Margarida Sampaio Fernandes

The Curious Case of Pseudocysts in Breast Cancer Specimens: An Alteration Associated With Hydrogel-Based Breast Biopsy Markers:

Communication between pathologists and clinicians is of the utmost importance for an adequate interpretation of histopathological findings. In early 2018, we started to observe, in breast cancer specimens of patients submitted to neoadjuvant treatment or with small or mammographic detected lesions, an unusual pathological finding. It consisted of pseudocystic lesions, lined by epithelioid histiocytes and multinucleate foreign body giant cells (Figure 1). The pseudocyst contained a basophilic substance, resembling mucinous material. The wall had minimal fibrosis and histiocytic infiltrate, sometimes with abundant hemosiderin pigment (Figure 1). It was common to find such lesions in close proximity to neoplastic lesions (carcinoma in situ, invasive carcinoma) or lymph nodes (Figure 2). Due to these unusual findings, we discussed with a surgeon of our multidisciplinary team what type of material was being used, that could be related to those alterations. Together, and after literature research, we came to the conclusion that such pseudocysts corresponded to a foreign body reaction to hydrogel-based, detectable breast biopsy markers. The histological pseudomucinous content, observed within the pseudocysts, correlates to the hydrogel component. These sonographically detectable devices, used to mark the lesion(s) for resection, have been recently introduced at our hospital. They are composed of a biodegradable hydrogel polymer and a central permanent metal marker, which accurately mark the breast lesion (HydroMark; Figure 3). The very small dimension of the device might be the reason for the failure of macroscopic detection in most cases. 769377 IJSXXX10.1177/1066896918769377International Journal of Surgical PathologyMarques et al research-article2018

A Comparative Analysis of Deep and Shallow Features for Multimodal Face Recognition in a Novel RGB-D-IR Dataset

With new trends like 3D and deep learning alternatives for face recognition becoming more popular, it becomes essential to establish a complete benchmark for the evaluation of such algorithms, in a wide variety of data sources and non-ideal scenarios. We propose a new RGB-depth-infrared (RGB-D-IR) dataset, RealFace, acquired with the novel Intel® RealSense™ collection of sensors, and characterized by multiple variations in pose, lighting and disguise. As baseline for future works, we assess the performance of multiple deep and “shallow” feature descriptors. We conclude that our dataset presents some relevant challenges and that deep feature descriptors present both higher robustness in RGB images, as well as an interesting margin for improvement in alternative sources, such as depth and IR.