Cervical cancer detection using K nearest neighbor imputer and stacked ensemble learningmodel.

Area of Science:

• Medical Informatics
• Machine Learning in Healthcare
• Oncology

Background:

• Cervical cancer is a major cause of death for women in developing countries.
• Early detection and treatment are crucial for reducing mortality.
• Automated cervical cancer detection using Pap smear images is hindered by missing data in datasets.

Purpose of the Study:

• To develop an automated system for cervical cancer prediction that effectively manages missing data.
• To achieve high accuracy in cervical cancer detection despite data imperfections.
• To improve the reliability of machine learning models for medical diagnostics.

Main Methods:

• A stacked ensemble voting classifier model was developed.
• The system integrates three distinct machine learning models.
• KNN Imputer was utilized to address missing values in the dataset.

Main Results:

• The proposed model achieved an accuracy of 99.41% with KNN imputation.
• Precision reached 97.63%, recall 95.96%, and F1 score 96.76%.
• Comparative analysis showed superior performance against seven other algorithms, with and without imputation.

Conclusions:

• The developed system effectively handles missing data in cervical cancer detection datasets.
• The findings support early detection and improved quality of care for cervical cancer patients.
• The proposed model demonstrates effectiveness compared to current state-of-the-art methods.