It is often assumed that topic models benefit from the use of a manually curated stopword list. Constructing this list is timeconsuming and often subject to user judgments about what kinds of words are important to the model and the application. Although stopword removal clearly affects which word types appear as most probable terms in topics, we argue that this improvement is superficial, and that topic inference benefits little from the practice of removing stopwords beyond very frequent terms. Removing corpus-specific stopwords after model inference is more transparent and produces similar results to removing those words prior to inference.

A caudal cerebellar infarction may easily be misdiagnosed clinically as a labyrinthine disorder, and was found to be the cause in one fourth of patients presenting with isolated acute vertigo.

Background and Purpose In a randomized study of hemiparetic stroke patients with a median age of 75 years, functional recovery was significantly better in those who received additional sensory stimulation (n=38), including electrostimulation, than in control patients (n=40) given the same physiotherapy and occupational therapy; group differences for balance, mobility, and activities of daily living were significant. The present study was designed to investigate postural control in patients who survived more than 2 years after stroke onset.Methods The 48 survivors (mean, 2.7 years; range, 2.0 to 3.8 years), 22 from the treatment group and 26 from the control group, were compared with 23 age-matched healthy subjects. Subjects were perturbed by vibrators applied to calf muscles or with galvanic vestibular stimulation. We evaluated postural control in terms of sway variances or sway velocities and the dynamics of postural control as a feedback system using system identification with a model previously validated for human postural control.

Background:A few studies indicated that hepatitis C and hepatitis B virus (HCV/HBV) might be associated with pancreatic cancer risk. The aim of this nationwide cohort study was to examine this possible association.Methods:Hepatitis C virus- and hepatitis B virus-infected individuals were identified from the national surveillance database from 1990 to 2006, and followed to the end of 2008. The pancreatic cancer risk in the study population was compared with the general population by calculation of Standardized Incidence Ratios (SIRs), and with a matched reference population using a Cox proportional hazards regression model to calculate hazard ratios (HRs).Results:In total 340 819 person-years in the HCV cohort and 102 295 in the HBV cohort were accumulated, with 34 and 5 pancreatic cancers identified, respectively. The SIRHCV was 2.1 (95% confidence interval, CI: 1.4, 2.9) and the SIRHBV was 1.4 (0.5, 3.3). In the Cox model analysis, the HR for HCV infection was 1.9 (95% CI: 1.3, 2.7), diminishing to 1.6 (1.04, 2.4) after adjustment for potential confounders.Conclusion:Our results indicated that HCV infection might be associated with an increased risk of pancreatic cancer but further studies are needed to verify such association. The results in the HBV cohort indicated an excess risk, however, without statistical significance due to lack of power.

Previously [The abstract has been shortened somewhat, and the headings removed, in order to conform to journal style], we have observed vestibular asymmetry in about one-third of healthy senior citizens and in about two-thirds of subjects with previous hip fractures and no other significant ailments. Wrist fractures are considered a harbinger of hip fractures. If vestibular asymmetry is correlated with falls and fractures among the elderly then it should also be reflected among subjects with wrist fractures. Sixty-six consecutive patients (mean age 67.8 years) who had sustained a fall-related wrist fracture during a 10-month period were included in the study. The frequency of head shake nystagmus among the patients was compared to that found among 49 healthy senior citizens (mean age 74.9 years). Nystagmus after head shaking, indicating asymmetric vestibular function, was found in 50 participants (76%) (p <0.001). Thirty-eight of these were graded with distinct or prominent nystagmus responses. Sixty percent of the subjects with horizontal nystagmus had a wrist fracture coinciding with the slow phase of nystagmus. Twenty-three subjects reported 30 previous fall-related fractures during the previous 10 years. Subjects with nystagmus after head shaking sustained 26 of these fractures. The frequency of signs of vestibular asymmetry was significantly higher (p < 0.001) among the subjects than among healthy senior citizens. These findings suggest that an asymmetric vestibular function could be an epidemiologically important contributory factor to falls and wrist fractures among the elderly population.