The incidence of invasive cervical cancer in white women under 50 has been increasing abut 3% a year since 1986; this follows a large increase in CIN that began in the mid 1980s. There is concern that the mortality of cervical cancer will increase, unless improvements are made in screening and detection methods. Techniques based on quantitative optical spectroscopy and imaging have the potential to fulfill this need. Computer based algorithms can be developed to correlate optical signals to histo- pathology, enabling semi-automated, real-time, minimally invasive identification of focal areas of pathology. Most of these techniques have bee tested only a very limited clinical trials (n>20). Further, the choice of illumination and detection wavelengths and geometries has been made arbitrarily or on the basis of small in vitro studies. We believe the benefits of optical approaches will only be realized by conducting mid- size clinical trials to identify the optical technique and choice of wavelengths which give the best performance. The goal of this proposal is to conduct a comprehensive set of clinical trials to develop and test optical methods for the screening and detection of CIN. We will measure optical spectra of colposcopically normal and abnormal sites in a group of 800 women with no history of abnormal smears. Algorithms for CIN will be developed and evaluated relative to the gold standard of colposcopically directed biopsy using fluorescence spectra at multiple excitation wavelengths, reflectance spectra, and the combination of fluorescence and reflectance.
The second aim i s to investigate the diagnostic ability of quantitative optical imaging techniques. We will conduct a series of clinical trials comparing the performance of various imaging technologies to the higher performing algorithm based on spectroscopy developed in the first aim.
The final aim of this proposal is to use objective data from clinical trials to determine the most effective technique (or combination of techniques) for the optical detection of CIN. Because optical spectra can be recorded remotely, in early real time without the need for tissue removal and data analysis can be automated, we believe that optical diagnosis affords many important advantages over traditional techniques including the potential to reduce the need for clinical expertise, reduce the number of unnecessary biopsies, to enable combined diagnosis and therapy in those patients who might benefit most and the potential to reduce health care costs.
Showing the most recent 10 out of 110 publications