Classification of rat mammary carcinoma with large scale in vivo microwave measurements

Abstract

Mammary carcinoma, breast cancer, is the most commonly diagnosed cancer type among women. Therefore, potential new technologies for the diagnosis and treatment of the disease are being investigated. One promising technique is microwave applications designed to exploit the inherent dielectric property discrepancy between the malignant and normal tissues. In theory, the anomalies can be characterized by simply measuring the dielectric properties. However, the current measurement technique is error-prone and a single measurement is not accurate enough to detect anomalies with high confidence. This work proposes to classify the rat mammary carcinoma, based on collected large-scale in vivo S$$_{11}$$ 11 measurements and corresponding tissue dielectric properties with a circular diffraction antenna. The tissues were classified with high accuracy in a reproducible way by leveraging a learning-based linear classifier. Moreover, the most discriminative S$$_{11}$$ 11 measurement was identified, and to our surprise, using the discriminative measurement along with a linear classifier an 86.92% accuracy was achieved. These findings suggest that a narrow band microwave circuitry can support the antenna enabling a low-cost automated microwave diagnostic system.

Dielectric Characterization of Ex-Vivo Breast Tissues: Differentiation of Tumor Types through Permittivity Measurements

“…However, no studies have shown histological variations in mammary tumors. At a fundamental level, this contrast is largely due to water, as the predominant adipose tissue has much lower dielectric properties than tumors with higher water content [23][24][25][26]. Nevertheless, more complex analyses have suggested that mechanisms such as bound water effects could also play a role in specific frequency ranges [24][25][26][27][28].…”

“…At a fundamental level, this contrast is largely due to water, as the predominant adipose tissue has much lower dielectric properties than tumors with higher water content [23][24][25][26]. Nevertheless, more complex analyses have suggested that mechanisms such as bound water effects could also play a role in specific frequency ranges [24][25][26][27][28]. Although characterization of breast tumor tissue has been carried out, histological classification of these tissues has not been performed.…”

Dielectric Characterization of Breast Biopsied Tissues as Pre-Pathological Aid in Early Cancer Detection: A Blinded Feasibility Study

“…Extensive literature supports the distinction between healthy and cancerous tissues based on their dielectric properties, wherein cancerous tissues exhibit higher dielectric properties primarily attributed to increased water content [ 20 , 21 , 22 , 23 ]. Additionally, some research focused on the application of the open-ended coaxial probe for the dielectric characterization of healthy and malignant breast tissues, including also in vivo measurements in animal models [ 18 , 24 , 25 ]. Lazebnik et al conducted a large-scale study of the dielectric properties of the breast tissues obtained both from cancer surgeries and breast reduction surgeries, utilizing a 0.5–20 GHz open-ended coaxial probe technique.…”

Deep Transfer Learning to Classify Mass and Calcification Pathologies from Screen Film Mammograms

“…Automatic detection of a tumor tissue from mammography is related to texture analysis, and many different approaches have been investigated to date by new textural feature definitions and classifier models [12], such as the spherical wavelet transform [13] and geometric and textural feature extraction [14]. Computer-aided breast cancer research has also been conducted by other imaging modalities, such as microwave applications [15]. Despite these and other successful results in the literature, feature extraction-based machine learning methods can be time-consuming, particularly for medical image-or video-based analysis.…”