Using acetowhite opacity index for detecting cervical intraepithelial neoplasia

Area of Science:

• Gynecologic pathology
• Medical imaging
• Computational pathology

Background:

• Cervical intraepithelial neoplasia (CIN) diagnosis relies on subjective colposcopic assessment of acetowhitening.
• Immature metaplastic and dysplastic epithelia exhibit characteristic whitening post-acetic acid application.
• Objective, automated analysis can improve diagnostic accuracy and consistency.

Purpose of the Study:

• To develop and evaluate an automated digital imaging system for analyzing the acetowhitening process in cervical tissue.
• To quantify temporal and spatial changes in acetowhitening using color and texture analysis.
• To assess the system's ability to discriminate high-grade dysplasia (CIN2+) from normal and low-grade lesions.

Main Methods:

• Acquisition of paired digital colposcopic images (pre- and post-acetic acid application).
• Automatic image alignment and extraction of spatial changes using color and texture.
• Calculation of an automated opacity index to quantify temporal whitening changes.
• Evaluation of the system on 99 human subjects, comparing results to histological diagnosis.

Main Results:

• The automated system successfully extracted spatial and temporal parameters of acetowhitening.
• The proposed opacity index demonstrated high performance in discriminating between different grades of cervical lesions.
• Sensitivity and specificity for detecting high-grade dysplasia (CIN2+) were 94% and 87%, respectively.

Conclusions:

• Automated analysis of the acetowhitening process using digital colposcopy shows significant promise for objective CIN diagnosis.
• The developed opacity index offers a reliable, quantitative measure for identifying high-grade cervical dysplasia.
• This technology has the potential to enhance diagnostic accuracy and support clinical decision-making in cervical cancer screening.