Radiofrequency impedance monitoring of the biological tissues under local heating by optical radiation

Abstract

Development of methods for simultaneous control of state of biological tissues during optical treatment is the important tasks in laser surgery. We introduce a novel approach for the monitoring of the state of biological tissues in the process of its local heating by optical radiation. It is based on measurements of the electrical radiofrequency impedance kinetics of the sample during irradiation. The obtained data are processed using interconnected mathematical modeling of corresponding thermodynamic, optical and electrical problems. Experimental applications of this approach, represented in the paper, reveal its high sensitivity, repeatability and consistency with the model. The introduced method can be used for the selection and optimization of radiation parameters of medical laser sources as an alternative or an addition to histological techniques. Radiofrequency impedance measurement can be used directly in the course of surgical operations for monitoring the treated tissues state, including its temperature and degree of damage.

Deep self‐supervised transformation learning for leukocyte classification

“…Leukocyte image classification plays an important role in clinical medicine, which is an essential branch in the field of medical image research. Early artificial representation-based methods [8,[24][25][26][27][28][29][30] expect to extract discriminative features of leukocytes by a well-designed strategy, and then the features are regarded as input to a classifier to distinguish leukocyte categories. For example, Adjouadi et al [31] utilize an SVM classifier to discriminate the features of flow cytometry data provided by Beckman-Coulter Corporation.…”

“…However, these methods depend on the pre-processing of cell segmentation, and errors in the process will have a negative impact on the classification of cells. Recently, researchers turn to deep learning methods [5][6][7][8][9][10][11] that use the model called neural networks to learn representations from data without any pre-processing and handcrafted features. Especially, convolutional neural network (CNN) [12], as a fast developing deep learning model, has been widely applied in leukocyte classification.…”