Context-Fused Guidance for Image Captioning Using Sequence-Level Training

Recent image captioning models based on the encoder-decoder framework have achieved remarkable success in humanlike sentence generation. However, an explicit separation between encoder and decoder brings out a disconnection between the image and sentence. It usually leads to a rough image description: the generated caption only contains main instances but neglects additional objects and scenes unexpectedly, which reduces the caption consistency of the image. To address this issue, we proposed an image captioning system within context-fused guidance in this paper. It incorporates regional and global image representation as the compositional visual features to learn the objects and attributes in images. To integrate image-level semantic information, the visual concept is employed. To avoid misleading decoding, a context fusion gate is introduced to calculate the textual context by selectively aggregating the information of visual concept and word embedding. Subsequently, the context-fused image guidance is formulated based on the compositional visual features and textual context. It provides the decoder with informative semantic knowledge. Finally, a captioner with a two-layer LSTM architecture is constructed to generate captions. Moreover, to overcome the exposure bias, we train the proposed model through sequence decision-making. The experiments conducted on the MS COCO dataset show the outstanding performance of our work. The linguistic analysis demonstrates that our model improves the caption consistency of the image.

Biomechanical Loads and Their Effects on Player Performance in NCAA D-I Male Basketball Games

Basketball games and training sessions are characterized by quick actions and many scoring attempts, which pose biomechanical loads on the bodies of the players. Inertial Measurement Units (IMUs) capture these biomechanical loads as PlayerLoad and Inertial Movement Analysis (IMA) and teams collect those data to monitor adaptations to training schedules. However, the association of biomechanical loads with game performance is a relatively unexplored area. The aims of the current study were to determine the statistical relations between biomechanical loads in games and training with game performance. Biomechanical training and game load measures and player-level and team-level game stats from one college basketball team of two seasons were included in the dataset. The training loads were obtained on the days before gameday. A three-step analysis pipeline modeled: (i) relations between team-level game stats and the win/loss probabilities of the team, (ii) associations between the player-level training and game loads and their game stats, and (iii) associations between player-level training loads and game loads. The results showed that offensive and defensive game stats increased the odds of winning, but several stats were subject to positional and individual performance variability. Further analyses, therefore, included total points [PTS], two-point field goals, and defensive rebounds (DEF REB) that were less subject to those influences. Increases in game loads were significantly associated with game stats. In addition, training loads significantly affected the game loads in the following game. In particular, increased loads 2 days before the game resulted in increased expected game loads. Those findings suggested that biomechanical loads were good predictors for game performance. Specifically, the game loads were good predictors for game stats, and training loads 2 days before gameday were good predictors for the expected game load. The current analyses accounted for the variation in loads of players and stats that enabled modeling the expected game performance for each individual. Coaches, trainers, and sports scientists can use these findings to further optimize training plans and possibly make in-game decisions for individual player performance.

The Impact of State Dementia Training Requirements for Nursing Homes on Resident Outcomes

Stakeholders, including policymakers, have prioritized the need to educate nursing home (NH) staff about Alzheimer’s disease and dementia. Despite this prioritization and the relationship between staff training and outcomes, dementia-specific knowledge is variable. This study examined state-level training policies between 2011-2016. During this time 12 states (regulators and payers) implemented NH dementia training requirements, creating an opportunity for a natural experiment between states with and without new requirements. We estimated difference-in-differences models to determine the effect of state requirements on outcomes. Data from Nursing Home Compare and LTCFocus.org were linked to data on state policies. Training requirements were associated with 0.39 and 0.17 percentage point reductions in antipsychotics use and restraint use, respectively, and no impact on falls or need for help with daily activities. State requirements for dementia training in NHs are associated with a small, but significant reduction in the use of antipsychotic medication and physical restraints.

88. Review of nursing home antibiotic stewardship citation deficiencies to identify opportunities to improve antibiotic stewardship implementation, 2018-2019

