Functional longevity of intraperioneal drains: an experimental evaluation

Various types of drains were inserted into the peritoneal cavity of twenty-eight dogs. After one to seven days, all drains failed to show the presence of 200 cc of colored fluid injected intraperitoneally. On autopsy, all tubes were surrounded and occluded by omentum.     

Upper extremity replantation: current concepts and patient selection

Dramatic advances in replantation and microsurgery have somewhat altered the criteria we use in selecting patients to be candidates for upper extremity replantation surgery. We suggest that contraindications for such replantation are: presence of associated life-threatening injuries; serious anesthetic risk; preexisting medical or psychiatric problems; previous injury or disease of the amputated part; warm ischemic time greater than 6-8 hours for extremities or greater than 10-12 hours for digits; and single-digit amputations (except thumb, for grasp). Replantation is feasible when: amputated part is properly preserved; injury type is sharp amputation, mild to moderate crush, or selected avulsion, and amputation is proximal to the DIP joint. Careful preservation of the amputated part, not in dry ice, is mandatory. On an individual basis, the decision to attempt replantation rests on the prediction that the patient may have better function with such surgery than with a prosthesis.     

Two-Dimensional Modeling and Simulation of Mass Transport in Microfabricated Preconcentrators

The adsorption and desorption behavior of a planar microfabricated preconcentrator (PC) has been modeled and simulated using the computational fluid dynamics (CFD) package CFDRC-ACE+trade. By comparison with the results of a designed experiment, model parameters were determined. Assuming a first-order reaction for the adsorption of a light hydrocarbon chemical analyte onto the PC adsorbent and a unity-value sticking coefficient, a rate constant of 36 500 s^-1 was obtained. This compares favorably with the value of 25 300 s^-1 obtained by application of the Modified-Wheeler equation. The modeled rate constant depends on the concentration of adsorbent sites, estimated to be 6.94 ldr 10^-8 kmol/m² for the Carboxen 1000 adsorbent used. Using the integral method, desorption was found to be first order with an Arrhenius temperature dependence and an activation energy of 30.1 kj/mol. Validation of this model is reported herein, including the use of Aris-Taylor dispersion to predict the influence of fluidics surrounding the PC. A maximum in desorption peak area with flow rate, predicted from a quadratic fit to the results of the designed experiment, was not observed in the 2-D simulation. Either approximations in the simulated model or the nonphysical nature of the quadratic fit are responsible. Despite the apparent simplicity of the model, the simulation is internally self consistent and capable of predicting performance of new device designs. To apply the method to other analytes and other adsorbent materials, only a limited number of comparisons to experiment are required to obtain the necessary rate constants.     

Limbic dysfunction in schizophrenia and mania. A study using 18F-labelled fluorodeoxyglucose and positron emission tomography

BACKGROUND: Diagnostic classes (derived from CATEGO) can be correlated with regional brain metabolism in patients with major psychiatric disorders. METHOD: Seventeen patients with schizophrenia, 15 with mania, 10 with depression and 10 healthy Volunteers were examined with positron emission tomography (PET) and 18F-labelled fluorodeoxyglucose, as a marker for glucose metabolism. The number of possible comparisons of regions of interest was reduced by principal-components analysis, and differences in factor scores were determined between diagnostic groups. RESULTS: Four independent factors, representing distributed brain systems, emerged: an anterior-posterior (1), a left-right temporal (2), a temporofrontal (3), and a mediofrontal (4) system, of which (1), (2) and (3) were abnormal in schizophrenia, (1) and (2) in mania, and (1) in depression. CONCLUSIONS: Abnormal patterns of metabolism could be detected, in decreasing order, in schizophrenia, mania and depression. Some of these abnormalities are likely to be due to medication, but others will be associated with structural or functional abnormalities of the frontolimbic system in the diagnostic groups.     

Survivable network system analysis: a case study

The Survivable Network Analysis method permits assessment of survivability at the architecture level. Steps include system mission and architecture definition, essential capability definition, compromisable capability definition, and survivability analysis of architectural soft-spots that are both essential and compromisable. The article summarizes application of the method to a subsystem of a large-scale, distributed health care system     

Managing Impressions Online: Self-Presentation Processes in the Online Dating Environment

This study investigates self-presentation strategies among online dating participants, exploring how participants manage their online presentation of self in order to accomplish the goal of finding a romantic partner. Thirty-four individuals active on a large online dating site participated in telephone interviews about their online dating experiences and perceptions. Qualitative data analysis suggests that participants attended to small cues online, mediated the tension between impression management pressures and the desire to present an authentic sense of self through tactics such as creating a profile that reflected their "ideal self," and attempted to establish the veracity of their identity claims. This study provides empirical support for Social Information Processing theory in a naturalistic context while offering insight into the complicated way in which "honesty" is enacted online.     

MIQR Active Learning on a Continuous Function and a Discontinuous Function

Active learning balances the cost of data acquisition against its usefulness for training. We select only those data points which are the most informative about the system being modelled. The MIQR (Maximum Inter-Quartile Range) criterion is defined by computing the inter-quartile range of the outputs of an ensemble of networks, and finding the input parameter values for which this is maximal. This method ensures data selection is not unduly influenced by ‘outliers’, but is principally dependent upon the ‘mainstream’ state of the ensemble. MIQR is more effective and efficient than contending methods¹ . The algorithm automatically regulates the training threshold and the network architecture as necessary. We compare active learning methods by applying them to a continuous function and a discontinuous function. Training is more difficult for a discontinuous function than a continuous function, and the volume of data for active learning is substantially less than for passive learning.