Polarization-Based Histopathology Classification of Ex Vivo Colon Samples Supported by Machine Learning

Abstract

In biophotonics, novel techniques and approaches are being constantly sought to assist medical doctors and to increase both sensitivity and specificity of the existing diagnostic methods. In such context, tissue polarimetry holds promise to become a valuable optical diagnostic technique as it is sensitive to tissue alterations caused by different benign and malignant formations. In our studies, multiple Mueller matrices were recorded for formalin-fixed, human, ex vivo colon specimens containing healthy and tumor zones. The available data were pre-processed to filter noise and experimental errors, and then all Mueller matrices were decomposed to derive polarimetric quantities sensitive to malignant formations in tissues. In addition, the Poincaré sphere representation of the experimental results was implemented. We also used the canonical and natural indices of polarimetric purity depolarization spaces for plotting our experimental data. A feature selection was used to perform a statistical analysis and normalization procedure on the available data, in order to create a polarimetric model for colon cancer assessment with strong predictors. Both unsupervised (principal component analysis) and supervised (logistic regression, random forest, and support vector machines) machine learning algorithms were used to extract particular features from the model and for classification purposes. The results from logistic regression allowed to evaluate the best polarimetric quantities for tumor detection, while the use of random forest yielded the highest accuracy values. Attention was paid to the correlation between the predictors in the model as well as both losses and relative risk of misclassification. Apart from the mathematical interpretation of the polarimetric quantities, the presented polarimetric model was able to support the physical interpretation of the results from previous studies and relate the latter to the samples’ health condition, respectively.

A Dual-Modality Imaging Method Based on Polarimetry and Second Harmonic Generation for Characterization and Evaluation of Skin Tissue Structures

“…Mueller matrix polarimetry has recently gained popularity in biomedical research due to its high potential for detecting the microstructural and optical properties of label-free samples [ 16 , 17 , 18 , 19 , 20 , 21 ]. Studies have demonstrated that Mueller matrix-derived parameters are sensitive to different structures and have significant application prospects in the pathological diagnosis of various abnormal tissues, such as skin cancer [ 22 ], breast cancer [ 23 , 24 ], liver cirrhosis and cancer [ 25 , 26 ], thyroid cancer [ 27 ], colon cancer [ 28 , 29 , 30 ], cervical cancer [ 31 , 32 , 33 ], and lung cancer [ 34 ]. Mueller matrix polarimetry can also be used to visualize brain white matter fiber tracts [ 35 ], explain the combined effects of scattering and absorption changes in cancer growth [ 36 ], and assess β-amyloidosis in Alzheimer’s disease [ 37 ].…”

A quantitative technique to analyze and evaluate microstructures of skin hair follicles based on mueller matrix polarimetry

“…Also, traditional optical equipment such as microscopes and endoscopes can be conveniently upgraded to the Mueller matrix microscope and endoscope by adding a polarization state generator (PSG) and analyzer (PSA) to their optical path [14]. Recent studies have demonstrated that the Mueller matrix microscope can be used to acquire structural characteristics of various pathological tissue samples and assist in diagnosis of skin cancer [15], liver fibrosis and cancer [16,17], cervical cancer [18][19][20], colon cancer [21], breast cancer [22][23][24], inflammatory bowel diseases [25], and screening and evaluating β-amyloidosis in Alzheimer's disease [26]. Especially, Mueller matrix polarimetry is suitable for detection of anisotropic structures, such as muscle fibers, collagen fibers, and elastin fibers, with prominent linear birefringence properties [11,13,27,28].…”

“…Combined with image segmentation, sliding window algorithm, and image texture analysis methods, it has great potential in skin structurerelated studies and clinical dermatological applications. Recently, the transmission Mueller matrix microscopy has been prevalently used for structural detection of various thin tissue samples [15][16][17][18][19][20][21][22][23][24][25]. The observation of histological tissue slices stained with a certain dye, which is the gold standard of pathological diagnosis, can help us extract groups of Mueller matrix parameters associated with microstructural features of skin hair follicles.…”

Differentiation of Human GBM From Non-GBM Brain Tissue With Polarization Imaging Technique

“…In particular, tissue polarimetry was used to effectively analyze how the myocardial tissues were affected by disease for definitive diagnostics in forensic medicine ( 31 ). Machine learning algorithms were used to extract particular features for human ex vivo colon specimen classification between healthy and tumor zones ( 32 ). As for the nervous system, PIT combined with optical correlation tomography (OCT) has been reported to distinguish brain white matter and glioma tissues ( 33 ), which confirmed the feasibility of PIT being clinically applied to detect glioma.…”