Immature and dysplastic cervical squamous epithelium whitens after the application of acetic acid during a colposcopic examination. The whitening process occurs visually over several minutes and subjectively discriminates between dysplastic and normal tissue. In this work, examples of the acetowhitening process are detailed in three ways: the color-imaged colposcopic appearance of the acetowhitening of high-grade cervical intraepithelial neoplasia (CIN 2/3), the kinetics of these reflectance patterns transformed to reduce noise in the signal, and a self-normalized green to red ratio measurement of the kinetics of these reflectance patterns. A total of six patients with biopsy confirmed CIN 2/3 were examined to obtain a set of timed images tracking the acetowhitening and the whitening-decay process over the course of 5-10 min. Regions of normal mature squamous epithelium within the same patients were also followed as an internal control. We determined that the temporal change over a 10 min time period in the ratio of green to red light intensities, taken from the respective color channels of the CCD, provides a reliable measure to clearly distinguish CIN 2/3 from normal cervical epithelium. This imaging and data normalization procedure may be applied to cervical lesions of different grades, to determine if a quantitative estimate provides predictive value during the colposcopic diagnosis.

Fluorescence emission and diffuse reflectance spectra of freshly excised cervical tissue were studied with two specially designed contact probes. The objective of the study was to reach a better understanding of the relationship between spectroscopic measurements and cervical tissue morphology. Tissue samples from loop electro-surgical excision and hysterectomy specimens were measured within 20 to 90 minutes of excision. Emission spectra with 337 nm excitation, and reflectance spectra were collected at wavelengths between 370 and 720 nm from different tissue sites. Hematoxylin-eosin stained slides of the measured zones were obtained and compared to the spectra.In one experiment, a contact probe with a central illumination fiber and two concentric rings of detection fibers (radii 0.1 and 1 mm), was placed in contact with the epithelium and used to measure spectra from ectocervix and endocervix. The influence of 5% acetic acid on fluorescence and reflectance spectra was also investigated. In another experiment, a single 100-micron fiber probe was placed perpendicular to a cut edge of tissue and scanned to measure spectra in depth. Depth scans were made over various areas of the cervix. Data were analyzed to estimate the distribution of common tissue fluorophores as a function of depth from the surface and location on the cervix. Significant differences in shape and in absolute intensity were observed in fluorescence and reflectance spectra taken from endocervical tissue and ectocervical tissue. Hemoglobin absorption was more pronounced for endocervical tissue. Changes with depth in NADH fluorescence were less pronounced with endocervical tissue. The influence of NADH decreases while collagen increases is one way to describe the differences observed in the shape of the spectra observed as a function of depth. The probe geometry, in particular the source/detector separation distance, plays an important role in defining the shape of fluorescence and reflectance spectra. Data collected in this study from nonneoplastic cervical tissue reveal changes after the application of acetic acid that are not negligible, the amplitude of fluorescence spectra decreases while the amplitude of reflectance spectra increases.