Image analysis for discrimination of cervical neoplasia.
Colposcopy involves visual imaging of the cervix for patients who have exhibited some prior indication of abnormality, and the major goals are to visually inspect for any malignancies and to guide biopsy sampling. Currently colposcopy equipment is being upgraded in many health care centers to incorporate digital image acquisition and archiving. These permanent images can be analyzed for characteristic features and color patterns which may enhance the specificity and objectivity of the routine exam. In this study a series of images from patients with biopsy confirmed cervical intraepithelia neoplasia stage 2/3 are compared with images from patients with biopsy confirmed immature squamous metaplasia, with the goal of determining optimal criteria for automated discrimination between them. All images were separated into their red, green, and blue channels, and comparisons were made between relative intensity, intensity variation, spatial frequencies, fractal dimension, and Euler number. This study indicates that computer-based processing of cervical images can provide some discrimination of the type of tissue features which are important for clinical evaluation, with the Euler number being the most clinically useful feature to discriminate metaplasia from neoplasia. Also there was a strong indication that morphology observed in the blue channel of the image provided more information about epithelial cell changes. Further research in this field can lead to advances in computer-aided diagnosis as well as the potential for online image enhancement in digital colposcopy.