Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jul 8, 2026

Simultaneous Scalp Electroencephalography (EEG), Electromyography (EMG), and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

Simultaneous Scalp Electroencephalography (EEG), Electromyography (EMG), and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

SeLECTS-assisted identification based on structured EEG reports: Development and external validation of a multicenter

Lijun Li1, Lingxiang Ao2, Lei Li1

  • 1Neurophysiology Center, Kunming Children's Hospital & Children's Hospital of Kunming Medical University, Kunming, China.

Medicine
|July 7, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Pediatric primary ciliary dyskinesia with rare genetic variants: Synonymous RSPH4A and potential NFE2L2 modifier of DNAH9 phenotype - 2-case report.

Medicine·2026
Same author

Vitamin D concentrations and headache risk in children and adolescents aged 6 to 19 years: The mediating role of body mass index.

Medicine·2026
Same author

Letter by Xie et al Regarding Article, "High-Volume Physical Activity and Clinical Coronary Artery Disease Outcomes: Findings From the Cooper Center Longitudinal Study".

Circulation·2026
Same author

A case report of venous air embolism induced by bone marrow cavity irrigation.

Medicine·2026
Same author

Association of C-reactive protein-albumin-lymphocyte (CALLY) index with all-cause mortality in patients with type 2 diabetes: Results from the retrospective cohort study of NHANES 1999-2010.

Medicine·2025
Same author

The role of machine learning methods in assessing the risk of neonatal sepsis: A study of biochemical markers and genetic variants.

Medicine·2025

A new machine learning model accurately identifies benign epilepsy with centrotemporal spikes (SeLECTS) using routine EEG reports and lab tests. This explainable tool improves early diagnosis and clinical workflow across institutions.

Area of Science:

  • Neurology
  • Pediatric Epilepsy
  • Machine Learning in Medicine

Background:

  • Benign epilepsy with centrotemporal spikes (SeLECTS) is a common pediatric epilepsy.
  • Early diagnosis is hindered by inter-reader variability and differing EEG protocols.
  • Current models lack generalizability due to reliance on single-center raw EEG data.

Purpose of the Study:

  • To develop and validate an explainable, cross-institutional machine learning model for SeLECTS assistance.
  • To improve early identification and clinical workflow using structured EEG reports and routine laboratory tests.
  • To evaluate model performance, calibration, and clinical net benefit on internal and external datasets.

Main Methods:

  • A multicenter retrospective study involving 1067 internal and 94 external participants.
Keywords:
BECTLASSOSHAPXGBoostcalibration curvedecision curveexternal validation

Related Experiment Videos

Last Updated: Jul 8, 2026

Simultaneous Scalp Electroencephalography (EEG), Electromyography (EMG), and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

Simultaneous Scalp Electroencephalography (EEG), Electromyography (EMG), and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

  • Feature selection using Least Absolute Shrinkage and Selection Operator identified 13 key variables.
  • Ten common machine learning classifiers were trained and tested, with eXtreme Gradient Boosting selected as the best performer.
  • Model performance was assessed using AUROC, accuracy, sensitivity, and specificity, with calibration and decision curve analyses for clinical utility.

    • Main Results:

    • The eXtreme Gradient Boosting model achieved high performance: internal validation (AUROC 0.97, accuracy 0.91) and external validation (AUROC 0.96, accuracy 0.88).
    • External validation results showed only a slight decline from internal performance.
    • Calibration analysis confirmed close agreement between predicted and observed risks, and decision curve analysis indicated positive net clinical benefit.

    Conclusions:

    • A machine learning model utilizing 13 routinely available features demonstrates high discrimination, good calibration, and clinical utility for SeLECTS assistance.
    • The interpretable and lightweight model is suitable for scalable deployment across institutions as a decision support tool.
    • This approach enhances risk alerts, tiered review, and supports quality control in SeLECTS diagnosis.