Predicting neurologic deterioration in patients with cerebellar hematomas

BACKGROUND: Patients with cerebellar hematomas may appear stable but may worsen suddenly. Whether certain clinical or CT scan findings predict worsening is not known. METHODS: We reviewed clinical and neuroimaging data in 72 patients with cerebellar hematomas at the Mayo Clinic from 1973 through 1993 to identify predictive features for neurologic deterioration. Patients presenting in coma and patients with vascular malformations or malignancies were excluded. Data were analyzed using chi-square or Fisher's exact test, with calculation of odds ratios with 95% confidence intervals. Multivariate logistic regression analysis was performed on appropriate variables. RESULTS: Thirty-three patients (46%) deteriorated, with a decrease in level of consciousness, new brainstem signs, or worsened motor response on the Glasgow Coma Scale. Clinical and neuroradiologic predictors for neurologic deterioration at p < 0.05 were admission systolic blood pressure greater than 200 mm Hg, pinpoint pupils and abnormal corneal or oculocephalic reflexes, hemorrhage extending into the vermis, hematoma size more than 3 cm in diameter, brainstem distortion, intraventricular hemorrhage, upward herniation, and acute hydrocephalus. Multivariate analysis demonstrated that hemorrhage located in the vermis (p = 0.03) and acute hydrocephalus (p = 0.0006) on admission CT scanning independently predicted deterioration. CONCLUSION: Patients with a cerebellar vermian hematoma or acute hydrocephalus are at high risk for neurologic deterioration. These patients should be carefully monitored and are more likely to require consideration for neurosurgical intervention.     

EXTENSION OF THE RELATIONAL DATABASE AND ITS ALGEBRA WITH ROUGH SET TECHNIQUES

This paper describes a database model based on the original rough sets theory. Its rough relations permit the representation of a rough set of tuples not definable in terms of the elementary classes, except through use of lower and upper approximations. The rough relational database model also incorporates indiscernibility in the representation and in all the operators of the rough relational algebra. This indiscernibility is based strictly on equivalence classes which must be defined for every attribute domain. There are several obvious applications for which the rough relational database model can more accurately model an enterprise than does the standard relational model. These include systems involving ambiguous, imprecise, or uncertain data. Retrieval over mismatched domains caused by the merging of one or more applications can be facilitated by the use of indiscernibility, and naive system users can achieve greater recall with the rough relational database. In addition, applications inherently “rough” could be more easily implemented and maintained in the rough relational database.     

Networking,Storytelling and Knowledge Production in First-Year Writing

As writing teachers, we seek to validate the knowledge that students bring to the classroom while introducing and cultivating new methods, strategies, techniques and technologies that may serve them now and in the future. In this article we describe an approach we take that integrates curricular and extracurricular co-apprenticeship strategies using public writing models in conjunction with social networking tools. With publication as a final goal, we urge students to move from expressing personal problems to addressing social issues, using a private, non-commercial learning network NoDiff <http://NoDiff.com> as a safe zone for skills development, knowledge production and social inquiry. In taking this approach, we resist dichotomies that place professional skills training in strict opposition to critical pedagogy. We argue that many of the so-called “extracurricular” composing activities that most of today's students engage in lay the groundwork for both professional and civic participation. Acknowledging that we as composition teachers are also beginners in the world of 21^st century communication, we emphasize both (1) the continual transference of skills and knowledge among teachers and learners; and (2) the importance of providing what is often lacking in skills-based ‘new literacy’ training, namely a critical pedagogical approach to learning as social engagement and critical intervention.     

Learning hierarchical probabilistic logic programs

Machine Learning - Probabilistic logic programming (PLP) combines logic programs and probabilities. Due to its expressiveness and simplicity, it has been considered as a powerful tool for learning...     

