Synaptic physiology of horizontal connections in the cat's visual cortex.

Horizontal connections are a principal component of intrinsic cortical circuitry. They arise mainly from pyramidal cells and course parallel to the brain's surface for distances as long as 8 mm, linking columns with shared orientation preference and allowing cells to integrate visual information from outside their receptive fields. We examined the synaptic physiology of the horizontal pathway in slices of the cat's striate cortex and found that activating lateral fibers produced both excitation and inhibition. We recorded the postsynaptic responses of identified pyramidal cells in layer 2 + 3 of area 17 to electrical shocks applied at three sites: in the home column of the impaled neuron either in layer 2 + 3 or 4, or at a lateral distance of 0.9-3 mm in layer 2 + 3. Within the home column, suprathreshold stimuli produced compound EPSPs with action potentials, followed by fast, GABAAergic IPSPs and a slower, GABABergic IPSP. For the distant stimulating site, the threshold response was an EPSP. Stronger shocks frequently evoked a disynaptic, GABAAergic IPSP that truncated the EPSP and could dominate the postsynaptic response. At the resting potential, the horizontally evoked EPSP was too small to elicit spikes. With depolarization of the membrane, however, it grew several hundred-fold. This amplification was blocked by N-(2,6-dimethylphenylcarbamoylmethyl)triethylammonium bromide (QX-314), but not by 2-amino-5-phosphonovalerate (APV), indicating that it was mediated by Na+ channels, rather than by NMDA receptors. We propose that the horizontal connections provide the means for stimuli outside the receptive field to modulate activity elicited within its confines. The voltage-dependent enhancement of the laterally evoked EPSP may explain why stimulating the surround by itself fails to drive cells but can facilitate their response to stimuli within the receptive field. The ability to initiate disynaptic inhibition from lateral sites shows that recruiting appropriate groups of horizontal fibers can also have a suppressive effect. Thus, the effect of horizontal input is state dependent, with the size and sign of the laterally evoked response changing according to the balance of converging inputs.     

A framework for evaluating image segmentation algorithms.

The purpose of this paper is to describe a framework for evaluating image segmentation algorithms. Image segmentation consists of object recognition and delineation. For evaluating segmentation methods, three factors-precision (reliability), accuracy (validity), and efficiency (viability)-need to be considered for both recognition and delineation. To assess precision, we need to choose a figure of merit, repeat segmentation considering all sources of variation, and determine variations in figure of merit via statistical analysis. It is impossible usually to establish true segmentation. Hence, to assess accuracy, we need to choose a surrogate of true segmentation and proceed as for precision. In determining accuracy, it may be important to consider different 'landmark' areas of the structure to be segmented depending on the application. To assess efficiency, both the computational and the user time required for algorithm training and for algorithm execution should be measured and analyzed. Precision, accuracy, and efficiency factors have an influence on one another. It is difficult to improve one factor without affecting others. Segmentation methods must be compared based on all three factors, as illustrated in an example wherein two methods are compared in a particular application domain. The weight given to each factor depends on application.     

A profile of homicides on the streets and in the homes of New York City.

Although prior research has found that homes containing firearms and illicit drug and ethanol users are more likely to be the scene of a homicide than homes that do not contain these elements, the authors studied homicides on the streets as well as in homes so as to assess the role of firearms, cocaine, and ethanol in both settings. Using the files of the Chief Medical Examiner, the authors reviewed all 4,468 homicides occurring in New York City in 1990 and 1991. The most frequent places of occurrence were the streets and other outdoor places (49.6 percent) and the victims'' homes (19.3 percent). Firearms were the cause of death for 49.6 percent of homicides in the home and 80.3 percent of those on the streets. Use of cocaine and ethanol was found more frequently among victims killed on the streets than those killed at home. In addition, victims killed on the streets were more likely to be male, ages 15-24 years, and African American. Further research in regard to prevention and intervention strategies is needed, keeping in mind the different patterns of homicide on the streets compared with those occurring in other settings.     

Gliomatosis cerebri: a case report with autopsy correlation

Summary MRI-autopsy correlation in a case of gliomatosis cerebri suggests that poor gray-white matter demarcation on MRI may be sign of neoplastic infiltration. The extent of infiltration is imperfectly assessed by current imaging modalities.     

Computed tomography of complications caused by manufacturing defects in synthetic grafts

Two unusual cases of delayed synthetic graft complications resulting from manufacturing defects of knitted Dacron are presented. Computed tomography appears to be a suitable noninvasive alternative examination to angiography in this condition.