Fusion of Imaging and Non-Imaging Data for Disease Trajectory Prediction for COVID-19 Patients

ABSTRCATPurposeThis study investigates whether graph-based fusion of imaging data with non-imaging EHR data can improve the prediction of disease trajectory for COVID-19 patients, beyond the prediction performance of only imaging or non-imaging EHR data.Materials and MethodsWe present a novel graph-based framework for fine-grained clinical outcome prediction (discharge, ICU admission, or death) that fuses imaging and non-imaging information using a similarity-based graph structure. Node features are represented by image embedding and edges are encoded with clinical or demographic similarity.ResultsOur experiments on data collected from Emory Healthcare network indicate that our fusion modeling scheme performs consistently better than predictive models using only imaging or non-imaging features, with f1-scores of 0.73, 0.77, and 0.66 for discharge from hospital, mortality, and ICU admission, respectively. External validation was performed on data collected from Mayo Clinic. Our scheme highlights known biases in the model prediction such as bias against patients with alcohol abuse history and bias based on insurance status.ConclusionThe study signifies the importance of fusion of multiple data modalities for accurate prediction of clinical trajectory. Proposed graph structure can model relationships between patients based on non-imaging EHR data and graph convolutional networks can fuse this relationship information with imaging data to effectively predict future disease trajectory more effectively than models employing only imaging or non-imaging data. Forecasting clinical events can enable intelligent resource allocation in hospitals. Our graph-based fusion modeling frameworks can be easily extended to other prediction tasks to efficiently combine imaging data with non-imaging clinical data.

How Do I Compare? The Effect of Work-Unit Demographics on Reactions to Pay Inequality

Prior research suggests that individuals react negatively when they perceive they are underpaid. Moreover, individuals frequently select pay referents who share their race and gender, suggesting that demographic similarity affects one’s knowledge of pay differences. Leveraging these insights, the authors examine whether the gender and racial composition of a work unit shapes individuals’ reactions to pay deprivation. Using field data from a large health care organization, they find that pay deprivation resulting from workers receiving less pay than their same-sex and same-race coworkers prompts a significantly stronger response than does pay deprivation arising from workers receiving less pay than their demographically dissimilar colleagues. A supplemental experiment reveals that this relationship likely results from individuals’ propensity to select same-category others as pay referents, shaping workers’ information about their colleagues’ pay. The study’s findings underscore the need to theoretically and empirically account for how demographically driven social comparison processes affect reactions to pay inequality.

Social network characteristics and alcohol use by ethnic origin: An ego-based network study on peer similarity, social relationships, and co-existing drinking habits among young Swedes

The study explores how social network determinants relate to the prevalence and frequency of alcohol use among peer dyads. It is studied how similar alcohol habits co-exist among persons (egos) and their peers (alters) when socio-demographic similarity (e.g., in ethnic origin), network composition and other socio-cultural aspects were considered. Data was ego-based responses derived from a Swedish national survey with a cohort of 23-year olds. The analytical sample included 7987 ego-alter pairs, which corresponds to 2071 individuals (egos). A so-called dyadic design was applied i.e., all components of the analysis refer to ego-alter pairs (dyads). Multilevel multinomial-models were used to analyse similarity in alcohol habits in relation to ego-alter similarity in ethnic background, religious beliefs, age, sex, risk-taking, educational level, closure in network, duration, and type of relationship, as well as interactions between ethnicity and central network characteristics. Ego-alter similarity in terms of ethnic origin, age and sex was associated with ego-alter similarity in alcohol use. That both ego and alters were non-religious and were members of closed networks also had an impact on similarity in alcohol habits. It was concluded that network similarity might be an explanation for the co-existence of alcohol use among members of peer networks.

