Synchronous Multiple Breast Cancers-Do We Need to Reshape Staging?

Background and Objectives: Current recommendations and treatment regimens in breast cancer are a reflection of its heterogeneity on multiple levels including histological subtypes, grading, molecular profiling, and numerous prognostic indices. Although based on extensive research, current guidelines are not explicit in the case of surgical specimens showing various degrees of mismatch between different parts of the same tumor and even more so between multicentric lesions. Synchronous breast cancer is the ideal prototype for studying inter- and intra-tumoral heterogeneity, therefore we envisaged that a study on patients with multicentric and multifocal lesions could contribute to the reshaping of the staging, prognosis, and treatment of breast malignancies. Material and Methods: A prospective observational study was conducted between January 2013 and May 2017 on 235 patients diagnosed with breast cancer (BC) and surgically treated at Emergency University Hospital, Bucharest. Thirty-seven patients had multiple breast tumors and were eligible for assessment of the heterogeneity of their lesions. Results: 6 were multicentric and 31 multifocal. The number of foci varied from 2 to 11. We encountered numerous mismatches between the index and the secondary tumors, as follows: 3 cases (8.1%) with histopathological mismatch, 13 (35.1%) with different grades of differentiation, 11 (29.8%) with ER (Estrogen Receptors) status mismatch, 12 (32.4%) with PR (Progesterone Receptors) status mismatch, 8 (21.6%) with molecular phenotype mismatch, and 17 (45.9%) cases with variable Ki-67. After careful analysis of index and secondary tumors, apart from the mismatches reported above, we discovered that the secondary tumors were actually dominant in 5 cases (13.5%), and therefore at least those cases had to be reclassified/restaged, as the supplementary data commanded changes in the therapeutic decision. Conclusions: For synchronous breast tumors, the current Tumor-Node-Metastasis (TNM) staging system ignores not only the histopathological and immunohistochemical characteristics of the secondary foci, but also their size. When secondary lesions are more aggressive or their cumulative mass is significantly bigger than that of the index tumor, the treatment plan should be adapted accordingly. We believe that information obtained from examining secondary foci in synchronous breast cancer and assessment of the cumulative tumoral mass should be reflected in the final staging and definitive treatment. The clinical benefit of staging the patients based on the most aggressive tumor and the cumulative tumoral burden rather than according to the biggest single tumor, will avoid under-treatment in cases with multifocal/multicentric BC displaying intertumoral mismatch.

Intelligent system based on multiple networks for accurate ovarian tumor semantic segmentation.

Ovarian tumors, especially malignant ones, represent a global concern, with increased prevalence in recent years. More accurate medical support systems are urgently needed to support medical staff in obtaining an efficient ovarian tumors diagnosis since detection in early stages could lead to immediately applying appropriate treatment, and implicitly improving the survival rate. The current paper aims to demonstrate that more accurate systems could be designed by combining different convolutional neural networks using different custom combination approaches and by selecting the appropriate networks to be involved in the ensemble model to achieve the best performance metrics. It is essential to understand if combining all experimented networks or only the best-performing ones could always lead to the most effective results or not. The current paper is structured in three main phases. The first step is to propose the individual networks involved in the experiments. Five DeepLab-V3+ networks with different encoders (ResNet-18, ResNet-50, MobileNet-V2, InceptionResNet-V2, and Xception) were used. In the second step, the paper proposes a custom algorithm to combine multiple individual semantic segmentation networks, while the last step describes the iterative selection approach for selecting all individual networks to be combined so that the most accurate ensemble is obtained. The system performing semantic segmentation for different types of ovarian tumors, covering both benign and malignant ones, achieved 91.18 % Intersection over union (IoU), thus overperforming all individual networks. The proposed method could be extended so that more powerful deep learning models could be used.

Dietary behavior during pregnancy.

During pregnancy, maternal diet is a modifiable factor that impacts the birth outcome. Since the nutritional needs of a pregnant women vary during preconception, gestational and breastfeeding period, it is necessary to adapt the diet and lifestyle, optimally under the personalized nutrition guidance of a specialist. High quality research regarding diet during pregnancy remains challenging as nutritional concerns also vary according to religion, financial income, age and education of the pregnant woman, as well as specific traditions of each country, limited number of dietitians with special training in maternal nutrition. The obstetrician usually plays a major role in dietary counselling and the use of nutritional monitoring tools can help identify pregnant women who may be at risk due to an inappropriate diet. The age of the pregnant women can be important when it comes to dietary and lifestyle changes. In this context, we addressed a questionnaire to 100 pregnant women in third trimester of pregnancy in order to evaluate a possible correlation between the age and the dietary behavior and lifestyle of a pregnant woman. Furthermore, the maternal compliance with dietary recommendations in pregnancy and the exposure to various risks due to unhealthy nutrition were also analysed in the study.