The Differential Diagnosis of Multiple Sclerosis Using Convex Combination of Infinite Kernels

Karaca, Yeliz; Zhang, Yudong; Cattani, Carlo; Ayan, Uğur

doi:10.2174/1871527315666161024142439

Cited by 14 publications

(7 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The images were obtained by using an MRI scanner of 1.5 Tesla. 14 MRI were taken from 120 MS subgroup patients (76 RRMS, 38 SPMS, 6 PPMS) and 19 healthy (Not-MS) individuals in three different nonconsecutive years (with a minimum of three years between the first and second MRI and a maximum of eight years between the second and third ones).…”

Section: Patient Detailsmentioning

confidence: 99%

See 1 more Smart Citation

Clustering Multiple Sclerosis Subgroups With Multifractal Methods and Self-Organizing Map Algorithm

Karaca

Cattani

2017

Fractals

Self Cite

View full text Add to dashboard Cite

Magnetic resonance imaging (MRI) is the most sensitive method to detect chronic nervous system diseases such as multiple sclerosis (MS). In this paper, Brownian motion Hölder regularity functions (polynomial, periodic (sine), exponential) for 2D image, such as multifractal methods were applied to MR brain images, aiming to easily identify distressed regions, in MS patients. With these regions, we have proposed an MS classification based on the multifractal method by using the Self-Organizing Map (SOM) algorithm. Thus, we obtained a cluster analysis by identifying pixels from distressed regions in MR images through multifractal methods and by diagnosing subgroups of MS patients through artificial neural networks.

show abstract

Section: Patient Detailsmentioning

confidence: 99%

“…MRI is an important medical practice to diagnose neurological diseases like MS. 14,15 MRI are used for displaying soft tissues in the brain and the spinal cord. An MRI can reveal inflammatory or damaged tissue in regions of the central nervous system.…”

Section: Magnetic Resonance Imagingmentioning

confidence: 99%

Clustering Multiple Sclerosis Subgroups With Multifractal Methods and Self-Organizing Map Algorithm

Karaca

Cattani

2017

Fractals

Self Cite

View full text Add to dashboard Cite

show abstract

“…In [22], the authors diagnose MS versus controls, comparing three machine-learning-based classifiers: the decision tree, k-NN and the support vector machine using the wavelet and wavelet entropy obtained with MRI. More recently, [23] classified MS subtypes based on features gathered from MRI and the Expanded Disability Status Scale using non-linear classification models (Convex Combination of Infinite Kernels).…”

Section: Introductionmentioning

confidence: 99%

A computer-aided diagnosis of multiple sclerosis based on mfVEP recordings

et al. 2019

View full text Add to dashboard Cite

Introduction The aim of this study is to develop a computer-aided diagnosis system to identify subjects at differing stages of development of multiple sclerosis (MS) using multifocal visual-evoked potentials (mfVEPs). Using an automatic classifier, diagnosis is performed first on the eyes and then on the subjects. Patients MfVEP signals were obtained from patients with Radiologically Isolated Syndrome (RIS) (n = 30 eyes), patients with Clinically Isolated Syndrome (CIS) (n = 62 eyes), patients with definite MS (n = 56 eyes) and 22 control subjects (n = 44 eyes). The CIS and MS groups were divided into two subgroups: those with eyes affected by optic neuritis (ON) and those without (non-ON). Methods For individual eye diagnosis, a feature vector was formed with information about the intensity, latency and singular values of the mfVEP signals. A flat multiclass classifier (FMC) and a hierarchical classifier (HC) were tested and both were implemented using the k-Nearest Neighbour (k-NN) algorithm. The output of the best eye classifier was used to classify the subjects. In the event of divergence, the eye with the best mfVEP recording was selected. Results In the eye classifier, the HC performed better than the FMC (accuracy = 0.74 and extended Matthew Correlation Coefficient (MCC) = 0.68). In the subject classification, accuracy = 0.95 and MCC = 0.93, confirming that it may be a promising tool for MS diagnosis. Conclusion In addition to amplitude (axonal loss) and latency (demyelination), it has shown that the singular values of the mfVEP signals provide discriminatory information that may be used to identify subjects with differing degrees of the disease.

