Sanaz Hariri Tabrizi1, Farah Farzaneh2, Seyed Mahmoud Reza Aghamiri3. 1. Department of Medical Radiation Engineering, Shahid Beheshti University, Tehran, Iran. F_farzaneh@sbmu.ac.ir. 2. Preventative Gynecology Research Center, Department of Obstetrics and Gynecology, Shahid Beheshti University of Medical Sciences, Tehran, Iran. 3. Department of Medical Radiation Engineering, Shahid Beheshti University, Tehran, Iran.
Abstract
BACKGROUND: We assessed the performance of single fiber reflectance spectroscopy (SFRS) in discriminating cervical squamous intraepithelial lesions (SIL) from non-SILs and the probable influence of environmental factors on its performance. METHODS: SFRS was used to measure the reflected light from cervical tissue of 157 patients undergoing standard colposcopy. Seven parameters extracted from the spectra in addition to two biographic parameters were used to compare the biopsy-confirmed SILs with non-SILs. The tissue classification capability was reported by receiver operating characteristic (ROC) curves. In addition, the effect of five interfering parameters, including the probe, clinician, hospital, menopausal status and age of the patient on spectroscopic parameters were investigated by the Kruskal-wallis test. RESULTS: The average vessel diameter and beta-carotene concentration were found to be the parameters contributing to tissue discrimination. SFRS could differentiate between SILs and non-SILs with sensitivity, specificity and area under the ROC curve of 63% ± 6%, 68% ± 6% and 0.69 ± 0.04, respectively. None of the five environmental parameters interfered with the discriminator spectroscopic parameters. CONCLUSIONS: SFRS was found as a noninvasive, fast, compact, cost-effective, independent, and acceptably accurate system to help the clinician to reduce the number of unnecessary biopsies during the colposcopy procedure.
BACKGROUND: We assessed the performance of single fiber reflectance spectroscopy (SFRS) in discriminating cervical squamous intraepithelial lesions (SIL) from non-SILs and the probable influence of environmental factors on its performance. METHODS: SFRS was used to measure the reflected light from cervical tissue of 157 patients undergoing standard colposcopy. Seven parameters extracted from the spectra in addition to two biographic parameters were used to compare the biopsy-confirmed SILs with non-SILs. The tissue classification capability was reported by receiver operating characteristic (ROC) curves. In addition, the effect of five interfering parameters, including the probe, clinician, hospital, menopausal status and age of the patient on spectroscopic parameters were investigated by the Kruskal-wallis test. RESULTS: The average vessel diameter and beta-carotene concentration were found to be the parameters contributing to tissue discrimination. SFRS could differentiate between SILs and non-SILs with sensitivity, specificity and area under the ROC curve of 63% ± 6%, 68% ± 6% and 0.69 ± 0.04, respectively. None of the five environmental parameters interfered with the discriminator spectroscopic parameters. CONCLUSIONS: SFRS was found as a noninvasive, fast, compact, cost-effective, independent, and acceptably accurate system to help the clinician to reduce the number of unnecessary biopsies during the colposcopy procedure.