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 Videos

Automatic embolus detection by a neural network.

V Kemény1, D W Droste, S Hermes

  • 1Department of Neurology, University of Münster, Germany. kemeny@uzu.net

Stroke
|April 3, 1999
PubMed
Summary
This summary is machine-generated.

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

[Structures of endovascular recanalization in Germany : Cross-sectional survey of certified stroke units 2016-2018].

Der Nervenarzt·2020
Same author

[Ultrasound diagnostics of extracranial cerebral arteries].

Der Radiologe·2019
Same author

Angioplasty in asymptomatic carotid artery stenosis vs. endarterectomy compared to best medical treatment: One-year interim results of SPACE-2.

International journal of stroke : official journal of the International Stroke Society·2019
Same author

A Second Family with Autosomal Dominant Burning Feet Syndrome.

Annals of the New York Academy of Sciences·2017
Same author

Hereditary Neuralgic Amyotrophy: Mutation Analysis of Candidate Genes.

Annals of the New York Academy of Sciences·2017
Same author

Middle cerebral artery blood flow velocity, end-tidal pC02 and blood pressure in patients with obstructive sleep apnea and in healthy subjects during continuous positive airway pressure breathing.

Neurological research·2016

Automated embolus detection using neural networks shows promise for cerebrovascular disease diagnosis. However, strong artifacts can cause misdiagnosis, requiring expert verification for accurate microembolic signal identification.

Area of Science:

  • Neuroscience
  • Biomedical Engineering
  • Medical Imaging

Background:

  • Transcranial Doppler ultrasound is crucial for identifying embolic sources in cerebrovascular diseases.
  • Automated systems aim to improve objectivity and efficiency in long-term recordings.
  • Evaluating automated embolus detection critically is essential for clinical application.

Purpose of the Study:

  • To assess the performance of a trained neural network for automated embolus detection.
  • To evaluate the critical conditions affecting the accuracy of the EMBotec V5.1 One system.

Main Methods:

  • Simultaneous middle or posterior cerebral artery recordings in 11 volunteers and 11 patients.
  • Generation of 1342 artifacts in normal subjects for false-positive testing.

Related Experiment Videos

  • Offline detection by a blinded investigator and a trained neural network.

    • Main Results:

    • The neural network rejected 85% of artifacts but misclassified 122 artifact signals as emboli.
    • Sensitivity was 73.4% and positive predictive value was 56.7% in microembolus-positive patients.
    • Artifact signals had significantly higher spectral power than true microembolic signals.

    Conclusions:

    • The neural network shows potential for automated embolus detection.
    • Misdiagnosis occurs with extreme signal qualities, particularly strong artifacts.
    • Verification of microembolic signals by experienced investigators remains mandatory.