show abstract

“…Zhan and Chen (2016) [22] used biorthogonal wavelet transform and logistic regression. Karaca (2017) [23] used convex combination of infinite kernels.…”

Section: Introductionmentioning

confidence: 99%

“…Zhan and Chen (2016) [22] used biorthogonal wavelet transform and logistic regression. Karaca (2017) [23] used convex combination of infinite kernels.Nevertheless, those methods are too complicated, and their models are difficult to train. In this study, we presented a novel and simple system, which was based on Haar wavelet transform, principle component, and logistic regression.…”

mentioning

confidence: 99%

Multiple Sclerosis Slice Identification by Haar Wavelet Transform and Logistic Regression

Wu¹,

Lopez²

2017

Proceedings of the Advances in Materials, Machinery, Electrical Engineering (AMMEE 2017)

View full text Add to dashboard Cite

Abstract. (Aim) Currently, scholars tend to use computer vision approaches to implement multiple sclerosis (MS) identification. (Method) In this study, we proposed a novel MS slice identification system, based on Haar wavelet transform, principal component analysis, and logistic regression. (Result) Simulation results showed the accuracies of our method using 2-level, 3-level, and 4-level decomposition are 83.25±1.62%, 89.72±1.18%, and 87.65±1.79%, respectively. (Conclusion) Our method with 3-level decomposition achieved the best. BackgroundMultiple sclerosis (MS) is a progressive disease that affects both brain and spinal cord [1]. Early symptoms are composed of tingling, weakness, blurred version, numbness, etc. [2]. The traditional MRI scanning based identification may meet with the "normal-appearing white matter" problem. Hence, it is necessary to develop new methods to identify MS plaques.Traditionally, scholars have proposed many computer vision methods [3][4][5] in identify abnormal brain diseases [6][7][8][9][10][11][12][13][14][15][16][17][18][19][20], including MS slice identification. For example, Zhou (2016) [21].used stationary wavelet entropy, combined with several machine learning methods. Zhan and Chen (2016) [22] used biorthogonal wavelet transform and logistic regression. Karaca (2017) [23] used convex combination of infinite kernels.Nevertheless, those methods are too complicated, and their models are difficult to train. In this study, we presented a novel and simple system, which was based on Haar wavelet transform, principle component, and logistic regression. Materials and MethodsThirty-four multiple sclerosis patients and thirty-three healthy subjects were enrolled from China local hospitals from 2013. Experienced radiologists were instructed to select the slice with MS plaques from the 34 patients, and select corresponding slices from 33 healthy subjects. In total, we selected 141 slices from MS patients and 148 slices from healthy controls.Our method contains three-stages. In the first stage, Haar wavelet [24] was chosen to transform the brain slices from spatial domain to wavelet domain [25][26][27][28][29][30][31]. The mother wavelet Haar wavelet λ(b) is defined with the form ofHere b denotes the spatial axis: either horizontal or vertical. The scaling function θ(b) is defined as:International Conference on Advances in Materials, Machinery, Electrical Engineering (AMMEE 2017) Copyright © 2017, the Authors. Published by Atlantis Press. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/). Advances in Engineering Research, volume 114We chose to use a two-level Haar wavelet transform. In the second stage, principle component was used to reduce the wavelet coefficients to only cover the threshold (THR) of total variances. We here set THR to 95%.In the third stage, logistic regression (LR) [32] was employed to be the classifier. The logistic regression is simpler and easier to train than multilayer perceptron [33][34][35][36], probabilis...

show abstract

The Differential Diagnosis of Multiple Sclerosis Using Convex Combination of Infinite Kernels

Cited by 14 publications

References 22 publications

Clustering Multiple Sclerosis Subgroups With Multifractal Methods and Self-Organizing Map Algorithm

Clustering Multiple Sclerosis Subgroups With Multifractal Methods and Self-Organizing Map Algorithm

A computer-aided diagnosis of multiple sclerosis based on mfVEP recordings

Multiple Sclerosis Slice Identification by Haar Wavelet Transform and Logistic Regression

Contact Info

Product

Resources

About