Synthetic Minority Oversampling Technique and Fractal Dimension for Identifying Multiple Sclerosis

Zhang, Yudong; Zhang, Yin; Phillips, P. Jonathon; Dong, Zhengchao; Wang, Shuihua

doi:10.1142/s0218348x17400102

Cited by 36 publications

(22 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Biological systems have been studied under the fractal perspective (Copley et al, 2012 ; Wedman et al, 2015 ; Lennon et al, 2016 ; Stankovic et al, 2016 ). There are reports suggesting that the fractal dimension discriminates between healthy and pathological conditions (Hiroshima et al, 2016 ; Zehani et al, 2016 ; Müller et al, 2017 ; Zhang et al, 2017 ; Popovic et al, 2018 ). In the cardiac context, structural remodeling generates significant fractal dimension variations (Zouein et al, 2014 ; Captur et al, 2016 ).…”

Section: Introductionmentioning

confidence: 99%

Atrial Rotor Dynamics Under Complex Fractional Order Diffusion

et al. 2018

View full text Add to dashboard Cite

The mechanisms of atrial fibrillation (AF) are a challenging research topic. The rotor hypothesis states that the AF is sustained by a reentrant wave that propagates around an unexcited core. Cardiac tissue heterogeneities, both structural and cellular, play an important role during fibrillatory dynamics, so that the ionic characteristics of the currents, their spatial distribution and their structural heterogeneity determine the meandering of the rotor. Several studies about rotor dynamics implement the standard diffusion equation. However, this mathematical scheme carries some limitations. It assumes the myocardium as a continuous medium, ignoring, therefore, its discrete and heterogeneous aspects. A computational model integrating both, electrical and structural properties could complement experimental and clinical results. A new mathematical model of the action potential propagation, based on complex fractional order derivatives is presented. The complex derivative order appears of considering the myocardium as discrete-scale invariant fractal. The main aim is to study the role of a myocardial, with fractal characteristics, on atrial fibrillatory dynamics. For this purpose, the degree of structural heterogeneity is described through derivatives of complex order γ = α + jβ. A set of variations for γ is tested. The real part α takes values ranging from 1.1 to 2 and the imaginary part β from 0 to 0.28. Under this scheme, the standard diffusion is recovered when α = 2 and β = 0. The effect of γ on the action potential propagation over an atrial strand is investigated. Rotors are generated in a 2D model of atrial tissue under electrical remodeling due to chronic AF. The results show that the degree of structural heterogeneity, given by γ, modulates the electrophysiological properties and the dynamics of rotor-type reentrant mechanisms. The spatial stability of the rotor and the area of its unexcited core are modulated. As the real part decreases and the imaginary part increases, simulating a higher structural heterogeneity, the vulnerable window to reentrant is increased, as the total meandering of the rotor tip. This in silico study suggests that structural heterogeneity, described by means of complex order derivatives, modulates the stability of rotors and that a wide range of rotor dynamics can be generated.

show abstract

Section: Introductionmentioning

confidence: 99%

Atrial Rotor Dynamics Under Complex Fractional Order Diffusion

et al. 2018

View full text Add to dashboard Cite

show abstract

“…These approaches can be mainly categorized into two levels: data level approaches and algorithm level approaches. Approaches at data level try to rebalance the data distribution by sampling the data space such that the conventional learning methods can capture the characteristics of the minority class [8][9][10][11][12][13][14][15][16][17]. Approaches at the algorithm level try to improve the generalization ability of existing algorithms on imbalanced data by adjusting the learning process of the algorithms.…”

Section: Introductionmentioning

confidence: 99%

Imbalanced Learning Based on Data-Partition and SMOTE

Guo

Zhou

2018

Information

View full text Add to dashboard Cite

Classification of data with imbalanced class distribution has encountered a significant drawback by most conventional classification learning methods which assume a relatively balanced class distribution. This paper proposes a novel classification method based on data-partition and SMOTE for imbalanced learning. The proposed method differs from conventional ones in both the learning and prediction stages. For the learning stage, the proposed method uses the following three steps to learn a class-imbalance oriented model: (1) partitioning the majority class into several clusters using data partition methods such as K-Means, (2) constructing a novel training set using SMOTE on each data set obtained by merging each cluster with the minority class, and (3) learning a classification model on each training set using convention classification learning methods including decision tree, SVM and neural network. Therefore, a classifier repository consisting of several classification models is constructed. With respect to the prediction stage, for a given example to be classified, the proposed method uses the partition model constructed in the learning stage to select a model from the classifier repository to predict the example. Comprehensive experiments on KEEL data sets show that the proposed method outperforms some other existing methods on evaluation measures of recall, g-mean, f-measure and AUC.

show abstract

“…The future work is to test other manual features, and try to enroll more students to enlarge our dataset. In addition, transfer learning [29] technique and fractal [30] will be tested. Besides, our method can be applied to detect abnormal breast [31][32][33] and other diseases.…”

Section: Resultsmentioning

confidence: 99%

Hu Moment Invariant: A New Method for Hearing Loss Detection

Tang¹,

Yixuan²,

Pereira³

2018

Proceedings of the 2017 International Conference Advanced Engineering and Technology Research (AETR 2017)

View full text Add to dashboard Cite

This paper proposed a novel hearing loss detection method. Our method first used seven Hu moment invariants to extract features. Afterwards, we used support vector machine to act as the classifier. The 10x5-fold cross validation shows our method yielded an overall accuracy of 77.47± 1.17%. The sensitivities of healthy control, left-sided hearing loss, and right-sided hearing loss are 77.60± 5.72%, 77.60± 4.30%, and 77.20± 5.98%, respectively. In all, our method is effective in hearing loss identification.

show abstract

Synthetic Minority Oversampling Technique and Fractal Dimension for Identifying Multiple Sclerosis

Cited by 36 publications

References 62 publications

Atrial Rotor Dynamics Under Complex Fractional Order Diffusion

Atrial Rotor Dynamics Under Complex Fractional Order Diffusion

Imbalanced Learning Based on Data-Partition and SMOTE

Hu Moment Invariant: A New Method for Hearing Loss Detection

Contact Info

Product

Resources

About