Citation: Kalbhor, M.; Shinde, S.
ColpoClassifier: A Hybrid
Framework for Classification of the
Cervigrams. Diagnostics 2023, 13,
1103. https://doi.org/10.3390/
diagnostics13061103
Academic Editor: Zoard T Krasznai
Received: 6 February 2023
Revised: 3 March 2023
Accepted: 6 March 2023
Published: 14 March 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
diagnostics
Article
ColpoClassifier: A Hybrid Framework for Classification
of the Cervigrams
Madhura Kalbhor and Swati Shinde *
Department of Computer Engineering, Pimpri Chinchwad College of Engineering, Pune 411044, India;
madhura.kalbhor@pccoepune.org
* Correspondence: swati.shinde@pccoepune.org
Abstract: Colposcopy plays a vital role in detecting cervical cancer. Artificial intelligence-based
methods have been implemented in the literature for the classification of colposcopy images. However,
there is a need for a more effective method that can accurately classify cervigrams. In this paper,
ColpoClassifier, a hybrid framework for the classification of cervigrams, is proposed, which consists
of feature extraction followed by classification. This paper uses a Gray-level co-occurrence matrix
(GLCM), a Gray-level run length matrix (GLRLM), and a histogram of gradients (HOG) for feature
extraction. These features are combined to form a feature fusion vector of the form GLCM + GLRLM +
HOG. The different machine learning classifiers are used for classification by using individual feature
vectors as well as feature fusion vectors. The dataset used in this paper is compiled by downloading
images from the WHO website. Two variants of this dataset are created, Dataset-I contains images of
the aceto-whitening effect, green filter, iodine application, and raw cervigram while Dataset-II only
contains images of the aceto-whitening effect. This paper presents the classification performance on
all kinds of images with the individual as well as hybrid feature fusion vector and concludes that
hybrid feature fusion vectors on aceto-whitening images have given the best results.
Keywords: colposcopy; feature extraction; machine learning; feature fusion; GLCM; GLRLM; HOG
1. Introduction
Cervical cancer affects the cervix of the vagina. Human Papillomavirus (HPV) is
a major cause of cervical cancer. Other causes include smoking, sexually transmitted
infections, and immune system dysfunction. Early detection of cervical cancer plays an
important role in its prevention and treatment. Cervical cancer screening is necessary for
early detection; however, less developed countries lack effective screening programs [1].
A Pap smear is the most common screening procedure that distinguishes abnormal
cells and predicts the risk of cervical cancer. However, the Pap smear has limitations as
incidences of both false negatives and false positives are high. If a Pap smear can predict
the possibility of cancer with its limitations, then we suggest a colposcopy for a more
accurate diagnosis. Colposcopy is a well-suited procedure for precancerous examination of
the cervix that can also be used in a low-cost setting. The availability of professionals and
technology plays an important role in the diagnosis and treatment of cervical cancer [1,2].
Colposcopy is a very subjective process as it depends on the knowledge and experience
of the doctors. The primary aim of this test is to identify premalignant or malignant
lesions as well as genital warts, polyps, and infections [3]. Visual inspection involves the
application of acetic acid to the visible part of the cervix.
The classification of a cervigram involves analyzing complicated patterns. Image
analysis and machine learning methods are used widely in the medical field. These
methods assist doctors by providing them with reasonable diagnoses. Primary features to
detect abnormal cervix are aceto-whitening of the cervix, punctuations and mosaic patterns,
erosion, and a rough surface [4].
Diagnostics 2023, 13, 1103. https://doi.org/10.3390/diagnostics13061103 https://www.mdpi.com/journal/diagnostics