Abstract:

Breast cancer is caused by uncontrolled cell growth and early detection is crucial for successful treatment. Diagnostic methods such as MRI, mammography, ultrasound and biopsy are used in the diagnosis of cancer. AI-based clinical decision support systems can predict treatment outcomes. A study was conducted to diagnose breast cancer using blood values and to increase performance metrics by applying discretization preprocessing. Machine learning methods were compared with 10-fold cross-validation and discretization applied and unapplied data, and the model with the best results was created. It gave the highest classification performance in breast cancer outcome prediction, with a sensitivity value of 0.828 for the Multi-Layer Sensor. In the model created by applying discretization, the Support Vector Machines increased from 55% to 72%.

Keywords: