional ◽  
Protein Sequences ◽  
Large Data ◽  
Human Protein ◽  
Dimensional Structure ◽  
Wild Type ◽  
Solvent Exposure ◽  
Data Set

Solvent accessibility (SASA) is a key feature of proteins for determining their folding and stability. SASA is computed from protein structures with different algorithms, and from protein sequences with machine-learning based approaches trained on solved structures. Here we ask the question as to which extent solvent exposure of residues can be associated to the pathogenicity of the variation. By this, SASA of the wild-type residue acquires a role in the context of functional annotation of protein single-residue variations (SRVs). By mapping variations on a curated database of human protein structures, we found that residues targeted by disease related SRVs are less accessible to solvent than residues involved in polymorphisms. The disease association is not evenly distributed among the different residue types: SRVs targeting glycine, tryptophan, tyrosine, and cysteine are more frequently disease associated than others. For all residues, the proportion of disease related SRVs largely increases when the wild-type residue is buried and decreases when it is exposed. The extent of the increase depends on the residue type. With the aid of an in house developed predictor, based on a deep learning procedure and performing at the state-of-the-art, we are able to confirm the above tendency by analyzing a large data set of residues subjected to variations and occurring in some 12,494 human protein sequences still lacking three-dimensional structure (derived from HUMSAVAR). Our data support the notion that surface accessible area is a distinguished property of residues that undergo variation and that pathogenicity is more frequently associated to the buried property than to the exposed one.

Image-guided vascular neurosurgery based on three-dimensional rotational angiography

2007 ◽  
Vol 106 (3) ◽  
pp. 501-506 ◽  
Author(s):  
Peter W. A. Willems ◽  
Theo Van Walsum ◽  
Peter A. Woerdeman ◽  
Everine B. Van De Kraats ◽  
Gerard A. P. De Kort ◽  
...  
Keyword(s):  
Image Registration ◽  
Navigation System ◽  
Three Dimensional ◽  
Region Of Interest ◽  
Small Region ◽  
Rotational Angiography ◽  
Vascular Tree ◽  
Data Set ◽  
3D Rotational Angiography ◽  
Vascular Neurosurgery

✓Three-dimensional rotational angiography is capable of exquisite visualization of cerebral blood vessels and their pathophysiology. Unfortunately, images obtained using this modality typically show a small region of interest without exterior landmarks to allow patient-to-image registration, precluding their use for neuronavigation purposes. The aim of this study was to find an alternative technique to enable 3D rotational angiography–guided vascular neurosurgery. Three-dimensional rotational angiograms were obtained in an angiographic suite with direct navigation capabilities. After image acquisition, a navigated pointer was used to touch fiducial positions on the patient's head. These positions were located outside the image volume but could nevertheless be transformed into image coordinates and stored in the navigation system. Prior to surgery, the data set was transferred to the navigation system in the operating room, and the same fiducial positions were touched again to complete the patient-to-image registration. This technique was tested on a Perspex phantom representing the cerebral vascular tree and on two patients with an intracranial aneurysm. In both the phantom and patients, the neuronavigation system provided 3D images representing the vascular tree in its correct orientation, that is, the orientation seen by the neurosurgeon through the microscope. In one patient, tissue shift was clearly observed without significant changes in the orientation of the structures. Results in this study demonstrate the feasibility of using 3D rotational angiography data sets for neuronavigation purposes. Determining the benefit of this type of navigation should be the subject of future studies.

scholarly journals Coheritability and Coenvironmentability as Concepts for Partitioning the Phenotypic Correlation

2019 ◽  
Author(s):  
Jorge Vasquez-Kool
Keyword(s):  
Genetic Parameter ◽  
Three Dimensional ◽  
Large Data ◽  
Phenotypic Correlation ◽  
Shared Environment ◽  
Parameter Estimates ◽  
Extensive Literature ◽  
Data Set ◽  
Phenotypic Characters ◽  
Relative Contribution

