Classification of cervical cancer from Pap smear images: a convolutional neural network approach
by Siti Noraini Sulaiman; Ajmal Hadi Ahmad Hishamuddin; Iza Sazanita Isa; Muhammad Khusairi Osman; Zainal Hisham Che Soh
International Journal of Intelligent Systems Technologies and Applications (IJISTA), Vol. 21, No. 3, 2023

Abstract: Cervical cancer is a significant global issue, with Pap smear tests being a common screening tool for precancerous stages. This study aims to develop a computer-aided diagnostics system that can classify precancerous cells from Pap smear images. The project employs convolutional neural networks (CNNs) trained using pre-processed images, adaptive fuzzy K-means (AFKM), and fuzzy C-means (FCM) to classify cervical cancer cell data as normal or abnormal. The datasets used in the project include normal, low-grade squamous intraepithelial lesion (LSIL), and high-grade squamous intraepithelial lesion (HSIL) categories. CNN1, CNN2, and CNN3 have been developed and CNN2 was chosen due to its highest accuracy of 87.71%. The CNN2 trained with AFKM outperformed other networks with an accuracy of 89.53%, precision of 0.870, recall of 0.870, specificity of 0.935, and F1-score of 0.870. This study demonstrates the potential of deep learning-based approaches for identifying and classifying cervical cell pre-cancerous stages.

Online publication date: Fri, 29-Sep-2023

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Intelligent Systems Technologies and Applications (IJISTA):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com