Quantitative spectroscopy has recently been extended from a contact-probe to wide-area spectroscopic imaging to enable mapping of optical properties across a wide area of tissue. We train quantitative spectroscopic imaging (QSI) to identify cervical high-grade squamous intraepithelial lesions (HSILs) in 34 subjects undergoing the loop electrosurgical excision procedure (LEEP subjects). QSI’s performance is then prospectively evaluated on the clinically suspicious biopsy sites from 47 subjects undergoing colposcopic-directed biopsy. The results show the per-subject normalized reduced scattering coefficient at 700 nm (A n ) and the total hemoglobin concentration are significantly different (p<0.05 ) between HSIL and non-HSIL sites in LEEP subjects. A n alone retrospectively distinguishes HSIL from non-HSIL with 89% sensitivity and 83% specificity. It alone applied prospectively on the biopsy sites distinguishes HSIL from non-HSIL with 81% sensitivity and 78% specificity. The findings of this study agree with those of an earlier contact-probe study, validating the robustness of QSI, and specifically A n , for identifying HSIL. The performance of A n suggests an easy to use and an inexpensive to manufacture monochromatic instrument is capable of early cervical cancer detection, which could be used as a screening and diagnostic tool for detecting cervical cancer in low resource countries.
It has long been speculated that underlying variations in tissue anatomy affect in vivo spectroscopic measurements. We investigate the effects of cervical anatomy on reflectance and fluorescence spectroscopy to guide the development of a diagnostic algorithm for identifying high-grade squamous intraepithelial lesions (HSILs) free of the confounding effects of anatomy. We use spectroscopy in both contact probe and imaging modes to study patients undergoing either colposcopy or treatment for HSIL. Physical models of light propagation in tissue are used to extract parameters related to tissue morphology and biochemistry. Our results show that the transformation zone, the area in which the vast majority of HSILs are found, is spectroscopically distinct from the adjacent squamous mucosa, and that these anatomical differences can directly influence spectroscopic diagnostic parameters. Specifically, we demonstrate that performance of diagnostic algorithms for identifying HSILs is artificially enhanced when clinically normal squamous sites are included in the statistical analysis of the spectroscopic data. We conclude that underlying differences in tissue anatomy can have a confounding effect on diagnostic spectroscopic parameters and that the common practice of including clinically normal squamous sites in cervical spectroscopy results in artificially improved performance in distinguishing HSILs from clinically suspicious non-HSILs.
Model-based light scattering spectroscopy (LSS) seemed a promising technique for in-vivo diagnosis of dysplasia in multiple organs. In the studies, the residual spectrum, the difference between the observed and modeled diffuse reflectance spectra, was attributed to single elastic light scattering from epithelial nuclei, and diagnostic information due to nuclear changes was extracted from it. We show that this picture is incorrect. The actual single scattering signal arising from epithelial nuclei is much smaller than the previously computed residual spectrum, and does not have the wavelength dependence characteristic of Mie scattering. Rather, the residual spectrum largely arises from assuming a uniform hemoglobin distribution. In fact, hemoglobin is packaged in blood vessels, which alters the reflectance. When we include vessel packaging, which accounts for an inhomogeneous hemoglobin distribution, in the diffuse reflectance model, the reflectance is modeled more accurately, greatly reducing the amplitude of the residual spectrum. These findings are verified via numerical estimates based on light propagation and Mie theory, tissue phantom experiments, and analysis of published data measured from Barrett's esophagus. In future studies, vessel packaging should be included in the model of diffuse reflectance and use of model-based LSS should be discontinued.