Hierarchical Decision Approach based on Neural Network and Genetic Algorithm method for single image classification of Pap smear

Cervical cancer is a cancer that attacks the cervix and is marked by abnormal growth in cervical cells. There is one method of early detection of cervical cancer, namely Pap smear that performs the examination and can help prevent early cervical cancer. Pap smear cell classification consists of 7 categories of classes (Normal Superficial, Normal Intermediate, Normal Colummar, Mild (Light) Dyplasia, Moderate Dyplasia, Servere Dyplasia and Carcinoma In Situ. There is still difficulty in classification process for seven classes on single cell image of Pap smear. The algorithm applied in the Pap smear image classification was Neural Network algorithm classification and feature selection using Genetic Algorithm. The best model of the classification result became the Hierarchical Decision Approach (HDA) model, so a new classification method approach for Pap smear image was proposed. Comparison of classification results was performed by using Neural Network algorithm and feature optimization using Genetic Algorithm to determine the increase of accuracy. The classification result of Pap smear image into 7 classes by using Hierarchical Decision Approach (HDA) method got the highest value of 79,78% while classification using Neural Network algorithm and feature optimization using Genetic Algorithm had the highest value of 68,48%.