A systematic review of the application of machine-learning algorithms in multiple sclerosis

INTRODUCTION

The applications of artificial intelligence, and in particular automatic learning or "machine learning" (ML), constitute both a challenge and a great opportunity in numerous scientific, technical, and clinical disciplines. Specific applications in the study of multiple sclerosis (MS) have been no exception, and constitute an area of increasing interest in recent years.

OBJECTIVE

We present a systematic review of the application of ML algorithms in MS.

MATERIALS AND METHODS

We used the PubMed search engine, which allows free access to the MEDLINE medical database, to identify studies including the keywords "machine learning" and "multiple sclerosis." We excluded review articles, studies written in languages other than English or Spanish, and studies that were mainly technical and did not specifically apply to MS. The final selection included 76 articles, and 38 were rejected.

CONCLUSIONS

After the review process, we established 4 main applications of ML in MS: 1) classifying MS subtypes; 2) distinguishing patients with MS from healthy controls and individuals with other diseases; 3) predicting progression and response to therapeutic interventions; and 4) other applications. Results found to date have shown that ML algorithms may offer great support for health professionals both in clinical settings and in research into MS.

Attentional networks in neurodegenerative diseases: anatomical and functional evidence from the Attention Network Test

INTRODUCTION

Understanding alterations to brain anatomy and cognitive function associated with neurodegenerative diseases remains a challenge for neuroscience today. In experimental neuroscience, several computerised tests have been developed to contribute to our understanding of neural networks involved in cognition. The Attention Network Test (ANT) enables us to measure the activity of 3 attentional networks (alertness, orienting, and executive function).

OBJECTIVES

The main aim of this review is to describe all the anatomical and functional alterations found in diverse neurological diseases using the ANT.

MATERIAL AND METHODS

We collected studies published since 2010 in the PubMed database that employed the ANT in different neurological diseases. Thirty-two articles were obtained, addressing multiple sclerosis, epilepsy, and Parkinson's disease, among other disorders.

CONCLUSIONS

Some of the anatomical structures proposed in the 3 attentional networks model were confirmed. The most relevant structures in the alertness network are the prefrontal cortex, parietal region, thalamus, and cerebellum. The thalamus is also relevant in the orienting network, together with posterior parietal regions. The executive network does not depend exclusively on the prefrontal cortex and anterior cingulate cortex, but also involves such subcortical structures as the basal ganglia and cerebellum and their projections towards the entire cortex.

A systematic review of the application of machine-learning algorithms in multiple sclerosis

INTRODUCTION

The applications of artificial intelligence, and in particular automatic learning or "machine learning" (ML), constitute both a challenge and a great opportunity in numerous scientific, technical, and clinical disciplines. Specific applications in the study of multiple sclerosis (MS) have been no exception, and constitute an area of increasing interest in recent years.

OBJECTIVE

We present a systematic review of the application of ML algorithms in MS.

MATERIALS AND METHODS

We used the PubMed search engine, which allows free access to the MEDLINE medical database, to identify studies including the keywords "machine learning" and "multiple sclerosis." We excluded review articles, studies written in languages other than English or Spanish, and studies that were mainly technical and did not specifically apply to MS. The final selection included 76 articles, and 38 were rejected.

CONCLUSIONS

After the review process, we established 4 main applications of ML in MS: 1) classifying MS subtypes; 2) distinguishing patients with MS from healthy controls and individuals with other diseases; 3) predicting progression and response to therapeutic interventions; and 4) other applications. Results found to date have shown that ML algorithms may offer great support for health professionals both in clinical settings and in research into MS.

Attentional networks in neurodegenerative diseases: anatomical and functional evidence from the Attention Network Test

INTRODUCTION

Understanding alterations to brain anatomy and cognitive function associated with neurodegenerative diseases remains a challenge for neuroscience today. In experimental neuroscience, several computerised tests have been developed to contribute to our understanding of neural networks involved in cognition. The Attention Network Test (ANT) enables us to measure the activity of 3 attentional networks (alertness, orienting, and executive function).

OBJECTIVES

The main aim of this review is to describe all the anatomical and functional alterations found in diverse neurological diseases using the ANT.

MATERIAL AND METHODS

We collected studies published since 2010 in the PubMed database that employed the ANT in different neurological diseases. Thirty-two articles were obtained, addressing multiple sclerosis, epilepsy, and Parkinson's disease, among other disorders.

CONCLUSIONS

Some of the anatomical structures proposed in the 3 attentional networks model were confirmed. The most relevant structures in the alertness network are the prefrontal cortex, parietal region, thalamus, and cerebellum. The thalamus is also relevant in the orienting network, together with posterior parietal regions. The executive network does not depend exclusively on the prefrontal cortex and anterior cingulate cortex, but also involves such subcortical structures as the basal ganglia and cerebellum and their projections towards the entire cortex.

