Enhancing Cancer Diagnosis through Machine Learning and Swarm Intelligence Techniques

Sep 17, 2026

Uploaded by Eliana

Machine learning has enabled high-performance detection and identification of cancer, particularly when medical images serve as input. To further strengthen model performance during processing, this research integrates feature identification with an optimization strategy. The proposed approach applies the ABC algorithm to select optimal features, using CT images of breast cancer from the WDBC dataset with 570 cases. The resulting model reaches a classification accuracy of 94.74% .

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Introduction

Artificial Bee Colony algorithm

Machin learning algorithms

Logistic Regression

Decision Tree

|
Pure Sciences International Journal of Kerbala | | | --- | --- | | Enhancing Cancer Diagnosis through Machine Learning and Swarm
Intelligence Techniques
Maha Sabri Altememe a*
a College of Computer Science and Information Technology, University of Kerbala, Iraq | | | P A P E R I N F O
Received: 12.09.2025
Accepted: 15.12.2025
Published: 31.03.2026 Keywords:
classification, breast cancer, medical images, ABC algorithm
| A b s t r a c t
Machine learning has achieved wide success in detecting and identifying cancer diseases, especially those based on medical images. To increase the performance of models, machine learning techniques are combined with other techniques in the processing stage. In this research, the ABC algorithm was used in the feature identification process to help the algorithm choose the optimal solution more accurately.The medical images used are CT images of breast cancer collected in the form of a standard dataset of 570 known as WDBC dataset. The proposed model achieved classification accuracy 94.74% .
DOI: 10.53851/psijk.v3.i9.63-68 | 1. INTRODUCTION Today, artificial intelligence has become widely important in many applications through the use of its multiple branches. Among these branches are machine learning techniques that are used to classify data and achieve high performance in implementation. The implementation of machine learning techniques requires the use of data for the purpose of training the model and then predicting the results and making decisions on new data (Zhang et al., 2020) . However, the effectiveness of the models depends on a number of determinants, including the percentage of data, its quality, the type of algorithm that was applied and its suitability for this data, in addition to the method of extracting features from this data. Therefore, to increase the efficiency of the model, algorithms are used that help extract important features. One of these algorithms is the artificial bee colony algorithm, which works to select the ideal features and can be applied to medical images, especially in medical image data for cancer. It is considered one of the major determinants due to the differences in cancer diseases, soit needs high-quality techniques to infer features(Tharwat et al., 2017) . In this research, a model is designed that combines machine learning techniques and ABC algorithm to select optimal features. Contributions to this study: Artificial Bee Colony algorithm-based featureselection for breast cancer detection. In order to improve patient care and diagnosis accuracy in the field of oncology, the project intends to create an efficient . 1. 1. Artificial Bee Colony algorithm Karaboga presented the Artificial Bee Colony (ABC) algorithm in 2005 (Karaboga, 2005) as a swarm-based technique to solve numerical problems (Rao et al., 2020) This algorithm's concept is derived from the clever ways that colonies of honey bees hunt for food. The work crew is divided into three primary groups. The first group is known as the scout bees, and their primary duty is to randomly hunt for food. The job of the second team, known as the worker bees, is to look for food sources around the ones that have already been found. The job of the third team, known as the observer bees, is to keep an eye on the worker bees and assess the food's worth and quality in order to choose the best (Tharwat et al., 2017) . In the ABC theory, the search process extracts the features present in the data in general, which represents finding specific solutions to the problem, and then the optimization stage is chosen by finding the best types of features, and this process is repeated repeatedly until the best features are obtained that help us in the classification stages. To clarify the work of the algorithm that is used to find the optimal solution, as mentioned previously, it consists of three groups: worker bees, spectator bees, and scout bees. The following code illustra

FAQs

What is the main method proposed for breast cancer diagnosis?

The study combines machine learning with the ABC (Artificial Bee Colony) algorithm for feature selection and optimization in medical image processing.

Which dataset and image type are used in the experiments?

CT images of breast cancer are used, collected in a standard dataset of 570 cases known as the WDBC dataset.

How accurate is the proposed classification model?

The proposed model achieves a classification accuracy of 94.74%.

Sep 17, 2026

Uploaded by Eliana