Calibrating Deep Learning Classifiers for Patient-Independent Electroencephalogram Seizure Forecasting.

The recent scientific literature abounds in proposals of seizure forecasting methods that exploit machine learning to automatically analyze electroencephalogram (EEG) signals. Deep learning algorithms seem to achieve a particularly remarkable performance, suggesting that the implementation of clinical devices for seizure prediction might be within reach. However, most of the research evaluated the robustness of automatic forecasting methods through randomized cross-validation techniques, while clinical applications require much more stringent validation based on patient-independent testing. In this study, we show that automatic seizure forecasting can be performed, to some extent, even on independent patients who have never been seen during the training phase, thanks to the implementation of a simple calibration pipeline that can fine-tune deep learning models, even on a single epileptic event recorded from a new patient. We evaluate our calibration procedure using two datasets containing EEG signals recorded from a large cohort of epileptic subjects, demonstrating that the forecast accuracy of deep learning methods can increase on average by more than 20%, and that performance improves systematically in all independent patients. We further show that our calibration procedure works best for deep learning models, but can also be successfully applied to machine learning algorithms based on engineered signal features. Although our method still requires at least one epileptic event per patient to calibrate the forecasting model, we conclude that focusing on realistic validation methods allows to more reliably compare different machine learning approaches for seizure prediction, enabling the implementation of robust and effective forecasting systems that can be used in daily healthcare practice.

Investigating the Psychometric Properties of Sex Addiction Screening Test-revised for Persian Population.

Introduction: Sexual addiction is known as a disorder that afflicts a person with difficulty in controlling or delaying sexual behaviors. To prevent social, physical, and psychological consequences, validated screening tests are needed to diagnose this disorder. One of these tests is established by Carnes with the name of sexual addiction screening test-revised (SAST-R). In this study, SAST-R has been translated and verified in the Persian language. Methods: The original screening test was translated into the Persian language and also back-translated for matching by two separate expert teams. The data was collected through an online survey of 1268 participants who were in the age range of 18 to 65 years (Mean±SD 29.44±6.90), and 56.1% and 43.9% of the population were women and men, respectively. Three questionnaires, including the SAST-R, the hypersexual behavior consequences scale, and the Connor-Davidson resilience scale as the principal, convergent, and divergent tests were administered to the participants. Results: The reliability of the test's internal consistency (Cronbach α=0.883), split-half (Cronbach α=0.779), and Guttman (lambda coefficients were between 0.773 to 0.883) tests were used. In addition, 4 methods of content validity (sexual hyperactivity specialist approved), convergent structure validity (P<0.001, R=0.731), the validity of divergent structure (P<0.09, R=-0.132), and factor validity (comparative fit index=0.884, goodness of fit index=0.873, root mean square error of approximation=0.047) were measured and confirmed the validity of the test. Conclusion: The Persian version of SAST-R is a reliable preclinical tool to assess the severity of sexual desire in patients. Highlights: The Persian version of sexual addiction screening test-revised (SAST-R) serves as a dependable pre-clinical instrument for evaluating the intensity of sexual desire in patients.Expose various subsets of the original questionnaire that explore different aspects, such as understanding sexual orientation.Proposed cutting-off scores as the guidelines for clinicians to distinguish various aspects of the original questionnaire. Plain Language Summary: This study translated and validated the sexual addiction screening test-revised (SAST-R) in Persian. The study involved 1,268 participants who completed the translated test and other questionnaires. Results showed that the Persian version of the SAST-R is a reliable tool for assessing the severity of sexual desire in patients. It demonstrated good internal consistency, reliability, and validity. This validated screening test is important for diagnosing sexual addiction in Persian-speaking individuals, enabling healthcare professionals to identify and support those struggling with the disorder.

Effect of Transcranial Direct Current Stimulation on Dorsolateral Prefrontal Cortex to Reduce the Symptoms of the Obsessive-Compulsive Disorder.

INTRODUCTION: Obsessive-Compulsive Disorder (OCD) is one of the most common debilitating mental disorders with a prevalence rate of 2% to 3% in the general population. Previous studies have indicated abnormalities in the dorsolateral prefrontal cortex (DLPFC) of OCD patients; thus, we decided to use transcranial Direct Current Stimulation (tDCS) to decline these patients' symptoms. METHODS: A total of 24 patients with OCD participated in this study with the hope of improvement after the application of tDCS. The subjects were randomly assigned to three groups of Sham, Right DLPFC, and Left DLPFC. tDCS was applied for five consecutive days and in each session, patients were subjected to 2 mA current flow for two 15 minutes followed by a 10-minute rest in between (every session lasted for 40 minutes). RESULTS: Subsequently, the changes in obsessive-compulsive level and cognitive functions were evaluated via Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) and Depression, Anxiety, and Stress Scale 21 (DASS-21) by comparing the results before (pre-test) and after (post-test) tDCS treatment. CONCLUSION: Ultimately, the scores of the Yale-Brown scale in the Left DLPFC group showed significant changes after treatment with tDCS (mean difference compared to the sham group: -6.18 and P≤0.05). Hereupon, this study demonstrated that transcranial direct current stimulation may cause improvements in symptoms of OCD.