Cervical Cancer Detection Techniques: A Chronological Review

Area of Science:

• Gynecologic Oncology
• Medical Imaging
• Artificial Intelligence in Healthcare

Background:

• Cervical cancer remains a significant global health challenge with high mortality and incidence rates.
• Traditional Pap smear tests for cervical cancer screening have limitations, including subjectivity and potential for missed diagnoses.
• There is a growing need for enhanced cervical cancer detection methods to improve accuracy and patient outcomes.

Purpose of the Study:

• To provide a chronological review of cervical cancer detection techniques.
• To evaluate the evolution and effectiveness of computer-aided detection (CAD) systems in cervical cancer screening.
• To highlight the potential of machine learning (ML)-based CAD systems for improved diagnosis.

Main Methods:

• A systematic literature review was conducted using the Scopus database (1996-2022).
• Search terms included "(cervix OR cervical) AND (cancer OR tumor) AND (detect* OR diagnosis)".
• Studies reporting on the development or evaluation of cervical cancer detection techniques, including traditional and CAD methods, were included.

Main Results:

• Early CAD systems using image processing had limited success due to low sensitivity and specificity.
• Machine learning (ML) algorithms introduced in the early 2000s significantly improved automated analysis of cervical cell images.
• ML-based CAD systems demonstrated improved sensitivity and specificity compared to traditional methods in several studies.

Conclusions:

• Significant advancements have been made in cervical cancer detection techniques over the past decades.
• ML-based CAD systems, such as HISCCD and ACSS, show promise for enhancing cervical cancer detection accuracy and sensitivity.
• Further validation and research are necessary for broad acceptance and to reduce the global burden of cervical cancer.