Development of an Anthropomorphic Phantom of the Axillary Region for Microwave Imaging Assessment.

We produced an anatomically and dielectrically realistic phantom of the axillary region to enable the experimental assessment of Axillary Lymph Node (ALN) imaging using microwave imaging technology. We segmented a thoracic Computed Tomography (CT) scan and created a computer-aided designed file containing the anatomical configuration of the axillary region. The phantom comprises five 3D-printed parts representing the main tissues of interest of the axillary region for the purpose of microwave imaging: fat, muscle, bone, ALNs, and lung. The phantom allows the experimental assessment of multiple anatomical configurations, by including ALNs of different size, shape, and number in several locations. Except for the bone mimicking organ, which is made of solid conductive polymer, we 3D-printed cavities to represent the fat, muscle, ALN, and lung and filled them with appropriate tissue-mimicking liquids. Existing studies about complex permittivity of ALNs have reported limitations. To address these, we measured the complex permittivity of both human and animal lymph nodes using the standard open-ended coaxial-probe technique, over the 0.5 GHz-8.5 GHz frequency band, thus extending current knowledge on dielectric properties of ALNs. Lastly, we numerically evaluated the effect of the polymer which constitutes the cavities of the phantom and compared it to the realistic axillary region. The results showed a maximum difference of 7 dB at 4 GHz in the electric field magnitude coupled to the tissues and a maximum of 10 dB difference in the ALN response. Our results showed that the phantom is a good representation of the axillary region and a viable tool for pre-clinical assessment of microwave imaging technology.

Confident texture-based laryngeal tissue classification for early stage diagnosis support.

Early stage diagnosis of laryngeal squamous cell carcinoma (SCC) is of primary importance for lowering patient mortality or after treatment morbidity. Despite the challenges in diagnosis reported in the clinical literature, few efforts have been invested in computer-assisted diagnosis. The objective of this paper is to investigate the use of texture-based machine-learning algorithms for early stage cancerous laryngeal tissue classification. To estimate the classification reliability, a measure of confidence is also exploited. From the endoscopic videos of 33 patients affected by SCC, a well-balanced dataset of 1320 patches, relative to four laryngeal tissue classes, was extracted. With the best performing feature, the achieved median classification recall was 93% [interquartile range [Formula: see text]]. When excluding low-confidence patches, the achieved median recall was increased to 98% ([Formula: see text]), proving the high reliability of the proposed approach. This research represents an important advancement in the state-of-the-art computer-assisted laryngeal diagnosis, and the results are a promising step toward a helpful endoscope-integrated processing system to support early stage diagnosis.