Computer-Aided Diagnosis of Vertebral Compression Fractures Using Convolutional Neural Networks and Radiomics

Journal of Digital Imaging - Vertebral Compression Fracture (VCF) occurs when the vertebral body partially collapses under the action of compressive forces. Non-traumatic VCFs can be secondary to...     

Longitudinal evaluation of a hearing protector fit training program

Objective:The present study evaluates a training program for fitting different hearing protection devices (HPDs) based on personal attenuation rating (PAR) before, immediately after, and six months after training.Methods:A total of 67 workers from a public university in the city of São Paulo, Brazil, were invited to participate in the measurement of PARs for foam and silicone protectors through the 3M™ E-A-Rfit Validation System. Two evaluations were performed for each protector at each sampling date: one after reading printed material (the package instructions) and another after being trained by an audiologist. The same procedures were repeated after six months. The final sample consisted of 30 individuals. ANOVA was used for statistical analysis.Results:Larger PAR values were observed after training by the audiologist, and smaller values were observed after six months. Then, after re-training, the values increased again. There were no statistically significant differences in PAR among the HPDs tested. Even after the two training sessions, 23 to 27% of the subjects did not obtain adequate PAR values.Conclusion:These findings emphasize the need for continual worker training in the correct fit of earplug HPDs and the importance of longitudinal PAR monitoring. In addition, some workers, despite the training provided, did not adapt to the HPDs used. Therefore, it is essential that other protection methods and/or other HPD types are made available to these individuals.Key words: Hearing, noise-induced hearing loss, occupational health, personal protective equipment, hearing protection devices     

Facial expression recognition in Alzheimer’s disease: A systematic review

Introduction: It is well established that behavioral variant frontotemporal dementia can impair social and emotional function. However, there is no consensus regarding how Alzheimer’s disease can affect facial expression recognition. We aim to systematically review all the literature addressing this issue over the last 10 years.

Method: We conducted a search based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). The search for literature was undertaken on 19 September 2017, using Pubmed, SciELO, BIREME, and Thomson Reuters Web of Science electronic databases. The key terms for the search were: Alzheimer’s disease, dementia, and facial expression recognition.

Results: We screened 173 articles, and 22 of them were selected. The most common methodology involved showing participants photographs of people expressing the six basic emotions—fear, anger, sadness, disgust, surprise, and happiness. Results were ambiguous. Among people with mild Alzheimer’s disease, happiness was easier to recognize than the other five basic emotions, with sadness and anger the most difficult to recognize. In addition, the intensity level of the emotions presented seems to be important, and facial expression recognition is related to specific cognitive capacities, including executive function and visuoperceptual abilities. Impairment in facial expression recognition does not appear to be a consistent neuropsychological finding in Alzheimer’s disease.

Conclusions: The lack of standardized assessment instruments and the heterogeneity of the methods and samples used across studies hamper comparisons. Future researches should investigate facial expression recognition through more ecological and standardized methods.     

Riesgo de recurrencia tras retirada de la anticoagulación en pacientes con enfermedad tromboembólica venosa no provocada: validación externa del nomograma de Viena y del modelo predictivo DASH

IntroductionScales for predicting venous thromboembolism (VTE) recurrence are useful for deciding the duration of the anticoagulant treatment. Although there are several scales, the most appropriate for our setting has not been identified. For this reason, we aimed to validate the DASH prediction score and the Vienna nomogram at 12 months.MethodsThis was a retrospective study of unselected consecutive VTE patients seen between 2006 and 2014. We compared the ability of the DASH score and the Vienna nomogram to predict recurrences of VTE. The validation was performed by stratifying patients as low-risk or high-risk, according to each scale (discrimination) and comparing the observed recurrence with the expected rate (calibration).ResultsOf 353 patients evaluated, 195 were analyzed, with an average age of 53.5 ± 19 years. There were 21 recurrences in 1 year (10.8%, 95% CI: 6.8%-16%). According to the DASH score, 42% were classified as low risk, and the rate of VTE recurrence in this group was 4.9% (95% CI: 1.3%-12%) vs. the high-risk group that was 15% (95% CI: 9%-23%) (p <.05). According to the Vienna nomogram, 30% were classified as low risk, and the rate of VTE recurrence in the low risk group vs. the high risk group was 4.2% (95% CI:0.5%-14%) vs. 16.2% (95% CI: 9.9%-24.4%) (p <.05).ConclusionsOur study validates the DASH score and the Vienna nomogram in our population. The DASH prediction score may be the most advisable, both because of its simplicity and its ability to identify more low-risk patients than the Vienna nomogram (42% vs. 30%).