AbstractCentral to the study of joint inheritance of quantitative traits is the determination of the degree of association between two phenotypic characters, and to quantify the relative contribution of shared genetic and environmental components influencing such relationship. One way to approach this problem builds on classical quantitative genetics theory, where the phenotypic correlation between two traits is modelled as the sum of a genetic component called the coheritability (hx,y), which reflects the degree of shared genetics influencing the phenotypic correlation, and an environmental component, namely the coenvironmentability (ex,y) that accounts for all other factors that exert influence on the observed trait-trait association. Here a mathematical and statistical framework is presented on the partition of the phenotypic correlation into these components. I describe visualization tools to analyze and ex,y concurrently, in the form of a three-dimensional (3DHER-plane) and a two-dimensional (2DHER-field) plots. A large data set of genetic parameter estimates (heritabilities, genetic and phenotypic correlations) was compiled from an extensive literature review, from which coheritability and coenvironmentability were derived, with the object to observe patterns of distribution, and tendency. Illustrative examples from a diverse set of published studies show the value of applying this partition to generate hypotheses proposing the differential contribution of shared genetics and shared environment to an observed phenotypic relationship between traits.

Rapidly detecting icebergs using ArcticDEM and Google Earth Engine

2021 ◽  
Author(s):  
Connor Shiggins ◽  
James Lea ◽  
Dominik Fahrner ◽  
Stephen Brough
Keyword(s):  
Region Of Interest ◽  
Large Data ◽  
Automated Analysis ◽  
Google Earth ◽  
Data Availability ◽  
Data Set ◽  
High Data ◽  
Computing Platform ◽  
Google Earth Engine ◽  
Automated Tool

&lt;p&gt;High resolution digital elevation models (DEMs) allow for the detection of icebergs and their size distribution, potentially giving insights into spatial and temporal changes in calving dynamics and iceberg cover. Here we present a fully automated tool for iceberg detection in glaciated fjords, utilising timestamped ArcticDEM tile data within the Google Earth Engine cloud computing platform. The automated tool requires only definition of a region of interest (ROI) through the following workflow:&lt;/p&gt;&lt;p&gt;1. Automatically filter timestamped ArcticDEM tiles to obtain only high-quality images with high data coverage within a ROI&lt;/p&gt;&lt;p&gt;2. Apply elevation correction to account for the geoid and tidal state, ensuring sea level is the equivalent to 0 m elevation&lt;/p&gt;&lt;p&gt;3. Apply an iceberg detection elevation threshold (any object at/or above 0.9 m)&lt;/p&gt;&lt;p&gt;4. Automatically delineate icebergs based on elevations above this threshold&lt;/p&gt;&lt;p&gt;5. Iceberg area, volume (total, below and above surface), freeboard height, mass and the ArcticDEM acquisition date are appended to each iceberg&lt;/p&gt;&lt;p&gt;This workflow allows for rapid, fully automated analysis of all available ArcticDEM tiles within a given ROI. The workflow does not require manual supervision, and can be easily related back to the original ArcticDEM data through Google Earth Engine. As an example, we apply our workflow to a 33 km&lt;sup&gt;2&lt;/sup&gt; ROI at Nuup Kangerlua (Godth&amp;#229;bsfjorden), southwest Greenland, detecting a total of 57,735 icebergs from 6 images with an execution time of 19 minutes. This workflow will provide a user-friendly platform for users of any coding ability requiring a large data set of icebergs with an area size greater than approximately 40 m&lt;sup&gt;2&lt;/sup&gt;. Results obtained from these data will be utilised to identify potential seasonal to multi-annual timescale changes in calving behaviour, though is dependent on ArcticDEM data availability.&amp;#160;&lt;/p&gt;

FRACTAL ASPECTS OF PROTEIN STRUCTURE AND DYNAMICS

Fractals ◽  
1993 ◽  
Vol 01 (02) ◽  
pp. 179-189 ◽  
Author(s):  
T. GREGORY DEWEY
Keyword(s):  
Dynamic Properties ◽  
Linear Chain ◽  
Protein Structures ◽  
Three Dimensional ◽  
Alpha Helix ◽  
Large Data ◽  
Thermal Fluctuations ◽  
Radius Of Gyration ◽  
Data Set ◽  
Long Range Correlations