The role of visual complexity and prototypicality regarding first impression of websites: Working towards understanding aesthetic judgments

This paper experimentally investigates the role of visual complexity (VC) and prototypicality (PT) as design factors of websites, shaping users' first impressions by means of two studies. In the first study, 119 screenshots of real websites varying in VC (low vs. medium vs. high) and PT (low vs. high) were rated on perceived aesthetics. Screenshot presentation time was varied as a between-subject factor (50 ms vs. 500 ms vs. 1000 ms). Results reveal that VC and PT affect participants' aesthetics ratings within the first 50 ms of exposure. In the second study presentation times were shortened to 17, 33 and 50 ms. Results suggest that VC and PT affect aesthetic perception even within 17 ms, though the effect of PT is less pronounced than the one of VC. With increasing presentation time the effect of PT becomes as influential as the VC effect. This supports the reasoning of the information-processing stage model of aesthetic processing (Leder et al., 2004), where VC is processed at an earlier stage than PT. Overall, websites with low VC and high PT were perceived as highly appealing.     

The HaLoop approach to large-scale iterative data analysis

The growing demand for large-scale data mining and data analysis applications has led both industry and academia to design new types of highly scalable data-intensive computing platforms. MapReduce has enjoyed particular success. However, MapReduce lacks built-in support for iterative programs, which arise naturally in many applications including data mining, web ranking, graph analysis, and model fitting. This paper (This is an extended version of the VLDB 2010 paper “HaLoop: Efficient Iterative Data Processing on Large Clusters” PVLDB 3(1):285–296, 2010.) presents HaLoop, a modified version of the Hadoop MapReduce framework, that is designed to serve these applications. HaLoop allows iterative applications to be assembled from existing Hadoop programs without modification, and significantly improves their efficiency by providing inter-iteration caching mechanisms and a loop-aware scheduler to exploit these caches. HaLoop retains the fault-tolerance properties of MapReduce through automatic cache recovery and task re-execution. We evaluated HaLoop on a variety of real applications and real datasets. Compared with Hadoop, on average, HaLoop improved runtimes by a factor of 1.85 and shuffled only 4 % as much data between mappers and reducers in the applications that we tested.     

A CP framework for testing CP

The success of several constraint-based modeling languages such as OPL, ZINC or COMET, appeals for better software engineering practices, particularly in the testing phase. This paper introduces a testing framework enabling automated test case generation for constraint programming. We propose a general framework of constraint program development which supposes that a first declarative and simple constraint model is available from the problem specifications analysis. Then, this model is refined using classical techniques such as constraint reformulation, surrogate, redundant, implied, global constraint and symmetry-breaking to form an improved constraint model that must be thoroughly tested before being used to address real-sized problems. We think that most of the faults are introduced in this refinement step and propose a process which takes the first declarative model as an oracle for detecting non-conformities and derive practical test purposes from this process. We implemented this approach in a new tool called CPTEST that was used to automatically detect non-conformities on classical benchmark programs, such as the Golomb rulers, n-queens, social golfer and the car-sequencing problems.     

EASEA: specification and execution of evolutionary algorithms on GPGPU

EASEA is a framework designed to help non-expert programmers to optimize their problems by evolutionary computation. It allows to generate code targeted for standard CPU architectures, GPGPU-equipped machines as well as distributed memory clusters. In this paper, EASEA is presented by its underlying algorithms and by some example problems. Achievable speedups are also shown onto different NVIDIA GPGPUs cards for different optimization algorithm families.     

A code tagging approach to software product line development

Software product line engineering seeks to systematise reuse when developing families of similar software systems so as to minimise development time, cost and defects. To realise variability at the code level, product line methods classically advocate usage of inheritance, components, frameworks, aspects or generative techniques. However, these might require unaffordable paradigm shifts for developers if the software was not thought at the outset as a product line. Furthermore, these techniques can be conflicting with a company’s coding practices or external regulations. These concerns were the motivation for the industry–university collaboration described in this paper in which we developed a minimally intrusive coding technique based on tags. The approach was complemented with traceability from code to feature diagrams which were exploited for automated configuration. It is supported by a toolchain and is now in use in the partner company for the development of flight-grade satellite communication software libraries.