Polarimetric imaging for cervical pre-cancer screening aided by machine learning: ex vivo studies

Abstract

.SignificanceWide-field imaging Mueller polarimetry is an optical imaging technique that has great potential to become a reliable, fast, non-contact in vivo imaging modality for the early detection of, e.g., deceases and tissue structural malformations, such as cervical intraepithelial neoplasia, in both clinical and low-resource settings. On the other hand, machine learning methods have established themselves as a superior solution in image classification and regression tasks. We combine Mueller polarimetry and machine learning, critically assess the data/classification pipeline, investigate the bias arising from training strategies, and demonstrate how higher levels of detection accuracy can be achieved.AimWe aim to automate/assist with diagnostic segmentation of polarimetric images of uterine cervix specimens.ApproachA comprehensive capture-to-classification pipeline is developed in house. Specimens are acquired and measured with imaging Mueller polarimeter and undergo histopathological classification. Subsequently, a labeled dataset is created within tagged regions of either healthy or neoplastic cervical tissues. Several machine learning methods are trained utilizing different training-test-set-split strategies, and their corresponding accuracies are compared.ResultsOur results include robust measurements of model performance with two approaches: a 90:10 training–test-set-split and leave-one-out cross-validation. By comparing the classifier’s accuracy directly with the ground truth obtained during histology analysis, we demonstrate how conventionally used shuffled split leads to an over-estimate of true classifier performance ( 0.964 ± 0.00 ) . The leave-one-out cross-validation, however, leads to more accurate performance ( 0.812 ± 0.21 ) with respect to newly obtained samples that were not used to train the models.ConclusionsCombination of Mueller polarimetry and machine learning is a powerful tool for the task of screening for pre-cancerous conditions in cervical tissue sections. Nevertheless, there is a inherent bias with conventional processes that can be addressed using more conservative classifier training approaches. This results in overall improvements of the sensitivity and specificity of the developed techniques for “unseen” images.

Assessment of the Impact of Nanowarming on Microstructure of Cryopreserved Fibroblast-Containing 3D Tissue Models Using Mueller Polarimetry

“…The minimum value of the entropy would be expected when all pixel values for each patch are equal, thus indicating the homogeneity of the measured medium. On the other hand, the maximum value of the entropy for a given image patch would be achieved when the image pixel values are uniformly distributed within the closed interval [0, 2 16 ], indicating the random fluctuations of the medium's properties.…”

“…In general, light-tissue interactions have been well studied, and using polarized light as a tool for the analysis of the inner structure of biological tissue has found applications in tissue characterization and the detection of both benign and malignant lesions [13][14][15][16][17][18][19][20][21]. Initially, the membrane integrity brake, caused by cryo-injury, is anticipated to be followed by changes in the tissue structure and its morphology [10,22,23].…”

Polarization-Based Digital Histology of Human Skin Biopsies Assisted by Deep Learning

“…Polarization-based optical modalities are sensitive to the micro-structure of biological tissue and have found its place in pre-clinical and clinical applications [1][2][3][4][5]. It was shown that polarized light can be used for non-invasive diagnosis of benign and malignant lesions [6][7][8][9][10][11][12]. Along with diffuse reflectance, fluorescence, Raman spectroscopy and other optical techniques, tissue polarimetry can be used as label-free modality providing supplementary information for skin diagnosis [10,13].…”