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

Bayesian fluorescence in situ hybridisation signal classification.

Boaz Lerner1

  • 1Pattern Analysis & Machine Learning Lab, Department of Electrical & Computer Engineering, Ben-Gurion University, Beer-Sheva, Israel. boaz@ee.bgu.ac.il

Artificial Intelligence in Medicine
|April 15, 2004
PubMed
Summary

Accurate classification of fluorescence in situ hybridisation (FISH) signals is crucial for detecting genetic abnormalities. This study compares a neural network (NN) with a naive Bayesian classifier (NBC), offering a trade-off between NN performance and NBC implementation simplicity.

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

Identifying diagnostic markers in the health records of prediagnostic amyotrophic lateral sclerosis patients.

Amyotrophic lateral sclerosis & frontotemporal degeneration·2025
Same author

Machine learning for predicting Crohn's disease from routine blood tests years before diagnosis: results from the epi-IIRN cohort.

Journal of Crohn's & colitis·2025
Same author

Lower prevalence of metabolic syndrome and its components in the prediagnostic phase of Parkinson's disease.

NPJ Parkinson's disease·2025
Same author

Early Detection of Pulmonary Embolism in a General Patient Population Immediately Upon Hospital Admission Using Machine Learning to Identify New, Unidentified Risk Factors: Model Development Study.

Journal of medical Internet research·2024
Same author

Medical concept embedding of real-valued electronic health records with application to inflammatory bowel disease.

Artificial intelligence in medicine·2023
Same author

Levodopa responsiveness in Parkinson's disease: harnessing real-life experience with machine-learning analysis.

Journal of neural transmission (Vienna, Austria : 1996)·2022

Area of Science:

  • Biomedical signal processing
  • Computational biology
  • Genetics

Background:

  • Accurate classification of fluorescence in situ hybridisation (FISH) signals is vital for identifying genetic abnormalities.
  • Previous systems utilized neural networks (NN) for high-accuracy FISH signal classification.
  • The NN system employed independent features, prompting exploration of alternative classifiers.

Purpose of the Study:

  • To evaluate the naive Bayesian classifier (NBC) as an alternative to NN for FISH signal classification.
  • To compare the performance of NBC with different density estimation methods (SGE, GMM, KDE).
  • To assess the trade-off between NN performance and NBC implementation simplicity.

Main Methods:

  • Implemented a naive Bayesian classifier (NBC) utilizing probability density estimation.

Related Experiment Videos

  • Evaluated density estimation using single Gaussian estimation (SGE), Gaussian mixture models (GMM), and kernel density estimation (KDE).
  • Compared NBC performance against a previously established neural network (NN) system.

    • Main Results:

    • NBC offers a simpler implementation compared to the resource-intensive NN.
    • Gaussian mixture models (GMM) showed good performance for low-dimensional data but struggled with dependent features.
    • Single Gaussian estimation (SGE) and NN provided inferior and superior performance, respectively, compared to GMM and KDE in certain scenarios.
    • The system supports both NN and NBC, allowing a balance between accuracy and implementation ease.

    Conclusions:

    • The naive Bayesian classifier (NBC) presents a viable, simpler alternative for FISH signal classification, particularly when implementation ease is prioritized.
    • The choice between NN and NBC depends on the specific requirements for accuracy versus computational resources and complexity.
    • This research provides a flexible system for genetic abnormality detection through optimized FISH signal classification.