Predicting Parkinson's disease using gradient boosting decision tree models with electroencephalography signals.

INTRODUCTION: Parkinson's disease (PD) is a neurodegenerative disorder with only symptomatic treatments currently available. Although correct, early diagnoses of PD are important, the existing diagnostic method based on pathologic examinations only has an accuracy of approximately 80.6%. Although electroencephalography (EEG)-based assistive technology has been introduced, it has been difficult to implement in practice due to the high computational complexity and low accuracy of the analysis methods. This study proposed a fast, accurate PD prediction method using the Hjorth parameter and the gradient boosting decision tree (GBDT) algorithm. METHOD: We used an open EEG dataset with 41 PD patients and 41 healthy controls (HCs); EEG signals were recorded from participants at the University of New Mexico (PD: 27 vs. HC: 27) and University of Iowa (PD: 14 vs. HC: 14). We explored the analytic time segment and frequency range in which the Hjorth parameter best represents the EEG characteristics of PD patients. RESULTS: Our best model (CatBoost-based) distinguished PD patients from controls with an accuracy of 89.3%, an area under the receiver operating characteristics curve (AUC) of 0.912, an F-score of 0.903, and an odds ratio of 115.5. These results showed that our models outperformed those of all other previous works and were even superior to previously known pathologic examination-based diagnoses with long-term follow-up (accuracy = 83.9%). CONCLUSION: The proposed methods are expected to be utilized as an effective method for improving the diagnosis of PD.

Development of the clinical assessment scale in autoimmune encephalitis.

OBJECTIVE: There is no scale for rating the severity of autoimmune encephalitis (AE). In this study, we aimed to develop a novel scale for rating severity in patients with diverse AE syndromes and to verify the reliability and validity of the developed scale. METHODS: The key items were generated by a panel of experts and selected according to content validity ratios. The developed scale was initially applied to 50 patients with AE (development cohort) to evaluate its acceptability, reproducibility, internal consistency, and construct validity. Then, the scale was applied to another independent cohort (validation cohort, n = 38). RESULTS: A new scale consisting of 9 items (seizure, memory dysfunction, psychiatric symptoms, consciousness, language problems, dyskinesia/dystonia, gait instability and ataxia, brainstem dysfunction, and weakness) was developed. Each item was assigned a value of up to 3 points. The total score could therefore range from 0 to 27. We named the scale the Clinical Assessment Scale in Autoimmune Encephalitis (CASE). The new scale showed excellent interobserver (intraclass correlation coefficient [ICC] = 0.97) and intraobserver (ICC = 0.96) reliability for total scores, was highly correlated with modified Rankin scale (r = 0.86, p < 0.001), and had acceptable internal consistency (Cronbach α = 0.88). Additionally, in the validation cohort, the scale showed high interobserver reliability (ICC = 0.99) and internal consistency (Cronbach α = 0.92). INTERPRETATION: CASE is a novel clinical scale for AE with a high level of clinimetric properties. It would be suitable for application in clinical practice and might help overcome the limitations of current outcome scales for AE. ANN NEUROL 2019;85:352-358.

Characteristics of patients with epilepsy who use a website providing healthcare information about epilepsy in South Korea.

We studied characteristics of epilepsy patients who use 'Epilia', a healthcare website for epilepsy patients, and the impact of 'Epilia' on patient satisfaction and attitude about epilepsy. A total of 153 epilepsy patients from 'Epilia' (online group) and 367 patients from epilepsy clinics (offline group) were recruited. A survey was done to investigate their socio-demographic, epilepsy-related, and psychological characteristics, as well as attitude alterations after beginning to use 'Epilia'. Factors affecting attitude were also identified in the online group. Online group had more severe and more frequent seizures, more adverse effects, more anxiety and depression, and worse quality of life than offline group. Online group patients who are more satisfied, who take two or more antiepileptic drugs, who visit to 'Epilia' more and longer had a greater positive attitude change than those without these characteristics. A healthcare website can contribute to self-management by providing useful information and education for epilepsy patients.

Assessment of language dominance by event-related oscillatory changes in an auditory language task: magnetoencephalography study.

The authors investigated the oscillatory changes induced by auditory language task to assess hemispheric dominance of language. Magnetoencephalography studies were conducted during word listening in 6 normal right-handed volunteers and 13 epilepsy patients who underwent Wada test. We carried out a time-frequency analysis of event-related desynchronization (ERD)/event-related synchronization (ERS) and intertrial coherence. We localized ERD/ERS on each subject's magnetic resonance images using beamformer. We compared ERD/ERS values between the left and right side of regions of interest in inferior frontal and superior temporal areas. We assessed the target frequency range that correlated best with the Wada test results. In all normal subjects, gamma ERD was lateralized to the left side in both the inferior frontal and superior temporal areas. In epilepsy patients, the concordance rate of gamma ERD and the Wada test results was 76.9% for the inferior frontal area and 69.2% for the superior temporal area. Gamma ERD can be considered as an indicator of language function, although it was not sufficient to replace the Wada test in the evaluation of epilepsy patients. The gamma ERD value of the inferior frontal area was more reliable for the assessment of language dominance compared with that obtained in the superior temporal area.

Adding or repositioning intracranial electrodes during presurgical assessment of neocortical epilepsy: electrographic seizure pattern and surgical outcome.

OBJECT: The aim of this study was to investigate changes in electroencephalography (EEG) patterns obtained from added or repositioned electrodes after those initially implanted had failed to indicate the true local ictal onset zone. The authors focused on the following matters: rationale for adding or repositioning electrodes, topographic and frequency characteristics of ictal onset before and after adding or repositioning electrodes, the effect of the procedures, and the relationship between changes in intracranial EEG onset patterns and surgical outcomes. METHODS: Of 183 patients with intracranial recordings, 18 experienced repositioning of existing or implanting of additional electrodes 7 or 10 days later. All patients underwent resection and were followed up for more than 1 year. In particular, the relationship between surgical outcome and distribution/frequency of intracranial seizure onset was analyzed. Results of noninvasive presurgical evaluations in patients who had undergone single and double invasive studies were also evaluated. By adding or repositioning electrodes, a new ictal onset zone was revealed in 13 patients. In another four, the second evaluation led to a change in defining the resection margin. Ictal onset in the partially sampled area, simultaneous or independent onset in two separate areas, and onset in the distal end of the electrode strip or grid were common reasons for failing to localize the ictal onset zone during the initial evaluation. Seven of 11 patients who were ultimately found to have a focal ictal onset zone on the second evaluation became seizure free after the operation. Only one of six patients with a regional ictal onset zone identified on the second evaluation became seizure free. There was no relationship between the frequency of the ictal rhythm and surgical outcome. Note, however, that surgical outcome was more favorable in patients who had undergone a single invasive study than in those who had undergone double invasive studies. The patients who needed a second evaluation had less localizing information and less concordant results on presurgical evaluations. When comparing nonlesional cases, surgical outcomes were not significantly different among patients with a single invasive study and those with double invasive studies. No additional morbidity or death occurred during the second study. CONCLUSIONS: The addition or reposition of intracranial electrodes with a short-term interval should be considered in selected patients. Spatial restriction of the ictal onset rhythm identified on repeated evaluation is the most important predictor of a good surgical outcome.