Nested dictionary learning for hierarchical organization of imagery and text

A tree-based dictionary learning model is developed for joint analysis of imagery and associated text. The dictionary learning may be applied directly to the imagery from patches, or to general feature vectors extracted from patches or superpixels (using any existing method for image feature extraction). Each image is associated with a path through the tree (from root to a leaf), and each of the multiple patches in a given image is associated with one node in that path. Nodes near the tree root are shared between multiple paths, representing image characteristics that are common among different types of images. Moving toward the leaves, nodes become specialized, representing details in image classes. If available, words (text) are also jointly modeled, with a path-dependent probability over words. The tree structure is inferred via a nested Dirichlet process, and a retrospective stick-breaking sampler is used to infer the tree depth and width.

Gene selection using iterative feature elimination random forests for survival outcomes.

Although many feature selection methods for classification have been developed, there is a need to identify genes in high-dimensional data with censored survival outcomes. Traditional methods for gene selection in classification problems have several drawbacks. First, the majority of the gene selection approaches for classification are single-gene based. Second, many of the gene selection procedures are not embedded within the algorithm itself. The technique of random forests has been found to perform well in high-dimensional data settings with survival outcomes. It also has an embedded feature to identify variables of importance. Therefore, it is an ideal candidate for gene selection in high-dimensional data with survival outcomes. In this paper, we develop a novel method based on the random forests to identify a set of prognostic genes. We compare our method with several machine learning methods and various node split criteria using several real data sets. Our method performed well in both simulations and real data analysis.Additionally, we have shown the advantages of our approach over single-gene-based approaches. Our method incorporates multivariate correlations in microarray data for survival outcomes. The described method allows us to better utilize the information available from microarray data with survival outcomes.

Primary vascularization of the graft determines the immunodominance of murine minor H antigens during organ transplantation.

Grafts can be rejected even when matched for MHC because of differences in the minor histocompatibility Ags (mH-Ags). H4- and H60-derived epitopes are known as immunodominant mH-Ags in H2(b)-compatible BALB.B to C57BL/6 transplantation settings. Although multiple explanations have been provided to explain immunodominance of Ags, the role of vascularization of the graft is yet to be determined. In this study, we used heart (vascularized) and skin (nonvascularized) transplantations to determine the role of primary vascularization of the graft. A higher IFN-γ response toward H60 peptide occurs in heart recipients. In contrast, a higher IFN-γ response was generated against H4 peptide in skin transplant recipients. Peptide-loaded tetramer staining revealed a distinct antigenic hierarchy between heart and skin transplantation: H60-specific CD8(+) T cells were the most abundant after heart transplantation, whereas H4-specific CD8(+) T cells were more abundant after skin graft. Neither the tissue-specific distribution of mH-Ags nor the draining lymph node-derived dendritic cells correlated with the observed immunodominance. Interestingly, non-primarily vascularized cardiac allografts mimicked skin grafts in the observed immunodominance, and H60 immunodominance was observed in primarily vascularized skin grafts. However, T cell depletion from the BALB.B donor prior to cardiac allograft induces H4 immunodominance in vascularized cardiac allograft. Collectively, our data suggest that immediate transmigration of donor T cells via primary vascularization is responsible for the immunodominance of H60 mH-Ag in organ and tissue transplantation.

B cells in rheumatoid synovitis.

In rheumatoid arthritis, T cells, B cells, macrophages, and dendritic cells invade the synovial membranes, establishing complex microstructures that promote inflammatory/tissue destructive lesions. B cell involvement has been considered to be limited to autoantibody production. However, recent studies suggest that B cells support rheumatoid disease through other mechanisms. A critical element of rheumatoid synovitis is the process of ectopic lymphoid neogenesis, with highly efficient lymphoid architectures established in a nonlymphoid tissue site. Rheumatoid synovitis recapitulates the pathways of lymph node formation, and B cells play a key role in this process. Furthermore, studies of rheumatoid lesions implanted in immunodeficient mice suggest that T cell activation in synovitis is B cell dependent, indicating the role played by B cells in presenting antigens and providing survival signals.

What the brain stem tells the frontal cortex. II. Role of the SC-MD-FEF pathway in corollary discharge.

One way we keep track of our movements is by monitoring corollary discharges or internal copies of movement commands. This study tested a hypothesis that the pathway from superior colliculus (SC) to mediodorsal thalamus (MD) to frontal eye field (FEF) carries a corollary discharge about saccades made into the contralateral visual field. We inactivated the MD relay node with muscimol in monkeys and measured corollary discharge deficits using a double-step task: two sequential saccades were made to the locations of briefly flashed targets. To make second saccades correctly, monkeys had to internally monitor their first saccades; therefore deficits in the corollary discharge representation of first saccades should disrupt second saccades. We found, first, that monkeys seemed to misjudge the amplitudes of their first saccades; this was revealed by systematic shifts in second saccade end points. Thus corollary discharge accuracy was impaired. Second, monkeys were less able to detect trial-by-trial variations in their first saccades; this was revealed by reduced compensatory changes in second saccade angles. Thus corollary discharge precision also was impaired. Both deficits occurred only when first saccades went into the contralateral visual field. Single-saccade generation was unaffected. Additional deficits occurred in reaction time and overall performance, but these were bilateral. We conclude that the SC-MD-FEF pathway conveys a corollary discharge used for coordinating sequential saccades and possibly for stabilizing vision across saccades. This pathway is the first elucidated in what may be a multilevel chain of corollary discharge circuits extending from the extraocular motoneurons up into cerebral cortex.

Diminishment of respiratory sinus arrhythmia foreshadows doxorubicin-induced cardiomyopathy.

BACKGROUND: The development of a microcomputer-based device permits quick, simple, and noninvasive quantification of the respiratory sinus arrhythmia (RSA) during quiet breathing. METHODS AND RESULTS: We prospectively and serially measured the radionuclide left ventricular ejection fraction and the RSA amplitude in 34 cancer patients receiving up to nine monthly bolus treatments with doxorubicin hydrochloride (60 mg/m2). Of the eight patients who ultimately developed symptomatic doxorubicin-induced congestive heart failure, seven (87.5%) demonstrated a significant decline in RSA amplitude; five of 26 subjects without clinical symptoms of cardiotoxicity (19.2%) showed a similar RSA amplitude decline. On average, significant RSA amplitude decline occurred 3 months before the last planned doxorubicin dose in patients destined to develop clinical congestive heart failure. CONCLUSION: Overall, RSA amplitude abnormality proved to be a more specific predictor of clinically significant congestive heart failure than did serial resting radionuclide ejection fractions.