ANALYSIS OF MACHINE LEARNING ALGORITHMS APPLIED TO BREAST CANCER DIAGNOSIS

Ana Beatriz Miranda Valentin; Glaucia Maria Bressan; Elisangela Ap. da Silva Lizzi; Fabricio Martins Lopes

All Papers

Submission

Paper Title

ANALYSIS OF MACHINE LEARNING ALGORITHMS APPLIED TO BREAST CANCER DIAGNOSIS

Authors

Ana Beatriz Miranda Valentin
Glaucia Maria Bressan
Elisangela Ap. da Silva Lizzi
Fabricio Martins Lopes

Modality

Poster

Subject area

Database and Software Development

Publishing Date

21/11/2024

Country of Publishing

Brazil | Brasil

Language of Publishing

Inglês

Paper Page

https://www.even3.com.br/anais/xmeeting-2024/822227-analysis-of-machine-learning-algorithms-applied-to-breast-cancer-diagnosis

ISBN

978-65-272-0843-3

Keywords

Breast cancer, tumor classification, feature selection, efficient diagnosis

Summary

Understanding the key features of breast tumors that lead to their identification as benign or malignant is fundamental to improving the detection and diagnosis of breast cancer, which contributes significantly to survival rates and treatment effectiveness. This study proposes a multidisciplinary approach that combines analytical methods and graphical visualizations to classify breast tumors as benign or malignant by considering supervised and unsupervised learning algorithms. The dataset from the repository of the University of Wisconsin (USA) was adopted. It comprises 569 breast tumor biopsy samples, with 32 features measured from digitized images of biopsy slides. Initially, Pearson correlation was adopted as a similarity metric for unsupervised learning. As a result, six clusters were identified through a dendrogram. The purity of each cluster was evaluated, highlighting the most predominant class in each cluster. In supervised learning, the Principal Component Analysis (PCA) was considered for reducing the number of features, achieving 10 most relevant ones. Then, the data set was divided into training (80%) and test (20%) sets for applying the Support Vector Machine (SVM). The comparison between Pearson’s correlation and SVM demonstrated a notable advantage of SVM in terms of accuracy in classifying breast tumors. The use of cross-validation showed the superiority of SVM over Pearson correlation for this specific purpose. The implementation of PCA resulted in a significant increase in the accuracy of the SVM classifier, reaching a score of 0.9825, compared to 0.9561 without applying PCA. This result highlights the importance of selecting relevant features to improve model performance. The analysis of breast tumor features and the classification using supervised and unsupervised machine learning methods offer valuable perspectives to improve breast cancer diagnosis and treatment practices. The use of algorithms such as SVM, together with dimensionality reduction techniques such as PCA, can result in significant improvements in diagnostic accuracy and effectiveness, directly benefiting patient care in this critical area of medicine.

Title of the Event: 20º Congresso Brasileiro de Bioinformática: X-Meeting 2024
City of the Event: Salvador
Title of the Proceedings of the event: X-Meeting presentations
Name of the Publisher: Even3
Means of Dissemination: Meio Digital

How to cite

VALENTIN, Ana Beatriz Miranda et al.. ANALYSIS OF MACHINE LEARNING ALGORITHMS APPLIED TO BREAST CANCER DIAGNOSIS.. In: X-Meeting presentations. Anais...Salvador(BA) Hotel Deville Prime, 2024. Available in: https//www.even3.com.br/anais/xmeeting-2024/822227-ANALYSIS-OF-MACHINE-LEARNING-ALGORITHMS-APPLIED-TO-BREAST-CANCER-DIAGNOSIS. Access in: 27/04/2025