A new approach to the classification of mammographic masses and normal breast tissue

A new approach to mammographic mass detection is presented in this paper. Although different algorithms have been proposed for such a task, most of them are application dependent. In contrast, our approach makes use of a kindred topic in computer vision adapted to our particular problem. In this sense, we translate the eigenfaces approach for face detection/classification problems to a mass detection. Two different databases were used to show the robustness of the approach. The first one consisted on a set of 160 regions of interest (RoIs) extracted from the MIAS database, being 40 of them with confirmed masses and the rest normal tissue. The second set of RoIs was extracted from the DDSM database, and contained 196 RoIs containing masses and 392 with normal, but suspicious regions. Initial results demonstrate the feasibility of using such approach with performances comparable to other algorithms, with the advantage of being a more general, simple and cost-effective approach

Modeling and Classifying Breast Tissue Density in Mammograms

We present a new approach to model and classify breast parenchymal tissue. Given a mammogram, first, we will discover the distribution of the different tissue densities in an unsupervised manner, and second, we will use this tissue distribution to perform the classification. We achieve this using a classifier based on local descriptors and probabilistic Latent Semantic Analysis (pLSA), a generative model from the statistical text literature. We studied the influence of different descriptors like texture and SIFT features at the classification stage showing that textons outperform SIFT in all cases. Moreover we demonstrate that pLSA automatically extracts meaningful latent aspects generating a compact tissue representation based on their densities, useful for discriminating on mammogram classification. We show the results of tissue classification over the MIAS and DDSM datasets. We compare our method with approaches that classified these same datasets showing a better performance of our proposal

HIV-1 immune activation induces siglec-1 expression and enhances viral trans-infection in blood and tissue myeloid cells

Background: Myeloid cells are key players in the recognition and response of the host against invading viruses. Paradoxically, upon HIV-1 infection, myeloid cells might also promote viral pathogenesis through trans-infection, a mechanism that promotes HIV-1 transmission to target cells via viral capture and storage. The receptor Siglec-1 (CD169) potently enhances HIV-1 trans-infection and is regulated by immune activating signals present throughout the course of HIV-1 infection, such as interferon α (IFNα). Results: Here we show that IFNα-activated dendritic cells, monocytes and macrophages have an enhanced ability to capture and trans-infect HIV-1 via Siglec-1 recognition of viral membrane gangliosides. Monocytes from untreated HIV-1-infected individuals trans-infect HIV-1 via Siglec-1, but this capacity diminishes after effective antiretroviral treatment. Furthermore, Siglec-1 is expressed on myeloid cells residing in lymphoid tissues, where it can mediate viral trans-infection. Conclusions: Siglec-1 on myeloid cells could fuel novel CD4+ T-cell infections and contribute to HIV-1 dissemination in vivo.

Comparative performance of the stable isotope signatures of carbon, nitrogen and oxygen in assessing early vigour and grain yield in durum wheat

The present paper studied the performance of the stable isotope signatures of carbon (δ13C), nitrogen (δ15N) and oxygen (δ18O) in plants when used to assess early vigour and grain yield (GY) in durum wheat growing under mild and moderate Mediterranean stress conditions. A collection of 114 recombinant inbred lines was grown under rainfed (RF) and supplementary irrigation (IR) conditions. Broad sense heritabilities (H2) for GY and harvest index (HI) were higher under RF conditions than under IR. Broad sense heritabilities for δ13C were always above 0·60, regardless of the plant part studied, with similar values for IR and RF trials. Some of the largest genetic correlations with GY were those shown by the δ13C content of the flag leaf blade and mature grains. Under both water treatments, mature grains showed the highest negative correlations between δ13C and GY across genotypes. Flag leaf δ13C was negatively correlated with GY only under RF conditions. The δ13C in seedlings was negatively correlated, under IR conditions only, with GY but also with early vigour. The sources of variation in early vigour were studied by stepwise analysis using the stable isotope signatures measured in seedlings. The δ13C was able to explain almost 0·20 of this variation under RF, but up to 0·30 under IR. In addition, nitrogen concentration in seedlings accounted for another 0·05 of variation, increasing the amount explained to 0·35. The sources of variation in GY were also studied through stable isotope signatures and biomass of different plant parts: δ13C was always the first parameter to appear in the models for both water conditions, explaining c. 0·20 of the variation. The second parameter (δ15N or N concentration of grain, or biomass at maturity) depended on the water conditions and the plant tissue being analysed. Oxygen isotope composition (δ18O) was only able to explain a small amount of the variation in GY. In this regard, despite the known and previously described value of δ13C as a tool in breeding, δ15N is confirmed as an additional tool in the present study. Oxygen isotope composition does not seem to offer any potential, at least under the conditions of the present study.