Recent advancement in cervical cancer diagnosis for automated screening: a detailed review

Chitra, B.; Kumar, S. S.

doi:10.1007/s12652-021-02899-2

Cited by 31 publications

(31 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CNN‐based classification models have been proven to perform better than traditional machine learning algorithms in natural and medical image analysis, but they require massive amounts of labeled images. Compared with numerous deep‐learning‐based classification models for cervical cytology, colposcopy, and histopathology images, few CNN‐based cervical OCT image classification models are available for comparison 39–42 . This study aims to propose a new SSL strategy for efficient OCT image classification model training rather than to design a new classification model.…”

Section: Discussionmentioning

confidence: 99%

Cervical optical coherence tomography image classification based on contrastive self‐supervised texture learning

Chen

Wang

2022

Medical Physics

View full text Add to dashboard Cite

Background Cervical cancer (CC) seriously affects the health of the female reproductive system. Optical coherence tomography (OCT) emerged as a noninvasive, high‐resolution imaging technology for cervical disease detection. However, OCT image annotation is knowledge‐intensive and time‐consuming, which impedes the training process of deep‐learning‐based classification models. Purpose This study aims to develop a computer‐aided diagnosis (CADx) approach to classifying in‐vivo cervical OCT images based on self‐supervised learning. Methods In addition to high‐level semantic features extracted by a convolutional neural network (CNN), the proposed CADx approach designs a contrastive texture learning strategy to leverage unlabeled cervical OCT images’ texture features. We conducted 10‐fold cross‐validation on the OCT image dataset from a multicenter clinical study on 733 patients from China. Results In a binary classification task for detecting high‐risk diseases, including high‐grade squamous intraepithelial lesion and CC, our method achieved an area‐under‐the‐curve value of 0.9798 ± 0.0157 with a sensitivity of 91.17% ± 4.99% and a specificity of 93.96% ± 4.72% for OCT image patches; also, it outperformed two out of four medical experts on the test set. Furthermore, our method achieved 91.53% sensitivity and 97.37% specificity on an external validation dataset containing 287 three‐dimensional OCT volumes from 118 Chinese patients in a new hospital using a cross‐shaped threshold voting strategy. Conclusions The proposed contrastive‐learning‐based CADx method outperformed the end‐to‐end CNN models and provided better interpretability based on texture features, which holds great potential to be used in the clinical protocol of “see‐and‐treat.”

show abstract

Section: Discussionmentioning

confidence: 99%

Cervical optical coherence tomography image classification based on contrastive self‐supervised texture learning

Chen

Wang

2022

Medical Physics

View full text Add to dashboard Cite

show abstract

“…CVF is a body fluid that is used for routine screening (e.g. for hrHPV detection) [27,28] as well as for research purposes [29,30]. To find reliable molecular markers in these samples would be highly beneficial provided sufficient sensitivity and specificity can be reached.…”

Section: Sample Typesmentioning

confidence: 99%

“…However, their implementation requires trained laboratory personnel and experienced pathologists for data interpretation with considerable inter-observer variability. Recent developments in computer-guided image analysis using artificial intelligence may alleviate this problem with the goal to perform such assays reliably even when trained personnel is not available [27,28]. In the following, we will give an overview of techniques that are used for the detection of molecular markers starting from the already widely accepted PCR-based hrHPV assays and assays to assess changes in DNA methylation patterns to discovery, validation, and screening techniques for protein markers.…”

Section: Assaysmentioning

confidence: 99%

Molecular markers for cervical cancer screening

Güzel

Hoff

Kok

et al. 2021

Expert Review of Proteomics

View full text Add to dashboard Cite

This review gives an overview of current screening practices for cervical cancer. In the introduction, we will cover approaches of population screening focusing on high-risk Human Papilloma Virus (hrHPV) and the need for a better triage assay. We will further assess the impact of current vaccination programs on screening. Subsequently, the review will cover various technological aspects of nucleic acid-and protein-based biomarker assays. We will then detail different molecular markers in view of their use in triage assays, emphasizing epigenetic and protein markers. Finally, we will place this in the context of cost-effectiveness considerations in view of their implementation in high-as well as in low-to middle-income countries. Introduction: Cervical cancer remains a significant healthcare problem, notably in low-to middle-income countries. While a negative test for hrHPV has a predictive value of more than 99.5%, its positive predictive value is less than 10% for CIN2+ stages. This makes the use of a so-called triage test indispensable for population-based screening to avoid referring women, that are ultimately at low risk of developing cervical cancer, to a gynecologist. This review will give an overview of tests that are based on epigenetic marker panels and protein markers. Areas covered: There is a medical need for molecular markers with a better predictive value to discriminate hrHPV-positive women that are at risk of developing cervical cancer from those that are not. Areas covered are epigenetic and protein markers as well as health economic considerations in view of the fact that most cases of cervical cancer arise in low-to-middle-income countries. Expert opinion: While there are biomarker assays based on changes at the nucleic acid (DNA methylation patterns, miRNAs) and at the protein level, they are not widely used in population screening. Combining nucleic acid-based and protein-based tests could improve the overall specificity for discriminating CIN2+ lesions that carry a low risk of progressing to cervical cancer within the screening interval from those that carry an elevated risk. The challenge is to reduce unnecessary referrals without an undesired increase in false-negative diagnoses resulting in cases of cervical cancer that could have been prevented. A further challenge is to develop tests for low-and middle-income countries, which is critical to reduce the worldwide burden of cervical cancer.

show abstract

“…Furthermore, this cancer can be successfully treated if detected early. Although, the recent advancement of using the Pap smear test in healthcare firms helps in diagnosing cervical cancer more effectively these days (Basak et al, 2021;Chitra and Kumar, 2021). This can be taken place by doctors, collecting a sample of cells from the surface of the cervix in women.…”

Section: Introductionmentioning

confidence: 99%

SKCV: Stratified K-fold cross-validation on ML classifiers for predicting cervical cancer

Prusty

Patnaik

Dash

2022

Front. Nanotechnol.

119

View full text Add to dashboard Cite

Cancer is the unregulated development of abnormal cells in the human body system. Cervical cancer, also known as cervix cancer, develops on the cervix’s surface. This causes an overabundance of cells to build up, eventually forming a lump or tumour. As a result, early detection is essential to determine what effective treatment we can take to overcome it. Therefore, the novel Machine Learning (ML) techniques come to a place that predicts cervical cancer before it becomes too serious. Furthermore, four common diagnosis testing namely, Hinselmann, Schiller, Cytology, and Biopsy have been compared and predicted with four common ML models, namely Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbors (K-NNs), and Extreme Gradient Boosting (XGB). Additionally, to enhance the better performance of ML models, the Stratified k-fold cross-validation (SKCV) method has been implemented over here. The findings of the experiments demonstrate that utilizing an RF classifier for analyzing the cervical cancer risk, could be a good alternative for assisting clinical specialists in classifying this disease in advance.

show abstract

Recent advancement in cervical cancer diagnosis for automated screening: a detailed review

Cited by 31 publications

References 67 publications

Cervical optical coherence tomography image classification based on contrastive self‐supervised texture learning

Cervical optical coherence tomography image classification based on contrastive self‐supervised texture learning

Molecular markers for cervical cancer screening

SKCV: Stratified K-fold cross-validation on ML classifiers for predicting cervical cancer

Contact Info

Product

Resources

About