EEG study on affective valence elicited by novel and familiar pictures using ERD/ERS and SVM-RFE

EEG signals have been widely explored in emotional processing analyses, both in time and frequency domains. However, in such studies, habituation phenomenon is barely considered in the discrimination of different emotional responses. In this work, spectral features of the event-related potentials (ERPs) are studied by means of event-related desynchronization/synchronization computation. In order to determine the most relevant ERP features for distinguishing how positive and negative affective valences are processed within the brain, support vector machine-recursive feature elimination is employed. The proposed approach was applied for investigating in which way the familiarity of stimuli affects the affective valence processing as well as which frequency bands and scalp regions are more involved in this process. In a group composed of young adult women, results prove that parietooccipital region and theta band are especially involved in the processing of novelty in emotional stimuli. Furthermore, the proposed method has shown to perform successfully using a moderated number of trials.

Individual EEG differences in affective valence processing in women with low and high neuroticism

OBJECTIVE

In this study, individual differences in brain electrophysiology during positive and negative affective valence processing in women with different neuroticism scores are quantified.

METHODS

Twenty-six women scoring high and low on neuroticism participated on this experiment. A support vector machine (SVM)-based classifier was applied on the EEG single trials elicited by high arousal pictures with negative and positive valence scores. Based on the accuracy values obtained from subject identification tasks, the most distinguishing EEG channels among participants were detected, pointing which scalp regions show more distinct patterns.

RESULTS

Significant differences were obtained, in the EEG heterogeneity between positive and negative valence stimuli, yielding higher accuracy in subject identification using negative pictures. Regarding the topographical analysis, significantly higher accuracy values were reached in occipital areas and in the right hemisphere (p < 0.001).

CONCLUSIONS

Mainly, individual differences in EEG can be located in parietooccipital regions. These differences are likely to be due to the different reactivity and coping strategies to unpleasant stimuli in individuals with high neuroticism. In addition, the right hemisphere shows a greater individual specificity.

SIGNIFICANCE

An SVM-based classifier asserts the individual specificity and its topographical differences in electrophysiological activity for women with high neuroticism compared to low neuroticism.

[Cognitive evoked potentials in remitting-relapsing and benign forms of multiple sclerosis]

INTRODUCTION

The assessment of the cognitive impairment in the multiple sclerosis disease is one of the most relevant challenges nowadays. An essential objective is to obtain diverse approaches that allow valuing objectively this impairment and its correlation with physiological variables.

AIM

To analyze the possible modulations in physiological components of the information processing (ERPs) related with an attentional deficit in diverse classes of patients with multiple sclerosis.

SUBJECTS AND METHODS

17 patients with remitting-relapsing multiple sclerosis, 9 patients with benign multiple sclerosis and 19 healthy subjects participated in the study. Behavioral performance in a visuo-spatial task (Posner paradigm) and later, an auditory oddball test was carried out where electroencephalography signal was registered to obtain ERPs. Correlation analyses were calculated between patient variables (EDSS or disease duration) and psychophysiological variables.

RESULTS

A delay in the reaction time during the development of the Posner task and a delay in the latency of the component P3 during the realization of the oddball task were found in both groups of patient with multiple sclerosis.

CONCLUSION

The data obtained in this experiment confirm the presence of attentional impairment in patients with diverse forms of multiple sclerosis. The exclusive modulation of the P3 latency component suggests that the impairment in these patients, at the beginning of the disease, is localized at a central level of cognitive processing and is product of the demyelinating process.

Abnormal ERPs and high frequency bands power in multiple sclerosis

Event-related potentials (ERPs) and power spectral density (PSD) were registered during an auditory-oddball paradigm in 11 MS patients. These patients showed a decrease in the amplitude of P2 and N2 components and a delayed P3 latency compared to control subjects suggesting that the attentional orienting mechanism in the auditory modality is affected in MS. The PSD analysis showed that MS patients exhibited an increased power in beta and gamma bands. The combined analysis of frequency and time domain suggested diverse phenomena that occurred in the MS patient group related with the EEG background or the motivational status.

[Gender-specific post-stimulus modulation in the alpha band during visual-spatial attention]

INTRODUCTION

Previous studies of the spontaneous activity of the EEG indicate that women present bigger absolute spectral power in the alpha band than men, but few studies have been done during cognitive activity.

AIM

To check possible gender differences in the EEG frequency domain during a visuo-spatial attentional task to understand better the temporary dynamics of alpha.

SUBJECTS AND METHODS

A classic oddball paradigm was used whose stimuli (90% standard-10% objectives) were complex. In each block (10 with 100 trials) it was requested to the subjects (20, 10 of each gender) to attend to one of the visual fields (left and right, with random stimuli and identical probability). Power spectral density values were calculated applying the fast Fourier transform in the alpha band (9-11 Hz) in two overlapped windows (sliding windows): of 0-512 ms and of 412-924 ms.

RESULTS

The normalized data were analyzed by means of ANOVAs, with the following factors: gender, attention, hemisphere, visual field and electrodes.

CONCLUSIONS

Different modulations were obtained attending to the gender variable: men showed smaller alpha or bigger suppression of the rhythm in posterior electrodes and women exhibited smaller power in anterior electrodes.