RESUMO
The presence of molecular mutations in colorectal cancer (CRC) is a decisive factor in selecting the most effective first-line therapy. However, molecular analysis is routinely performed only in a limited number of patients with remote metastases. We propose to use tissue stiffness as a marker of the presence of molecular mutations in CRC samples. For this purpose, we applied compression optical coherence elastography (C-OCE) to calculate stiffness values in regions corresponding to specific CRC morphological patterns (n = 54). In parallel to estimating stiffness, molecular analysis from the same zones was performed to establish their relationships. As a result, a high correlation between the presence of KRAS/NRAS/BRAF driver mutations and high stiffness values was revealed regardless of CRC morphological pattern type. Further, we proposed threshold stiffness values for label-free targeted detection of molecular alterations in CRC tissues: for KRAS, NRAS, or BRAF driver mutation-above 803 kPa (sensitivity-91%; specificity-80%; diagnostic accuracy-85%), and only for KRAS driver mutation-above 850 kPa (sensitivity-90%; specificity-88%; diagnostic accuracy-89%). To conclude, C-OCE estimation of tissue stiffness can be used as a clinical diagnostic tool for preliminary screening of genetic burden in CRC tissues.
Assuntos
Biomarcadores Tumorais , Neoplasias Colorretais , Técnicas de Imagem por Elasticidade , GTP Fosfo-Hidrolases , Mutação , Proteínas Proto-Oncogênicas B-raf , Proteínas Proto-Oncogênicas p21(ras) , Humanos , Neoplasias Colorretais/genética , Neoplasias Colorretais/patologia , Neoplasias Colorretais/diagnóstico , Técnicas de Imagem por Elasticidade/métodos , Biomarcadores Tumorais/genética , Proteínas Proto-Oncogênicas B-raf/genética , Proteínas Proto-Oncogênicas p21(ras)/genética , GTP Fosfo-Hidrolases/genética , Feminino , Masculino , Elasticidade , Idoso , Proteínas de Membrana/genética , Pessoa de Meia-IdadeRESUMO
The paper is devoted to the derivation of a bidirectional distribution function for crystals, which specifies all outgoing rays for a ray coming to the boundary of two transparent crystalline media with different optical properties, i.e., a particular mineral, directions of optical axes if they exist, and other features. A local model of interaction based on the notion of polarized light ray is introduced, which is specified by a geometric ray, its polarization state, light intensity, and so on. The computational algorithm that is suggested allows computing the directions and other properties of all (up to four) outgoing rays. In this paper, isotropic, uniaxial, and biaxial crystals are processed in a similar manner. The correctness of the model is validated by comparison of photos of real uniaxial crystals with corresponding computed images. The case of biaxial crystals is validated by testing the effect of conical refraction. Specifications of a series of tests devoted to rendering of optically different objects is presented also.