Analysis of Radiographic Images of Patients with COVID-19 Using Fractal Dimension and Complex Network-Based High-Level Classification

2022 ◽  
pp. 16-26
Author(s):  
Weiguang Liu ◽  
Jianglong Yan ◽  
Yu-tao Zhu ◽  
Everson José de Freitas Pereira ◽  
Gen Li ◽  
...  
Keyword(s):  
Fractal Dimension ◽  
Complex Network ◽  
Radiographic Images ◽  
High Level

ALGORITHMS FOR FRACTAL DIMENSION CALCULATION

2008 ◽  
Vol 22 (07) ◽  
pp. 459-466 ◽  
Author(s):  
O. SHANKER
Keyword(s):  
Fractal Dimension ◽  
Complex Network ◽  
System Size

Algorithms to calculate the fractal dimension of a complex network are presented. One of the algorithms is applied to a parametrized class of models whose fractal dimension transitions from one to two. For the system size we considered here (16384 nodes), the transition takes place from one at p = 0 to essentially two at the small value p = 0.03. This seems to indicate that the transition is likely to become infinitely sharp and occur at p = 0 as the system size increases to infinity.

PLANT LEAF IDENTIFICATION BASED ON VOLUMETRIC FRACTAL DIMENSION

2009 ◽  
Vol 23 (06) ◽  
pp. 1145-1160 ◽  
Author(s):  
ANDRÉ RICARDO BACKES ◽  
DALCIMAR CASANOVA ◽  
ODEMIR MARTINEZ BRUNO
Keyword(s):  
Fractal Dimension ◽  
Fourier Analysis ◽  
Complex Task ◽  
Classification Rate ◽  
Method Performance ◽  
Analysis Process ◽  
Novel Approach ◽  
Visual Attribute ◽  
Novel Method ◽  
High Level

Texture is an important visual attribute used to describe the pixel organization in an image. As well as it being easily identified by humans, its analysis process demands a high level of sophistication and computer complexity. This paper presents a novel approach for texture analysis, based on analyzing the complexity of the surface generated from a texture, in order to describe and characterize it. The proposed method produces a texture signature which is able to efficiently characterize different texture classes. The paper also illustrates a novel method performance on an experiment using texture images of leaves. Leaf identification is a difficult and complex task due to the nature of plants, which presents a huge pattern variation. The high classification rate yielded shows the potential of the method, improving on traditional texture techniques, such as Gabor filters and Fourier analysis.

scholarly journals Integrated Approach for Automatic Crackle Detection Based on Fractal Dimension and Box Filtering

2016 ◽  
Vol 5 (4) ◽  
pp. 34-50 ◽  
Author(s):  
Cátia Pinho ◽  
Ana Oliveira ◽  
Cristina Jácome ◽  
João Manuel Rodrigues ◽  
Alda Marques
Keyword(s):  
Cystic Fibrosis ◽  
Fractal Dimension ◽  
Gold Standard ◽  
Integrated Approach ◽  
Respiratory Sounds ◽  
Clinical Settings ◽  
Algorithm Performance ◽  
Overall Performance ◽  
High Level ◽  
Respiratory Conditions

Crackles are adventitious respiratory sounds (RS) that provide valuable information on different respiratory conditions. Nevertheless, crackles automatic detection in RS is challenging, mainly when collected in clinical settings. This study aimed to develop an algorithm for automatic crackle detection/characterisation and to evaluate its performance and accuracy against a multi-annotator gold standard. The algorithm is based on 4 main procedures: i) recognition of a potential crackle; ii) verification of its validity; iii) characterisation of crackles parameters; and iv) optimisation of the algorithm parameters. Twenty-four RS files acquired in clinical settings were selected from 10 patients with pneumonia and cystic fibrosis. The algorithm performance was assessed by comparing its results with a multi-annotator gold standard agreement. High level of overall performance (F-score=92%) was achieved. The results highlight the potential of the algorithm for automatic crackle detection and characterisation of RS acquired in clinical settings.

