RESUMEN
Profiling cellular phenotypes from microscopic imaging can provide meaningful biological information resulting from various factors affecting the cells. One motivating application is drug development: morphological cell features can be captured from images, from which similarities between different drug compounds applied at different doses can be quantified. The general approach is to find a function mapping the images to an embedding space of manageable dimensionality whose geometry captures relevant features of the input images. An important known issue for such methods is separating relevant biological signal from nuisance variation. For example, the embedding vectors tend to be more correlated for cells that were cultured and imaged during the same week than for those from different weeks, despite having identical drug compounds applied in both cases. In this case, the particular batch in which a set of experiments were conducted constitutes the domain of the data; an ideal set of image embeddings should contain only the relevant biological information (e.g., drug effects). We develop a general framework for adjusting the image embeddings in order to "forget" domain-specific information while preserving relevant biological information. To achieve this, we minimize a loss function based on distances between marginal distributions (such as the Wasserstein distance) of embeddings across domains for each replicated treatment. For the dataset we present results with, the only replicated treatment happens to be the negative control treatment, for which we do not expect any treatment-induced cell morphology changes. We find that for our transformed embeddings (i) the underlying geometric structure is not only preserved but the embeddings also carry improved biological signal; and (ii) less domain-specific information is present.
RESUMEN
We investigate central issues such as invertibility, stability, synchronization, and frequency characteristics for nonlinear wavelet transforms built using the lifting framework. The nonlinearity comes from adaptively choosing between a class of linear predictors within the lifting framework. We also describe how earlier families of nonlinear filter banks can be extended through the use of prediction functions operating on a causal neighborhood of pixels. Preliminary compression results for model and real-world images demonstrate the promise of our techniques.
RESUMEN
OBJECTIVES: To determine the clinical utility of upper endoscopy in patients who have upper gastrointestinal bleeding after hospitalization. METHODS: Patients were studied who underwent upper endoscopy for an indication of suspected upper gastrointestinal bleeding that developed more than 48 hours after hospitalization. Demographic, clinical, and endoscopic data were extracted by chart review. Bleeding was characterized as clinically important (defined as overt bleeding in association with hemodynamic compromise or the need for blood transfusion) or non-clinically important. RESULTS: Eighty-six patients met inclusion criteria. Clinically important bleeding occurred in 17%. Peptic ulcer disease and gastritis were the most common sources of bleeding in the clinically important and non-clinically important groups, respectively. The bleeding source was not found in 24% of patients. Endoscopic therapy was required in 11% (all of whom had clinically important bleeding). Upper endoscopy prompted no treatment changes in the non-clinically important bleeding group. CONCLUSIONS: Endoscopic therapy was needed only in the few patients with clinically important bleeding. Nonendoscopic treatment can be recommended for upper gastrointestinal bleeding developing in hospitalized patients who do not meet established criteria for a clinically important bleed.