EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Unsupervised Learning-Based Classification of Breast Cancer Using Gaussian Mixture Model

Hussein Alkattan, Salam Abdulkhaleq Noaman, Abrar Ryadh, Mostafa Abotaleb, H.K. Al-Mahdawi, Pushan Kumar Dutta

SHIFAA · 2025

DOI: 10.70470/shifaa/2024/100 ↗

Kurz gefasst

Fachartikel in der Zeitschrift SHIFAA, 2025. Verfasst von Hussein Alkattan und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study reports the use of unsupervised learning approaches for breast cancer classification on the GBSG dataset. Application of GMMs to segment the data into four clusters sought to identify latent group structures with the patients based upon the absence of class labels. Application of PCA was also used for dimensionality reduction for visualization, capturing over 95% of the variance in the first two components. The distribution of clusters indicated that Cluster 0 and Cluster 1 had 223 (≈41.6%) and 303 (≈56.6%) patients respectively, whereas Clusters 2 and 3 had merely 9 (≈1.7%) and 13 (≈2.4%) patients. The adjusted rand index (ARI) between the GMM clusters and the true cancer status was −0.0033, reflecting minimal concordance. Moreover, a correlation heatmap indicated high positive correlations among various clinical features, enabling the detection of redundant variables. Although unsupervised clustering did not concur strongly with labeled classes, it provided insightful information concerning data distribution and potential feature interactions. These findings support the implementation of GMM-based unsupervised learning in preprocessing pipelines or semi-supervised classification frameworks for predicting breast cancer.

Zitieren

APA

Alkattan, H., Noaman, S. A., Ryadh, A., Abotaleb, M., Al-Mahdawi, H., & Dutta, P. K. (2025). Unsupervised Learning-Based Classification of Breast Cancer Using Gaussian Mixture Model. SHIFAA. https://doi.org/10.70470/shifaa/2024/100

Vancouver

Alkattan H, Noaman SA, Ryadh A, Abotaleb M, Al-Mahdawi H, Dutta PK. Unsupervised Learning-Based Classification of Breast Cancer Using Gaussian Mixture Model. SHIFAA. 2025. doi:10.70470/shifaa/2024/100

BibTeX
@article{Alkattan2025,
  author = {Alkattan, Hussein and Noaman, Salam Abdulkhaleq and Ryadh, Abrar and Abotaleb, Mostafa and Al-Mahdawi, H.K. and Dutta, Pushan Kumar},
  title = {Unsupervised Learning-Based Classification of Breast Cancer Using Gaussian Mixture Model},
  journal = {SHIFAA},
  year = {2025},
  doi = {10.70470/shifaa/2024/100}
}
Kostenlos starten →