Establishing basal cell carcinoma (BCC) subtype is sometimes challenging for pathologists. Deep-learning (DL) algorithms are an emerging approach in image classification due to their performance, accompanied by a new concept -transfer learning, which implies replacing the final layers of a trained network and retraining it for a new task, while keeping the weights from the imported layers. A DL convolution-based software, capable of classifying 10 subtypes of BCC, was designed. Transfer learning from three general-purpose image classification networks (AlexNet, GoogLeNet, and ResNet-18) was used. Three pathologists independently labeled 2249 patches. Ninety percent of data was used for training and 10% for testing on 100 independent training sequences. Each of the resulted networks independently labeled the whole dataset. Mean and standard deviation (SD) accuracy (ACC) [%]/sensitivity (SN) [%]/specificity (SP) [%]/area under the curve (AUC) for all the networks was 82.53±2.63/72.52±3.63/97.94±0.3/0.99. The software was validated on another 50-image dataset, and its results are comparable with the result of three pathologists in terms of agreement. All networks had similar classification accuracies, which demonstrated that they reached a maximum classification rate on the dataset. The software shows promising results, and with further development can be successfully used on histological images, assisting pathologists' diagnosis and teaching.

Two deep-learning algorithms designed to classify images according to the Gleason grading system that used transfer learning from two well-known general-purpose image classification networks (AlexNet and GoogleNet) were trained on Hematoxylin-Eosin histopathology stained microscopy images with prostate cancer. The dataset consisted of 439 images asymmetrically distributed in four Gleason grading groups. Mean and standard deviation accuracy for AlexNet derivate network was of 61.17±7 and for GoogleNet derivate network was of 60.9±7.4. The similar results obtained by the two networks with very different architecture, together with the normal distribution of classification error for both algorithms show that we have reached a maximum classification rate on this dataset. Taking into consideration all the constraints, we conclude that the resulted networks could assist pathologists in this field, providing first or second opinions on Gleason grading, thus presenting an objective opinion in a grading system which has showed in time a great deal of interobserver variability.

Basal cell carcinoma (BCC) is the most frequent cancer of the skin and comprises low-risk and high-risk subtypes. We selected a low-risk subtype, namely, nodular (N), and a high-risk subtype, namely, micronodular (MN), with the aim to identify differences between them using a classical morphometric approach through a gray-level co-occurrence matrix and histogram analysis, as well as an approach based on deep learning semantic segmentation. From whole-slide images, pathologists selected 216 N and 201 MN BCC images. The two groups were then manually segmented and compared based on four morphological areas: center of the BCC islands (tumor, T), peripheral palisading of the BCC islands (touching tumor, TT), peritumoral cleft (PC) and surrounding stroma (S). We found that the TT pattern varied the least, while the PC pattern varied the most between the two subtypes. The combination of two distinct analysis approaches yielded fresh insights into the characterization of BCC, and thus, we were able to describe two different morphological patterns for the T component of the two subtypes.

Background: Paraganglioma is a rare neuroendocrine tumor derived from chromaffin cells. The overproduction of catecholamines accounts for the presenting symptoms and cardiovascular complications. The clinical presentation frequently overlaps with the associated cardiac diseases, delaying the diagnosis. Multimodality imaging and a multidisciplinary team are essential for the correct diagnosis and adequate clinical management. Case Summary: A 37-year-old woman with a personal medical history of long-standing arterial hypertension and radiofrequency ablation for atrioventricular nodal reentry tachycardia presented with progressive exertional dyspnea and elevated blood pressure values, despite a comprehensive pharmacological treatment with six antihypertensive drugs. The echocardiography showed a bicuspid aortic valve and severe aortic regurgitation. The computed tomography angiography revealed a retroperitoneal space-occupying solid lesion, with imaging characteristics suggestive of a paraganglioma. The multidisciplinary team concluded that tumor resection should be completed first, followed by an aortic valve replacement if necessary. The postoperative histopathology examination confirmed the diagnosis of paraganglioma. After the successful resection of the tumor, the patient was asymptomatic, and the intervention for aortic valve replacement was delayed. Discussion: This was a rare case of a late-detected paraganglioma in a young patient with resistant hypertension overlapping the clinical presentation and management of severe aortic regurgitation. A multimodality imaging approach including transthoracic and transesophageal echocardiography, computed tomography, and magnetic resonance imaging had an emerging role in establishing the diagnosis and in guiding patient management and follow-up. The resection of paraganglioma was essential for the optimal timing of surgical correction for severe aortic regurgitation. We further reviewed various cardiovascular complications induced by pheochromocytomas and paragangliomas.