Identification of clusters in multifocal electrophysiology recordings to maximize discriminant capacity (patients vs. control subjects) (original) (raw)
Related papers
A computer-aided diagnosis of multiple sclerosis based on mfVEP recordings
PLOS ONE, 2019
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.
IEEE Transactions on Information Technology in Biomedicine, 2000
This paper describes the application of a novel unsupervised pattern recognition system to the classification of the Visual Evoked Potentials (VEP's) of normal and multiple sclerosis (MS) patients. The method combines a traditional statistical feature extractor with a fuzzy clustering method, all implemented in a parallel neural network architecture. The optimization routine, ALOPEX, is used to train the network while decreasing the likelihood of local solutions. The unsupervised system includes a feature extraction and clustering module, trained by the optimization routine ALOPEX. Through maximization of the output variance of each node, and an architecture which excludes redundancy, the feature extraction network retains the most significant Karhunen-Loève expansion vectors. The clustering module uses a modification to the Fuzzy-Means (FCM) clustering algorithms, where ALOPEX adjusts a set of cluster centers to minimize an objective error function. The result combines the power of the FCM algorithms with the advantage of a more global solution from ALOPEX. The new pattern recognition system is used to cluster the VEP's of 13 normal and 12 MS subjects. The classification with this technique can, without supervision, separate the patient population into two groups which largely correspond to the MS and control subject groups. A suitable threshold can be chosen so that the recognizer chooses no false negatives. The use of multiple stimulation patterns appears to improve the reliability of the decision. The reasoning of most neural networks in their decision making cannot easily be extracted upon the completion of training. However, due to the linearity of the network nodes, the cluster prototypes of this unsupervised system can be reconstructed to illustrate the reasoning of the system. In this application, this analysis hints at the usefulness of previously unused portions of the VEP in detecting MS. It also indicates a possible use of the system as a training aide.
2016
In multiple sclerosis (MS), the combination of visual, somatosensory and motor evoked potentials (EP) has been shown to be highly correlated with the Expanded Disability Severity Scale (EDSS) and to predict the disease course. In the present study, we explored whether the significance of the visual EP (VEP) can be improved with multichannel recordings (204 electrodes) and topographic analysis (tVEP). VEPs were analyzed in 83 MS patients (median EDSS 2.0; 52 % with history of optic neuritis; hON) and 47 healthy controls (HC). TVEP components were automatically defined on the basis of spatial similarity between the scalp potential fields (topographic maps) of single subjects ’ VEPs and reference maps generated from HC. Non-ambiguous measures of latency, amplitude and configuration were derived from the maps reflecting the P100 component. TVEP was compared to conventional analysis (cVEP) with respect to reliability in HC, validity using descriptors of logistic regression models, and se...
Sensors
As multiple sclerosis (MS) usually affects the visual pathway, visual electrophysiological tests can be used to diagnose it. The objective of this paper is to research methods for processing multifocal electroretinogram (mfERG) recordings to improve the capacity to diagnose MS. MfERG recordings from 15 early-stage MS patients without a history of optic neuritis and from 6 control subjects were examined. A normative database was built from the control subject signals. The mfERG recordings were filtered using empirical mode decomposition (EMD). The correlation with the signals in a normative database was used as the classification feature. Using EMD-based filtering and performance correlation, the mean area under the curve (AUC) value was 0.90. The greatest discriminant capacity was obtained in ring 4 and in the inferior nasal quadrant (AUC values of 0.96 and 0.94, respectively). Our results suggest that the combination of filtering mfERG recordings using EMD and calculating the corre...
Detection of steady-state visual evoked potentials based on the multisignal classification algorithm
2007
This paper presents a new method for detection of steady-sate visual evoked potential (SSVEP) in a 26-class brain-computer interface (BCI) using principal component analysis (PCA). PCA is used to decompose the multi-channel EEG signals into the components which are orthogonal. After PCA processing, the principal components (PCs) can be grouped into components related to SSVEPs and components related to brain activities. A major issue in using PCA to detect the SSVEP is the selection of proper components. In this work, we use high-order statistics to automatically identify the SSVEP components and AR power spectra to detect the frequency of SSVEP. The results of experiments on three subjects and each subject with 8 experiment sessions show that an average detection accuracy between 76.4% and 91.8% can be achieved.
Investigative Ophthalmology & Visual Science, 2007
Multifocal visual evoked potentials provide a topographic measure of visual response amplitude and latency. The objective of this study was to evaluate the sensitivity and specificity of the multifocal visual evoked potential technique in detecting visual abnormalities in patients with multiple sclerosis. Multifocal visual evoked potentials were recorded from 74 patients with multiple sclerosis with history of optic neuritis (MS-ON, n = 74 eyes) or without (MS-noON , n = 71 eyes), and 50 normal subjects (controls, n = 100 eyes) using a 60-sector pattern reversal dartboard stimulus (VERIS). Amplitude and latency for each sector were compared with normative data and assigned probabilities. Size and location of clusters of adjacent abnormal sectors (p < 0.05) were examined. Mean response amplitudes were (AE SE) 0.39 AE 0.02, 0.53 AE 0.02, and 0.60 AE 0.01 for MS-ON, MS-noON , and control groups, respectively, with significant differences between all groups (p < 0.0001). Mean latencies (ms; AESE relative to normative data) were 12.7 AE 1.3 (MS-ON), 4.3 AE 1.1 (MS-noON), and 0.3 AE 0.4 (controls); group differences again significant (p < 0.0001). Half the MS-ON eyes had clusters larger than five sectors compared with 13% in MS-noON and 2% in controls. Abnormal sectors were distributed diffusely, although the largest cluster was smaller than 15 sectors in two-thirds of MS-ON eyes. Cluster criteria combining amplitude and latency showed an area of 0.96 under the receiver operating characteristic curve, yielding a criterion with 91% sensitivity and 95% specificity. We conclude that the multifocal visual evoked potential provides high sensitivity and specificity in detecting abnormalities in visual function in multiple sclerosis patients.
Improved Characterization of Visual Evoked Potentials in Multiple Sclerosis by Topographic Analysis
Brain Topography, 2014
In multiple sclerosis (MS), the combination of visual, somatosensory and motor evoked potentials (EP) has been shown to be highly correlated with the Expanded Disability Severity Scale (EDSS) and to predict the disease course. In the present study, we explored whether the significance of the visual EP (VEP) can be improved with multichannel recordings (204 electrodes) and topographic analysis (tVEP). VEPs were analyzed in 83 MS patients (median EDSS 2.0; 52 % with history of optic neuritis; hON) and 47 healthy controls (HC). TVEP components were automatically defined on the basis of spatial similarity between the scalp potential fields (topographic maps) of single subjects' VEPs and reference maps generated from HC. Non-ambiguous measures of latency, amplitude and configuration were derived from the maps reflecting the P100 component. TVEP was compared to conventional analysis (cVEP) with respect to reliability in HC, validity using descriptors of logistic regression models, and sensitivity derived from receiver operating characteristics curves. In tVEP, reliability tended to be higher for measurement of amplitude (p = 0.06). Regression models on diagnosis (MS vs. HC) and hON were more favorable using tVEP-versus cVEP-predictors. Sensitivity was increased in tVEP versus cVEP: 72 % versus 60 % for diagnosis, and 88 % versus 77 % for hON. The advantage of tVEP was most pronounced in pathological VEPs, in which cVEPs were often ambiguous. TVEP is a reliable, valid, and sensitive method of objectively quantifying pathological VEP in particular. In combination with other EP modalities, tVEP may improve the monitoring of disease course in MS.