Instruction via a mobile social media app, in conjunction with regular instruction, increases subjective measures of adequacy of bowel preparation. Use of the app significantly increased the proportion of patients with successful cecal intubation and in whom adenomas were detected, indicating increased quality of colonoscopy. ClinicalTrials.gov number: NCT02140827.
We propose using faster regions with convolutional neural network features (faster R-CNN) in the TensorFlow tool package to detect and number teeth in dental periapical films. To improve detection precisions, we propose three post-processing techniques to supplement the baseline faster R-CNN according to certain prior domain knowledge. First, a filtering algorithm is constructed to delete overlapping boxes detected by faster R-CNN associated with the same tooth. Next, a neural network model is implemented to detect missing teeth. Finally, a rule-base module based on a teeth numbering system is proposed to match labels of detected teeth boxes to modify detected results that violate certain intuitive rules. The intersection-over-union (IOU) value between detected and ground truth boxes are calculated to obtain precisions and recalls on a test dataset. Results demonstrate that both precisions and recalls exceed 90% and the mean value of the IOU between detected boxes and ground truths also reaches 91%. Moreover, three dentists are also invited to manually annotate the test dataset (independently), which are then compared to labels obtained by our proposed algorithms. The results indicate that machines already perform close to the level of a junior dentist.
Compared to sequential learning models, graph-based neural networks exhibit some excellent properties, such as ability capturing global information. In this paper, we investigate graph-based neural networks for text classification problem. A new framework TensorGCN (tensor graph convolutional networks), is presented for this task. A text graph tensor is firstly constructed to describe semantic, syntactic, and sequential contextual information. Then, two kinds of propagation learning perform on the text graph tensor. The first is intra-graph propagation used for aggregating information from neighborhood nodes in a single graph. The second is inter-graph propagation used for harmonizing heterogeneous information between graphs. Extensive experiments are conducted on benchmark datasets, and the results illustrate the effectiveness of our proposed framework. Our proposed TensorGCN presents an effective way to harmonize and integrate heterogeneous information from different kinds of graphs.
SummaryThe rapid increase of ~omics datasets generated by microarray, mass spectrometry and next generation sequencing technologies requires an integrated platform that can combine results from different ~omics datasets to provide novel insights in the understanding of biological systems. MADMAX is designed to provide a solution for storage and analysis of complex ~omics datasets. In addition, analysis results (such as lists of genes) can be merged to reveal candidate genes supported by all datasets. The system constitutes an ISA-Tab compliant LIMS part, which is linked to the different analysis pipelines. A pilot study of different type of ~omics data in Brassica rapa demonstrates the possible use of MADMAX. The web-based user interface provides easy access to data and analysis tools on top of the database.
IntroductionTo better understand how phenotypes emerge, increasingly series of ~omics technologies (genomics, transcriptomics, proteomics, metabolomics) rather than individual measurements are necessarily used within a single study. Such efforts boost the demands of both data storage and data analysis of different high-throughput approaches. However, in the past it was hardly possible to store metadata from different ~omics technologies in the same repository. To accommodate this demand the ISA-Tab [1] format was proposed to build up a common structured representation of the metadata of studies from a combination of technologies. This also triggered attempts to develop data processing tools tailored to the needs of biologists. Unfortunately most of these tools have high demands on hardware requirements, or contain non-intuitive command line-based interfaces.* To whom correspondence should be addressed: jack.leunissen@wur.nl Here we present MADMAX, a multi-purpose database for the management and analysis of data from multiple ~omics experiments. It includes an ISA-Tab compliant backend database and a series of analysis pipelines for transcriptomics, metabolomics and genomics datasets; these pipelines are connected to the database through webservices such that other pipelines can be easily integrated into the current system ( Figure 1 Through the web interface, the user can store a complete experiment with all fields required in ISA-Tab format, sufficient to allow for subsequent analysis or even repeating the experiment later. Another section on the website is the central access to different analysis pipelines. Both individual analysis results and combined gene lists can be retrieved in the system for download. Centrally stored experiments and analysis results can only be accessed by the creator by default and will be accessible for other users only if the creator desires to share the data. The system is on an automatic backup schedule.MADMAX can be reached at http://madmax2.bioinformatics.nl/ and is available upon request by sending an email to madmax.request@bioinformatics.nl.
ImplementationMADMAX is built upon an Oracle relational database on a Linux server and a computational analysis engine for different...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.