RESUMO
Objective: The purpose of this study was to compare donepezil at 5 mg and 10 mg/day against a placebo to systematically evaluate its effectiveness in improving cognitive function among patients suffering from dementia at any stage. Method: For this systematic review and meta-analysis, we looked up Medline, Scopus, Embase, Web of Science, and The Cochrane Library for articles on the efficacy of donepezil in dementia published in the past 20 years and summarized the placebo and intervention data. Initially, a total of 2,272 articles were extracted using our search query and after the inclusion and exclusion criteria set for extraction of data, 18 studies were included in this review using PRISMA flowchart. The ADAS-cog and MMSE assessment scales were used for measuring the outcomes using IBM SPSS 29.0 for the meta-analysis. Result: The meta-analysis comprised a total of 18 RCTs (randomized controlled trials) that were randomized to receive either donepezil 5 mg/day (n = 1,556), 10 mg/day (n = 2050) or placebo (n = 2,342). Meta-analysis concerning efficacy showed that donepezil at 10 mg/day significantly improved the MMSE score (g: 2.27, 95%CI: 1.25-3.29) but could not substantially reduce the ADAS-cog. At 5 mg/day donepezil, an overall slight improvement in MMSE score (Hedges' g: 2.09, 95%CI: 0.88-3.30) was observed. Conclusion: Both donepezil 5 mg/day and 10 mg/day doses demonstrated improved cognitive functions for patients with dementia, however results indicated that the 10 mg/day dose was more efficacious.
RESUMO
Unlike 2-dimensional (2D) images, direct 3-dimensional (3D) point cloud processing using deep neural network architectures is challenging, mainly due to the lack of explicit neighbor relationships. Many researchers attempt to remedy this by performing an additional voxelization preprocessing step. However, this adds additional computational overhead and introduces quantization error issues, limiting an accurate estimate of the underlying structure of objects that appear in the scene. To this end, in this article, we propose a deep network that can directly consume raw unstructured point clouds to perform object classification and part segmentation. In particular, a Deep Feature Transformation Network (DFT-Net) has been proposed, consisting of a cascading combination of edge convolutions and a feature transformation layer that captures the local geometric features by preserving neighborhood relationships among the points. The proposed network builds a graph in which the edges are dynamically and independently calculated on each layer. To achieve object classification and part segmentation, we ensure point order invariance while conducting network training simultaneously-the evaluation of the proposed network has been carried out on two standard benchmark datasets for object classification and part segmentation. The results were comparable to or better than existing state-of-the-art methodologies. The overall score obtained using the proposed DFT-Net is significantly improved compared to the state-of-the-art methods with the ModelNet40 dataset for object categorization.