Analysis of Complex Network Measures for Multi-label Classification

2021 ◽  
Vol 30 (04) ◽  
pp. 2150023
Author(s):  
Vinícius H. Resende ◽  
Murillo G. Carneiro
Keyword(s):  
Complex Network ◽  
Predictive Performance ◽  
Clustering Coefficient ◽  
Data Sets ◽  
Real World Data ◽  
Network Measures ◽  
Formation Pattern ◽  
Network Properties ◽  
High Level ◽  
Physical Features

Most multi-label learning (MLL) techniques perform classification by analyzing only the physical features of the data, which means they are unable to consider high-level features, such as structural and topological ones. Consequently, they have trouble to detect the semantic meaning of the data (e.g., formation pattern). To handle this problem, a high-level framework has been recently proposed to the MLL task, in which the high-level features are extracted using the analysis of complex network measures. In this paper, we extend that work by evaluating different combinations of four complex networks measures, namely clustering coefficient, assortativity, average degree and average path length. Experiments conducted over seven real-world data sets showed that the low-level techniques often can have their predictive performance improved after being combined with high-level ones, and also demonstrated that there is no a unique measure that provides the best results, i.e., different problems may ask for different network properties in order to have their high-level patterns efficiently detected.

scholarly journals An Object-Oriented HLA Simulation Study

2015 ◽  
Vol 15 (5) ◽  
pp. 121-130
Author(s):  
Georgi Kirov
Keyword(s):  
Software Architecture ◽  
Complex Network ◽  
Simulation Study ◽  
Data Exchange ◽  
Object Oriented ◽  
Object Oriented Programming ◽  
Network Programming ◽  
High Level Architecture ◽  
Distributed Simulations ◽  
High Level

Abstract The study is dedicated to High Level Architecture (HLA) standard for software architecture of interoperable distributed simulations. The paper discusses the differences between object-oriented programming and HLA. It presents an extended simulation architecture providing a mechanism for HLA data exchange through Object-Oriented (OO) objects. This eliminates the complex network programming for HLA distributed simulations. The paper shows a sample code that implements the architecture for OO HLA/RTI simulation.

scholarly journals How to calculate the fractal dimension of a complex network: the box covering algorithm

2007 ◽  
Vol 2007 (03) ◽  
pp. P03006-P03006 ◽  
Author(s):  
Chaoming Song ◽  
Lazaros K Gallos ◽  
Shlomo Havlin ◽  
Hernán A Makse
Keyword(s):  
Fractal Dimension ◽  
Complex Network

scholarly journals Integrated Approach for Automatic Crackle Detection Based on Fractal Dimension and Box Filtering

2020 ◽  
pp. 815-832
Author(s):  
Cátia Pinho ◽  
Ana Oliveira ◽  
Cristina Jácome ◽  
João Manuel Rodrigues ◽  
Alda Marques
Keyword(s):  
Cystic Fibrosis ◽  
Fractal Dimension ◽  
Gold Standard ◽  
Integrated Approach ◽  
Respiratory Sounds ◽  
Clinical Settings ◽  
Algorithm Performance ◽  
Overall Performance ◽  
High Level ◽  
Respiratory Conditions

Crackles are adventitious respiratory sounds (RS) that provide valuable information on different respiratory conditions. Nevertheless, crackles automatic detection in RS is challenging, mainly when collected in clinical settings. This study aimed to develop an algorithm for automatic crackle detection/characterisation and to evaluate its performance and accuracy against a multi-annotator gold standard. The algorithm is based on 4 main procedures: i) recognition of a potential crackle; ii) verification of its validity; iii) characterisation of crackles parameters; and iv) optimisation of the algorithm parameters. Twenty-four RS files acquired in clinical settings were selected from 10 patients with pneumonia and cystic fibrosis. The algorithm performance was assessed by comparing its results with a multi-annotator gold standard agreement. High level of overall performance (F-score=92%) was achieved. The results highlight the potential of the algorithm for automatic crackle detection and characterisation of RS acquired in clinical settings.

