Classification of adeno carcinoma, high squamous intraephithelial lesion, and squamous cell carcinoma in Pap smear images based on extreme learning machine

Cervical cancer is a malignant tumour that attacks the female genital area originating from epithelial
metaplasia in the squamous protocol junction area. One method of diagnosis of cervical cancer is to do
a Pap smear examination by taking a cervical cell smear from the woman’s cervix and observing its cell
development. However, examination of cervical cancer from Pap smear results usually takes a long time.
This is because medical practitioners still rely on visual observations in the analysis of the results of Pap
smear so that the results are subjective. Therefore, we need a programme that can help the classification
process in establishing a diagnosis of cervical cancer with high accuracy results. In this study, a cervical
cancer classification program was developed using a combination of the Grey Level Co-occurrence Matrix
(GLCM) and Extreme Learning Machine (ELM) methods. There are three classes of cervical cell images
classified, namely adenocarcinoma, High Squamous Intraepithelial Lesion (HSIL) and Squamous Cell
Carcinoma (SCC). From the results of the training program obtained an accuracy 100% and from the
testing program obtained an accuracy of 80%.