Applied Optics

Assessment of anisotropy of collagen structures through spatial frequencies of Mueller matrix images for cervical pre-cancer detection

Analysis of spatial frequency of Mueller matrix (MM) images in the Fourier domain yields quantifying parameters of anisotropy in the stromal region in normal and precancerous tissue sections of human uterine cervix. The spatial frequencies of MM elements reveal reliable information of microscopic structural organization arising from the different orientations of collagen fibers in the connective tissue and their randomization with disease progression. Specifically, the local disorder generated in the normal periodic and regular structure of collagen during the growth of the cervical cancer finds characteristic manifestation in the Fourier spectrum of the selected Mueller matrix elements encoding the anisotropy effects through retardance and birefringence. In contrast, Fourier spectra of differential polarization gated images are limited to only one orientation of collagen. Fourier spectra of first row elements M11, M12, M13, and M14 and first column elements M11, M21, M31, and M41 discriminates cervical inter-epithelial neoplasia (CIN)-I from normal cervical tissue samples with 95%–100% sensitivity and specificity. FFT spectra of first and fourth row elements classify CIN-I and CIN-II grades of cervical cancerous tissues with 90%–100% sensitivity and 87%–100% specificity. Normal and CIN-II grade samples are successfully discriminated through Fourier spectra of every MM element while that of M31 element arises as the key classifier among normal, CIN-I, and CIN-II grades of cervical cancer with 100% sensitivity and specificity. These results demonstrate the promise of spatial frequency analysis of Mueller matrix images as a novel, to the best of our knowledge, approach for cancer/precancer detection.

Fluorescence Mueller matrix imaging of the stromal region in unstained cervical tissue sections: quantitative categorization of CIN grades

This study introduces a fluorescence Mueller matrix imaging system coupled with an appropriate analysis method for assessing anisotropy in the stromal regions of human ectocervical tissue sections. The system effectively captures detailed changes in tissue morphology and biochemistry, driven by interactions between intrinsic fluorophores and their microenvironment. Measurements from the connective tissue near the epithelium reveal significant alterations in fluorescence diattenuation and polarizance in both ground and excited states. Pronounced differences in the decomposition-derived anisotropy parameters are observed for unstained cervical tissue sections varying from healthy to different grades of cervical intraepithelial neoplasia (CIN). A significant reduction in the parameters—circular depolarization (M44 decreases from 0.65 in healthy tissue to 0.05 in CIN-III), fluorescence linear diattenuation (α L drops from 0.25 to 0.07), and fluorescence linear polarizance (β L drops from 0.27 to 0.08)—is observed due to the degradation of the collagen fibrous structure present within the stromal region of the human cervix. The polarization parameters demonstrate a linear relationship between healthy and different CIN grades as the cancer progresses. Additionally, a strong correlation was identified between α L and β L , confirmed through linear curve fitting across all sites examined at different CIN grades. This fluorescence Mueller matrix imaging technique, coupled with a suitable decomposition method, serves as a diagnostic tool that minimizes the need for extensive clinical tests. It effectively detects early stage cervical cancer by identifying significant decreases in fluorescence linear diattenuation and polarizance, offering valuable quantitative insights into tissue changes across various pre-cancerous grades.

Enhanced epithelial fluorescence sensitivity using an oblique-angled fiber optic probe for epithelial cancer diagnosis: a modified Monte Carlo simulation study

This study aims to analyze spatially resolved fluorescence for probing various depths of the epithelial tissue using an optimized illumination and collection configuration. The enhanced epithelial sensitivity enables precise detection of alterations in the tissue’s optical properties associated with disease progression. The elastic scattering and fluorescence of a cervical tissue-mimicking phantom were simulated using the Monte Carlo method, which was modified to incorporate oblique illumination, oblique collection, and the propagation of fluorescence photons. The results demonstrate that oblique illumination and collection in a parallel configuration (O-O-P) exhibit high epithelial sensitivity at both small and large source–detector separations. Additionally, spatially resolved fluorescence enables high-sensitivity probing of various depths within the epithelium layer. We modeled a fiber-optic probe employing an O-O-P configuration with 45° beveled fibers. It was observed that this multi-collection fiber-optic probe effectively differentiates cervical precancer grades by analyzing fluorescence intensity variations, which correspond to changes in tissue optical properties during disease progression.

Smartphone-based fluorescence spectroscopic device for cervical precancer diagnosis: a random forest classification of in vitro data

Cervical cancer can be treated and cured if diagnosed at an early stage. Optical devices, developed on smartphone-based platforms, are being tested for this purpose as they are cost-effective, robust, and field portable, showing good efficiency compared to the existing commercial devices. This study reports on the applicability of a 3D printed smartphone-based spectroscopic device (3D-SSD) for the early diagnosis of cervical cancer. The proposed device has the ability to evaluate intrinsic fluorescence (IF) from the collected polarized fluorescence (PF) and elastic-scattering (ES) spectra from cervical tissue samples of different grades. IF spectra of 30 cervical tissue samples have been analyzed and classified using a combination of principal component analysis (PCA) and random forest (RF)-based multi-class classification algorithm with an overall accuracy above 90%. The usage of smartphone for image collection, spectral data analysis, and display makes this device a potential contender for use in clinics as a regular screening tool.