Proteins have well-defined three dimensional structures which are dictated by their amino acid sequence. Despite this great specificity, general structural and dynamic properties exist. Scaling relationships for the radius of gyration and surface area of a large data set of proteins are demonstrated in this work. These results show that proteins scale as collapsed polymers. Thermal fluctuations are examined for two different proteins by an analysis of the Debye-Waller factors derived from X-ray crystallographic data. Long-range correlations exist between fluctuations along the backbone. A disordered Ising model is presented which gives similar correlations. To further examine the role of multiple connectivity in protein structures, the vibrational spectrum for an alpha helix (linear chain with H-bonds) is analyzed from recursive relationships derived using a decimation technique.

scholarly journals Visualization of High-Dimensional Data by Pairwise Fusion Matrices Using t-SNE

Symmetry ◽  
2019 ◽  
Vol 11 (1) ◽  
pp. 107 ◽  
Author(s):  
Mujtaba Husnain ◽  
Malik Missen ◽  
Shahzad Mumtaz ◽  
Muhammad Luqman ◽  
Mickaël Coustaty ◽  
...  
Keyword(s):  
Local Structure ◽  
High Dimensional Data ◽  
Three Dimensional ◽  
Principal Component ◽  
Large Data ◽  
High Dimensional ◽  
Data Set ◽  
Novel Approach ◽  
Critical Issues ◽  
Low Dimensional

We applied t-distributed stochastic neighbor embedding (t-SNE) to visualize Urdu handwritten numerals (or digits). The data set used consists of 28 × 28 images of handwritten Urdu numerals. The data set was created by inviting authors from different categories of native Urdu speakers. One of the challenging and critical issues for the correct visualization of Urdu numerals is shape similarity between some of the digits. This issue was resolved using t-SNE, by exploiting local and global structures of the large data set at different scales. The global structure consists of geometrical features and local structure is the pixel-based information for each class of Urdu digits. We introduce a novel approach that allows the fusion of these two independent spaces using Euclidean pairwise distances in a highly organized and principled way. The fusion matrix embedded with t-SNE helps to locate each data point in a two (or three-) dimensional map in a very different way. Furthermore, our proposed approach focuses on preserving the local structure of the high-dimensional data while mapping to a low-dimensional plane. The visualizations produced by t-SNE outperformed other classical techniques like principal component analysis (PCA) and auto-encoders (AE) on our handwritten Urdu numeral dataset.

A COMPUTER-AIDED SYSTEM FOR MASS DETECTION AND CLASSIFICATION IN DIGITIZED MAMMOGRAMS

2005 ◽  
Vol 17 (05) ◽  
pp. 215-228 ◽  
Author(s):  
SHENG-CHIH YANG ◽  
CHUIN-MU WANG ◽  
YI-NUNG CHUNG ◽  
GIU-CHENG HSU ◽  
SAN-KAN LEE ◽  
...  
Keyword(s):  
Probabilistic Neural Network ◽  
Diagnostic System ◽  
Region Of Interest ◽  
Texture Features ◽  
Computer Assisted ◽  
Mass Detection ◽  
Difference Matrix ◽  
Data Set ◽  
System A ◽  
Mass Classification

This paper presents a computer-assisted diagnostic system for mass detection and classification, which performs mass detection on regions of interest followed by the benign-malignant classification on detected masses. In order for mass detection to be effective, a sequence of preprocessing steps are designed to enhance the intensity of a region of interest, remove the noise effects and locate suspicious masses using five texture features generated from the spatial gray level difference matrix (SGLDM) and fractal dimension. Finally, a probabilistic neural network (PNN) coupled with entropic thresholding techniques is developed for mass extraction. Since the shapes of masses are crucial in classification between benignancy and malignancy, four shape features are further generated and joined with the five features previously used in mass detection to be implemented in another PNN for mass classification. To evaluate our designed system a data set collected in the Taichung Veteran General Hospital, Taiwan, R.O.C. was used for performance evaluation. The results are encouraging and have shown promise of our system.

