A Bayesian propensity score adjustment for latent variable modeling and MCMC algorithm

The estimation of the differences among groups in observational studies is frequently inaccurate owing to a bias caused by differences in the distributions of covariates. In order to estimate the average treatment effects when the treatment variable is binary, Rosenbaum and Rubin [1983. The central role of the propensity score in observational studies for causal effects. Biometrika 70, 41-55] proposed an adjustment method for pre-treatment variables using propensity scores. Imbens [2000. The role of the propensity score in estimating dose-response functions. Biometrika 87, 706-710] extended the propensity score methodology for estimation of average treatment effects with multivalued treatments.However, these studies focused only on estimating the marginal mean structure. In many substantive sciences such as the biological and social sciences, a general estimation method is required to deal with more complex analyses other than regression, such as testing group differences on latent variables. For latent variable models, the EM algorithm or the traditional Monte Carlo methods are necessary. However, in propensity score adjustment, these methods cannot be used because the full distribution is not specified.In this paper, we propose a quasi-Bayesian estimation method for general parametric models that integrate out the distributions of covariates using propensity scores. Although the proposed Bayes estimates are shown to be consistent, they can be calculated by existing Markov chain Monte Carlo methods such as Gibbs sampler. The proposed method is useful to estimate parameters in latent variable models, while the previous methods were unable to provide valid estimates for complex models such as latent variable models.We also illustrated the procedure using the data obtained from the US National Longitudinal Survey of Children and Youth (NLSY1979-2002) for estimating the effect of maternal smoking during pregnancy on the development of the child's cognitive functioning.     

A technique of minilaparotomy-assisted vaginal hysterectomy

PURPOSE: To report a patient with acute retinal pigment epitheliitis examined less than 24 hours after onset of symptoms. METHOD: One day after the onset of blurred vision in her left eye, a 33-year-old woman had a best-corrected visual acuity of LE, 20/60 -2. The left eye had classic uniform golden-colored nodules in a honeycomb pattern in the foveal retinal pigment epithelium. Intravenous fundus fluorescein angiography disclosed staining of the foveal pigment epithelium. RESULTS: One month after initial examination, visual acuity was LE, 20/20, and fine subfoveal pigmentary clumping was present. CONCLUSION: The pigmentary maculopathy of acute retinal pigment epitheliitis may be nonspecific, resulting from more than one type of primary foveal inflammation.     

Hematopoietic and lymphopoietic responses in human granulocyte-macrophage colony-stimulating factor (GM-CSF) receptor transgenic mice injected with human GM-CSF

Using a clonal assay of bone marrow (BM) cells from transgenic mice (Tg-mice) expressing the human granulocyte-macrophage colony-stimulating factor receptor (hGM-CSFR), we found in earlier studies that hGM-CSF alone supported the development not only of granulocyte-macrophage colonies, but also of erythrocytes, megakaryocytes, mast cells, blast cells, and mixed hematopoietic colonies. In this report, we evaluated the in vivo effects of hGM-CSF on hematopoietic and lymphopoietic responses in the hGM-CSFR Tg-mice. Administration of this factor to Tg-mice resulted in dose-dependent increases in numbers of reticulocytes and white blood cells (WBCs) in the peripheral blood. Morphological analysis of WBCs showed that the numbers of all types of the cell, including neutrophils, eosinophils, monocytes, and lymphocytes increased; the most remarkable being in lymphocytes that contained a number of large granular lymphocytes (LGLs) in addition to mature T and B cells. However, total cellularity of the BM of the Tg-mice decreased in a dose-dependent manner when hGM-CSF was injected. In sharp contrast to the BM, spleens of the Tg-mice were grossly enlarged. Although all types of blood cells and hematopoietic progenitors increased in the spleen, erythroid cells and their progenitors showed the most significant increase. Increased numbers of megakaryocytes and LGLs were also observed in spleen and liver of the treated Tg-mice. Flow cytometric analysis showed that LGLs expanded in Tg-mice expressed Mac-1+ CD3- NK1.1+. The thymus of Tg-mice treated with hGM-CSF exhibited a dose-dependent shrinkage and a remarkable decrease in CD4+ CD8+ cells. Thus, hGM-CSF stimulated not only myelopoiesis but also erythropoiesis and megakaryopoiesis of hGM-CSFR Tg-mice in vivo, in accordance with our reported in vitro findings. In addition, hGM-CSF affected the development of lymphoid cells, including natural killer cells of these Tg-mice.     