Swelling characteristics of bentonite after long-term dissolution in alkaline solution

Clay Minerals ◽  
2019 ◽  
Vol 54 (4) ◽  
pp. 409-416
Author(s):  
Guo-sheng Xiang ◽  
Wei-min Ye ◽  
Li-yong Lv
Keyword(s):  
Fractal Dimension ◽  
Reaction Time ◽  
Alkaline Solution ◽  
Void Ratio ◽  
Swelling Properties ◽  
Swelling Characteristics ◽  
Naoh Solution ◽  
Swelling Coefficient ◽  
High Level

AbstractIn a high-level radioactive waste repository, bentonite may react with the alkaline solution produced by cement degradation. In this study, bentonite was mixed with alkaline solution in a closed system and reacted for 3–24 months. Furthermore, swelling tests were conducted on the alkaline-dissolved bentonite immersed in distilled water. The swelling deformation decreased significantly with increases in the concentration of NaOH solution and reaction time, and this was mainly due to montmorillonite dissolution. The fractal e–p relationship (e is the void ratio and p is the vertical pressure) with two calculation coefficients (the swelling coefficient and the fractal dimension) was employed to determine the swelling of alkaline-dissolved bentonite. The fractal dimension increased slightly with increasing reaction time or concentration of NaOH solution, as the dissolution traces caused by the alkaline solution favoured an increase in the irregularity and fractality of the bentonite surface. The swelling coefficient decreased linearly with decreasing montmorillonite content. In addition, the swelling coefficient and the fractal dimension were related exponentially to the reaction time in alkaline solution. A relationship between the swelling of alkaline-dissolved samples and the reaction time was proposed, which might be used to assess the swelling properties of bentonite barriers that would be affected by long-term dissolution of the alkaline solution in a closed repository.

scholarly journals Correlation Between Bone Density and Fractal Dimension: A Pilot Study

2020 ◽  
Vol 23 (2) ◽  
pp. 130-132
Author(s):  
Vittoria Perrotti ◽  
Giovanna Iezzi ◽  
Angela De Sanctis ◽  
Antonio Pasculli ◽  
Adriano Piattelli ◽  
...  
Keyword(s):  
Fractal Dimension ◽  
Pilot Study ◽  
Linear Dependence ◽  
Structural Changes ◽  
Method Comparison ◽  
P Value ◽  
Strong Positive Correlation ◽  
Radiographic Images ◽  
Non Invasive ◽  
Bone Trabeculae

Fractal analysis is a mathematical method used to describe the internal architecture of complex structures, such as the bone tissue. The aim of this study was to determine whether fractal dimension (FD) is able to distinguish between different bone densities (BD), assessed histomorphometrically, and whether there is a linear dependence between the FD and BD in order to support the use of FD as a supplementary non-invasive method for BD measurement. Microscopic photographs of bone specimens stained with acid fuchsin and toluidine blue obtained during block biopsies from nine segments of bovine ribs were used. A total of 42 regions of interest (ROI) were cut off from the original photo, converted into a bitmap and binarized. The evaluation of FD was carried out using the box counting method. Comparison of FD values in the four different densities bone sites (D1-D4) was made by means of Kruskal-Wallis test followed by Dunn's multiple comparison test. The linear dependence between the two variables (FD and BD) was investigated calculating the Pearson's r test correlation coefficient, which was considered significant when P < .05. The more the bone was compact, the higher were FD values. A strong positive correlation between BD evaluated by histomorphometry and FD (R : 0:9651), p-value < .00001 was found. The increase in the values of the FD strongly correlated with the increase of the percentage of the bone trabeculae observed in the histological slides. This pilot study demonstrates that FD might be able to distinguish different densities in bone sites and that there is a linear correlation between BD and FD on histological samples. Future studies will be addressed at evaluating whether this data can be confirmed on a larger samples size and on radiographic images. This will be useful for the early and non-invasive detection of structural changes in the trabecular bone pattern during healing, inflammatory processes and pathologies associated to bone breakdown.

Sign in / Sign up

Export Citation Format

Share Document