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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Discovery of Anti-Japanese Encephalitis Compounds: Based on Natural Compound Library, Bioinformatics, Network Pharmacology and Experimental Validation.

Current pharmaceutical design·2026
Same author

Differences in Posterior Ocular Morphology Following Red Light Therapy Versus 0.01% Atropine for Myopia Control: A Randomized Controlled Trial Analysis.

Investigative ophthalmology & visual science·2026
Same author

<b>A new species of <i>Casinaria</i> (Hymenoptera: Ichneumonidae) parasitizing on <i>Aporia hippia</i> (Lepidoptera: Pieridae) from China, and its complete mitochondrial genome</b>.

Zootaxa·2026
Same author

Author Correction: Natural selection and genetic diversity maintenance in a parasitic wasp during continuous biological control application.

Nature communications·2026
Same author

DGCD-3D: Difference-guided conditional diffusion model for low-field 3D MRI enhancement to assist stroke assessment.

Medical image analysis·2026
Same author

Assessing the Therapeutic Efficacy of Xihuang Pill in Lung Adenocarcinoma by Single-Cell Raman Spectroscopy.

Analytical chemistry·2026

Related Experiment Video

Updated: Sep 7, 2025

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

7.6K

A multi-classification model for non-small cell lung cancer subtypes based on independent subtask learning.

Jinkai Li1,2, Fan Song1, Peng Zhang1

  • 1Beijing Advanced Innovation Center for Biomedical Engineering, School of Biological Science and Medical Engineering, Beihang University, Beijing, China.

Medical Physics
|June 17, 2022
PubMed
Summary

A novel Independent Subtask Learning (ISTL) method effectively classifies non-small cell lung cancer (NSCLC) subtypes. This approach significantly outperforms traditional methods, offering improved accuracy and interpretability for clinical computer-aided detection.

Keywords:
machine learningmulti-classificationnon-small cell lung cancerone-vs-one (OVO)radiomics

More Related Videos

Unilateral Lung Volume Analysis Using Micro-CT for Enhanced Assessment of Pulmonary Fibrosis in Preclinical Models
03:38

Unilateral Lung Volume Analysis Using Micro-CT for Enhanced Assessment of Pulmonary Fibrosis in Preclinical Models

Published on: June 20, 2025

321
Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
07:13

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Published on: April 18, 2025

192

Related Experiment Videos

Last Updated: Sep 7, 2025

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

7.6K
Unilateral Lung Volume Analysis Using Micro-CT for Enhanced Assessment of Pulmonary Fibrosis in Preclinical Models
03:38

Unilateral Lung Volume Analysis Using Micro-CT for Enhanced Assessment of Pulmonary Fibrosis in Preclinical Models

Published on: June 20, 2025

321
Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
07:13

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Published on: April 18, 2025

192

Area of Science:

  • Medical Imaging Analysis
  • Machine Learning in Oncology
  • Computational Pathology

Background:

  • Non-small cell lung cancer (NSCLC) classification into subtypes like adenocarcinoma (ADC), squamous cell carcinoma (SCC), large cell carcinoma (LCC), and not otherwise specified (NOS) is critical for treatment decisions.
  • Existing research faces challenges in NSCLC multi-classification due to data imbalance, feature space unification difficulties, and complex decision boundaries.
  • Traditional machine learning methods, such as one-vs-one (OVO), struggle to effectively address these challenges in NSCLC subtyping.

Purpose of the Study:

  • To develop a novel Independent Subtask Learning (ISTL) method for improved multi-classification of NSCLC subtypes.
  • To address the limitations of traditional methods in handling data imbalance and complex decision boundaries in NSCLC classification.
  • To enhance the interpretability of machine learning models for distinguishing between different NSCLC subtypes.

Main Methods:

  • Proposed a novel Independent Subtask Learning (ISTL) method incorporating four strategies: independent data expansion, independent feature selection (IFS), independent model construction, and a hybrid voting strategy (majority voting with Bayesian prior).
  • Utilized a dataset of 1036 CT scans from eight international databases, featuring diverse data acquisition conditions and an imbalanced distribution of NSCLC subtypes (ADC:SCC:LCC:NOS = 600:268:105:63).
  • Compared the ISTL method against traditional OVO-support vector machine and OVO-random forest models.

Main Results:

  • The ISTL method achieved a significantly higher accuracy of 0.812 on the independent test cohort compared to OVO-support vector machine (0.691) and OVO-random forest (0.710).
  • Independent feature selection within the ISTL method revealed distinct feature sets for different binary tasks, enhancing model interpretability for NSCLC subtype differentiation.
  • Auxiliary experiments confirmed the effectiveness of each of the four proposed strategies within the ISTL framework.

Conclusions:

  • The ISTL method demonstrates superior performance and interpretability for the multi-classification of NSCLC subtypes.
  • This approach holds significant potential for advancing clinical computer-aided detection systems for lung cancer.
  • The ISTL methodology offers a promising framework applicable to a wide range of multi-classification problems beyond NSCLC.