The anti-tumor arotinoid Ro 40-8757 protects bone marrow from the toxic effects of 5-fluorouracil

We have previously shown that strychnine mimics the cytoprotective properties of glycine in renal proximal tubule (RPT) suspensions exposed to antimycin A (AA). The aims of this study were to determine whether the cytoprotective properties of strychnine applied to various types of nephrotoxicants and to examine the temporal aspects of the cytoprotection of glycine and strychnine. Tubular release of LDH activity was used as a marker of cell death. Glycine (2 mM) or strychnine (1 mM) added 5 min prior to the toxicant decreased LDH release in rabbit RPT suspensions exposed to 25 microM tetrafluoroethyl-L-cysteine (TFEC), 10 microM HgCl2, 0.5 mM t-butyl hydroperoxide (TBHP), or 0.2 mM bromohydroquinone (BHQ) for 4 hr, or 2 mM sodium cyanide (NaCN) for 2 hr. The relative rank order of effectiveness of glycine and strychnine was NaCN = TFEC > BHQ > DCVC > TBHP > HgCl2. The temporal aspects of strychnine and glycine protection were examined by exposing RPT to either AA or TFEC for 1 or 3 hr, respectively, and then adding either 1 mM glycine or 1 mM strychnine. Glycine and strychnine decreased LDH release in AA-treated RPT at 1.25 and 2 hr and TFEC-treated RPT at 4 hr. In addition, when RPT exposed to AA or TFEC and treated with strychnine or glycine were washed at either 1 or 4 hr, protection was eliminated at later time points. When glycine was added to RPT treated with either PCBC, TFEC, or DCVC 5 min prior to or 30, 60, 120, and 180 min following toxicant addition, LDH release was reduced at all time points. These results demonstrate that strychnine and glycine protect RPT from a variety of diverse nephrotoxicants, strychnine and glycine do not need to be present at the time of toxic insult, strychnine and glycine cytoprotection is reversible, and strychnine and glycine act in the late phase of necrotic cell injury.     

Noise transfer characteristics in wavelength conversion based on cross-gain saturation in a semiconductor optical amplifier

Wavelength conversion based on cross-gain saturation in a semiconductor optical amplifier is studied from the viewpoint of how fluctuation in the input signal is transferred to the converted signal. Experiments in which a signal light with amplifier ASE noise is converted show that fluctuation in the "on"-level is suppressed while fluctuation in the "off"-level is enhanced due to the in/out relation of the converter.     

Production of a TiN film with nanoscale particle size by a combined method of plasma-alloy reaction and spray deposition

On leave from: Honda Research and Development Co., Ltd, Wako 351-01, Japan.     

Fault section detection system for 66 kV underground branchtransmission lines using optical magnetic field sensors

A fault section detection system that uses optical magnetic field sensors and instantly detects the section in which a ground fault occurred was developed for 66 kV underground multiterminal systems having Y-branch joint boxes. The optical magnetic field sensor, which is based on the Faraday effect in Bi-doped YIG ((BiYbGd)₃Fe₅O₁₂) having a large Verdet constant, detects cable conductor currents of 0 to 2000 A with high precision with the use of a laminated magnetic ring core of silicon steel plates. Sensors and a fault section detector/indicator of a system are connected with optical fibers capable of nonrepeated transmission of over 6 km     

IBM's Linux watch, the challenge of miniaturization

Nearly four years in development, the IBM Linux watch contains a complete computer system that runs Linux, displays X11 graphics, and has wireless connectivity. The system fits in a case that could pass as a slightly unusual analog timepiece with a somewhat odd shape and an extraordinarily brilliant face. The developers have created two versions of the watch, one with an organic light-emitting diode display and the other with a liquid crystal display. Still considered a research prototype, the watch already runs some personal information management applications, and it can communicate with PCs, PDAs, and other wireless-enabled devices, viewing condensed e-mail and directly receiving pager-like messages. Eventually, users will be able to access various Internet-based services, such as up-to-the-minute information about weather, traffic conditions, the stock market, and sports     

Tetramethylsuccinonitrile in polyvinyl chloride products for food and its release into food-simulating solvents

The content of tetramethylsuccinonitrile (TMSN), the main decomposition product of 2,2'-azobis-isobutyronitrile, in polyvinyl chloride (PVC) products used for food packaging, were examined as well as food-simulating solvents. The TMSN concentration in 17 PVC products ranged from below the detection limit, 0.05 mg/kg, up to 523 mg/kg. The release of TMSN from two PVC products into five kinds of food-simulating solvents at 60 degrees C for 30 min was observed, except for 1 +/- 1 micrograms/kg of TMSN in n-heptane from a PVC bottle containing 523 +/- 30.4 mg/kg of TMSN. The detection limit of TMSN in the food-simulating solvents was 1 micrograms/kg. When pieces of the bottle were stored in olive oil at 40 degrees C for 120 days, 5 +/- 1 micrograms/kg of TMSN was detected in the oil. The release of TMSN from the pieces of the bottle into olive oil between 80 and 140 degrees C depended on the formula ln y = 0.08786x-5.696, r = 0.9927, where y is the concentration (microgram/kg) of TMSN in olive oil, x is the temperature (degrees C), and r is the correlation coefficient.