Background Centers for Medicare & Medicaid Services (CMS) set standards for nursing homes (NH) and conduct inspections to assess adherence to regulatory requirements, including antibiotic stewardship implementation. NHs not meeting requirements are issued a citation. We reviewed text summaries for citations related to antibiotic stewardship to assess implementation in NHs. Methods We obtained publicly available antibiotic stewardship deficiency citations issued to NHs from 9/2018—7/2019 and NH characteristics data from CMS Nursing Home Compare. We used the χ2 test to compare characteristics of NHs with and without citations. We did a qualitative review of a randomly selected subset (318/635) of antibiotic stewardship citations and classified citations into one or more of four categories based on Centers for Disease Control and Prevention’s Core Elements of Antibiotic Stewardship: 1) Leadership & Accountability, 2) Action, 3) Tracking & Reporting, 4) Education (Fig 1). We developed subcategories based on our iterative review process to further describe the citations. Each citation was reviewed by two reviewers and yielded 95% agreement in categorization. Discordant citations were reviewed by a third reviewer, and core element categories with agreement by ≥2/3 reviewers were assigned, resulting in 99% agreement. Antibiotic Stewardship Citation Deficiency Category Common Themes and Examples from Citation Text from Qualitative Review of Antibiotic Stewardship Citation Deficiencies. Results There were 635 NH stewardship citation deficiencies across 44 states from 9/2018—7/2019. NHs with a citation were more likely to have < 100 beds (60% vs. 50%) and for-profit ownership (75% vs. 70%) (Table 1). Of the 318 reviewed citations, Action was cited in 67% of NHs; 115/213 had missing or incomplete criteria documented for antibiotic initiation. Tracking & Reporting was cited in 40% of NHs; 117/126 had missing or incomplete antibiotic or infection tracking logs. Leadership & Accountability was cited in 23% of NHs; 41/72 NHs had no stewardship policy available. Education was cited in 13% of NHs (Fig 2). Table 1. Characteristics of US Nursing Homes With and Without an Antibiotic Stewardship Citation Deficiency from 9/2018-7/2019. Figure 2. Types of Antibiotic Stewardship Citation Deficiencies in US Nursing Homes, 2018-2019, N=318. Conclusion The most common opportunities for improvement fell under appropriate assessment and documentation of criteria for antibiotic initiation at the resident and facility-level. Training NH staff to use available resources for antibiotic stewardship activities may improve implementation. Further evaluation to identify barriers to implementation is needed. Disclosures David Gifford, MD, MPH, American Healthcare Association (Employee)

Innovative methods and technologies of training specialists in the conditions of postgraduate pedagogical education

In the context of postgraduate pedagogical education reform, transition to multi-level training of teachers, ensuring competitiveness of specialists in a reasonable labor market, issue of preparing future teachers for professional and pedagogical activities is very relevant. Since the Republic of Kazakhstan joined ranks of independent States, changes in public awareness, along with all its areas have increased the requirements for the level of teachers’ professionalism, achievements in science and industry, and introduction of advanced technologies. In education of a person who meets the new requirements of society, as a factor in ensuring national security, we consider the training of creative teachers who think innovatively, make complex pedagogical decisions, formulate problems, and change their stereotypes. The research uses interview to collect data on the topic. The study discusses the benefits of the undergraduate, masters and PhD programs. Keywords: design training technology; hardware platforms; modern technologies; SAS-study.

Innovative Methods and Technologies of Training Specialists in Postgraduate Pedagogical Education

In the context of postgraduate pedagogical education reform, transition to multi-level training of teachers, ensuring competitiveness of specialists in a reasonable labor market, issue of preparing future teachers for professional and pedagogical activities is very relevant. Since the Republic of Kazakhstan joined ranks of independent States, changes in public awareness, along with all its areas, have increased the requirements for the level of teachers’ professionalism, achievements in science and industry, and introduction of advanced technologies. The aim of this study is to consider the training of creative teachers who think innovatively, make complex pedagogical decisions, formulate problems, and change their stereotypes. The research uses a review of literature and an interview method to arrive at its findings. As part of the findings, the research reports that using research projects are a good way to train teachers

Detection of m6A from direct RNA sequencing using a Multiple Instance Learning framework

RNA modifications such as m6A methylation form an additional layer of complexity in the transcriptome. Nanopore direct RNA sequencing captures this information in the raw current signal for each RNA molecule, enabling the detection of RNA modifications using supervised machine learning. However, experimental approaches provide only site-level training data, whereas the modification status for each single RNA molecule is missing. Here we present m6Anet, a neural network-based method that leverages the Multiple Instance Learning framework to specifically handle missing read-level modification labels in site-level training data. m6Anet outperforms existing computational methods, shows similar accuracy as experimental approaches, and generalises to different cell lines with almost identical accuracy. We demonstrate that m6Anet captures the underlying read-level stoichiometry that can be used to approximate differences in modification rates. m6Anet achieves this without retraining model parameters, enabling the transcriptome-wide identification and quantification of m6A from a single run of direct RNA sequencing.