Background and Purpose-Despite the evidence for a positive association of dietary salt and blood pressure, the few prospective studies that have assessed the association between dietary salt and stroke have reported inconsistent results. The purpose of this study was to examine the relation between sodium intake and death from stroke in a population-based cohort of Japanese men and women. Methods-In 1992, usual diet including sodium intake was determined in 13 355 men and 15 724 women in Takayama City, Gifu, with the use of a validated food frequency questionnaire. Results-There were 269 stroke deaths (137 men and 132 women) between baseline and 1999. In men, the highest compared with the lowest tertile of sodium intake was significantly positively associated with death from total stroke after controlling for covariates (hazard ratio [HR]), 2.33; 95% CI, 1.23 to 4.45). Significantly positive associations were also observed between sodium intake and death from ischemic stroke (HR, 3.22; 95% CI, 1.22 to 8.53) as well as death from intracerebral hemorrhage (HR, 3.85; 95% CI, 1.16 to 12.7). A positive association between sodium intake and death from stroke in women was suggested, although the associations for total stroke and ischemic stroke were of borderline significance (HR, 1.70; 95% CI, 0.96 to 3.02 and HR, 2.10; 95% CI, 0.96 to 4.62, respectively). Conclusions-These prospective data support the hypothesis that dietary salt increases the risk of death from stroke.

Recent advances in 3D fully convolutional networks (FCN) have made it feasible to produce dense voxel-wise predictions of volumetric images. In this work, we show that a multi-class 3D FCN trained on manually labeled CT scans of several anatomical structures (ranging from the large organs to thin vessels) can achieve competitive segmentation results, while avoiding the need for handcrafting features or training class-specific models. To this end, we propose a two-stage, coarse-to-fine approach that will first use a 3D FCN to roughly define a candidate region, which will then be used as input to a second 3D FCN. This reduces the number of voxels the second FCN has to classify to ∼10% and allows it to focus on more detailed segmentation of the organs and vessels. We utilize training and validation sets consisting of 331 clinical CT images and test our models on a completely unseen data collection acquired at a different hospital that includes 150 CT scans, targeting three anatomical organs (liver, spleen, and pancreas). In challenging organs such as the pancreas, our cascaded approach improves the mean Dice score from 68.5 to 82.2%, achieving the highest reported average score on this dataset. We compare with a 2D FCN method on a separate dataset of 240 CT scans with 18 classes and achieve a significantly higher performance in small organs and vessels. Furthermore, we explore fine-tuning our models to different datasets. Our experiments illustrate the promise and robustness of current 3D FCN based semantic segmentation of medical images, achieving state-of-the-art results..

No abstract

Colorectal cancer incidence in relation to body size, smoking, and alcohol consumption was studied in a cohort of 29 051 city residents of Japan. In 1992, each participant completed a self-administered questionnaire on sociodemographic characteristics, drinking, cigarette smoking, diet, exercise, and reproductive and medical histories. The response rate was 92%. From 1993 to 2000, 161 men and 134 women were diagnosed with colorectal cancer at two major hospitals in the city. Relative risks and 95% confidence intervals were calculated by using Cox proportional hazard models. A positive relation between height and colorectal cancer was seen in both sexes, controlling for age, body mass index (BMI), smoking and drinking habits, and years of education. The findings were statistically significant only for men (relative risk 2.13 for the tallest compared with the shortest height tertile; 95% confidence interval ¼ 1.26 -3.58). Body mass index was also associated positively with colon cancer risk for men, whereas the pattern for women was not clear. There was a positive association between pack-years of cigarette smoking and the risk of rectal cancer in men. A positive dose -response relation between alcohol consumption and colon cancer risk was observed for men and women.

Ultraviolet (UV) irradiation stimulates stress-activated protein kinase/c-Jun N-terminal kinase (SAPK/ JNK), which is a member of the mitogen-activated protein kinase (MAPK) superfamily and implicated in stress-induced apoptosis. UV also induces the activation of another MAPK member, extracellular signal-regulated kinase (ERK), which is typically involved in a growth-signaling cascade. However, the UV-induced signaling pathway leading to ERK activation, together with the physiological role, has remained unknown. Here we examined the molecular mechanism and physiological function of UV-induced ERK activation in human epidermoid carcinoma A431 cells that retain a high number of epidermal growth factor (EGF) receptors. UV-induced ERK activation was accompanied with the Tyr phosphorylation of EGF receptors, and both responses were completely abolished in the presence of a selective EGF receptor inhibitor (AG1478) or the Src inhibitor PP2 and by the expression of a kinase-dead Src mutant. On the other hand, SAPK/JNK activation by UV was partially inhibited by these inhibitors. UV stimulated Src activity in a manner similar to the ERK activation, but the Src activation was insensitive to AG1478. UV-induced cell apoptosis measured by DNA fragmentation and caspase 3 activation was enhanced by AG1478 and an ERK kinase inhibitor (U0126) but inhibited by EGF receptor stimulation by the agonist. These results indicate that UV-induced ERK activation, which provides a survival signal against stress-induced apoptosis, is mediated through Src-dependent Tyr phosphorylation of EGF receptors.