scholarly journals Reappraisal of shear wave elastography as a diagnostic tool for identifying thyroid carcinoma

2019 ◽  
Vol 8 (8) ◽  
pp. 1195-1205 ◽  
Author(s):  
Kristine Zøylner Swan ◽  
Steen Joop Bonnema ◽  
Marie Louise Jespersen ◽  
Viveque Egsgaard Nielsen
Keyword(s):  
Diagnostic Accuracy ◽  
Shear Wave ◽  
Thyroid Nodules ◽  
Area Under The Curve ◽  
Region Of Interest ◽  
Shear Wave Elastography ◽  
Thyroid Malignancy ◽  
Operating Characteristics ◽  
Factors Affecting ◽  
Benign Nodules

Thyroid nodular disease is common, but predicting the risk of malignancy can be difficult. In this prospective study, we aimed to assess the diagnostic accuracy of shear wave elastography (SWE) in predicting thyroid malignancy. Patients with thyroid nodules were enrolled from a surgical tertiary unit. Elasticity index (EI) measured by SWE was registered for seven EI outcomes assessing nodular stiffness and heterogeneity. The diagnosis was determined histologically. In total, 329 patients (mean age: 55 ± 13 years) with 413 thyroid nodules (mean size: 32 ± 13 mm, 88 malignant) were enrolled. Values of SWE region of interest (ROI) for malignant and benign nodules were highly overlapping (ranges for SWE-ROImean: malignant 3–100 kPa; benign 4–182 kPa), and no difference between malignant and benign nodules was found for any other EI outcome investigated (P = 0.13–0.96). There was no association between EI and the histological diagnosis by receiver operating characteristics analysis (area under the curve: 0.51–0.56). Consequently, defining a cut-off point of EI for the prediction of malignancy was not clinically meaningful. Testing our data on previously proposed cut-off points revealed a low accuracy of SWE (56–80%). By regression analysis, factors affecting EI included nodule size >30 mm, heterogeneous echogenicity, micro- or macrocalcifications and solitary nodule. In conclusion, EI, measured by SWE, showed huge overlap between malignant and benign nodules, and low diagnostic accuracy in the prediction of thyroid malignancy. Our study supports that firmness of thyroid nodules, as assessed by SWE, should not be a key feature in the evaluation of such lesions.

Optimization of Rock Berms for Pipeline Stabilization Subject to Intense Hydrodynamic Forcing

2015 ◽  
Author(s):  
Andrew Cornett ◽  
Scott Baker ◽  
Peter Riedel ◽  
Paul Knox
Keyword(s):  
Tropical Cyclones ◽  
Three Dimensional ◽  
Large Data ◽  
Physical Modelling ◽  
Cost Effective ◽  
Data Set ◽  
Waves And Currents ◽  
2D And 3D ◽  
Water Depths ◽  
Comprehensive Study

This article describes a comprehensive study in which 2D and 3D physical modelling at 1:40 scale was used to optimize the design and validate the performance of dynamically stable rock berms to be used for stabilizing several large pipelines traversing water depths from 5m to 65m and potentially exposed to large waves and strong currents generated by intense tropical cyclones. For added realism, all of the model rock berms were constructed using a scaled simulation of rock installation by fall pipe vessel to be used in the field. Special attention was also given to simulating the self-stability of the model pipeline segments, including special end constraints designed to mimic the behaviour of a continuous pipeline. A large data set concerning the behaviour of dynamically reshaping rock berms in a range of water depths under intense hydrodynamic forcing due to three-dimensional waves and currents was produced and used to develop efficient and cost-effective rock berm designs for all depth zones.

Sign in / Sign up

Export Citation Format

Share Document