The characteristic of Leader-member Exchange (LMX) relationship between leader and follower

The objective of this research is to understand the characteristic of the LMX relationship between leader and follower in the construction industry (in the context of a project-based organization) with followers as the provider of ideas and knowledge. This research is conducted in a construction state-own enterprise (SOE) in Indonesia, selected due to tight competition and the consequential need for innovation within short timescales. This study involves all 121 projects that existed at the time of data collection, and the unit of analysis is a dyadic relationship between general managers (as leader) and project managers (as a follower). Of these dyads, only 118 dyads can be analyzed (97.52 percent participation rate) and the analysis method used in this study is descriptive statistical analysis. The findings indicate the presence of statistic similarity between leaders and followers in case: the man’s gender 98,3%, regional origin match 42,4%, similarity level of education 39,8%, the similarity of ownership of construction certification 71,2%, the duration of the current working relationship with the leader for 13 to 24 months (49,2%), the duration of knowing current leader in the company for above 5 years (60%). Therefore, the demographic similarity is an important characteristic of high-quality LMX.

Party Elites’ Preferences in Candidates: Evidence from a Conjoint Experiment

Party elites selecting candidates are crucial for the composition of parliament. Yet, despite their pivotal position within the party, we know only little about their preferences for potential candidates and how their own backgrounds shape these preferences. This paper presents results from a conjoint experiment carried out with party delegates chosen to select the candidates for five German parties in the run-up to three state elections. Theoretical expectations derived from the principle-agent framework on delegates’ preferences in candidates are evaluated. Analyses show that delegates prefer attributes indicative of quality and socio-demographic similarity in candidates. Additionally, I show that these preferences for candidates differ between inexperienced and experienced delegates, the latter showing a stronger preference for valence attributes in candidates. These findings contribute to our understanding of the role of personal attributes of selectors for candidate selection and hold crucial implications for the composition of legislatures and long-term effects on public policy.

“Looking at Myself in the Future”: how mentoring shapes scientific identity for STEM students from underrepresented groups

Background Mentorship has been well-established in the literature as fostering scientific identity and career pathways for underrepresented minority students in science, technology, engineering, and mathematics (STEM) fields. Mentorship is prioritized by programs that aim to increase diversity and support future leadership in STEM fields, but in-depth understanding of mentorship in these contexts remains limited. Drawing on qualitative interview data, we sought to understand the relationship between mentoring and scientific identity among a diverse sample of 24 students in one such program, in order to inform program development. Results Qualitative analysis of the data revealed that mentorship, especially research mentorship, was common and played a role in formation of scientific identity. Students with research mentors tended to say they strongly identified as scientists, whereas those who lacked research mentorship varied in their level of scientific identity. In interviews, research-mentored students described mentors as colleagues who gave them opportunities to grow and as examples to look up to. Students valued mentors with whom they identified on the basis of demographic similarity or shared values, as well as those who challenged them in their academic and research endeavors. Conclusions Our analysis highlights how different mentoring experiences can contribute to development of future STEM leadership. We discuss implications for practice, including the need for tailored mentoring approaches and research-focused mentoring, and offer several recommendations for research and programming.

Open Campus Policies: How Built, Food, Social, and Organizational Environments Matter for Oregon’s Public High School Students’ Health

Open campus policies that grant access to the off-campus food environment may influence U.S. high school students’ exposure to unhealthy foods, yet predictors of these policies are unknown. Policy holding and built (walkability), food (access to grocery stores), social (school-to-neighborhood demographic similarity), and organizational (policy holding of neighboring schools) environment data were collected for 200 Oregon public high schools. These existing data were derived from the Oregon School Board Association, WalkScore.com, the 2010 Decennial Census, the 2010–2014 American Community Survey, the Supplemental Nutrition Assistance Program, TDLinex, Nielson directories, the U.S. Department of Education, the National Center for Education Statistics, and the Common Core of Data. Most (67%) of Oregon public high schools have open campus policies. Logistic regression analyses modeled open campus policy holding as a function of built, food, social, and organizational environment influences. With health and policy implications, the results indicate that the schools’ walkability, food access, and extent of neighboring open campus policy-schools are significantly associated with open campus policy holding in Oregon.

Open Campus Policies: How Built, Food, Social, and Organizational Environments Matter for Oregon’s Public High School Students’ Health

Open campus policies that grant access to the off-campus food environment influence U.S. high school students’ exposure to unhealthy foods, yet predictors of these policies are unknown. Policy holding and built (walkability), food (access to grocery stores), social (school-to-neighborhood demographic similarity), and organizational (policy holding of neighboring schools) environment data were collected for 200 Oregon public high schools. These existing data derived from the Oregon School Board Association, WalkScore.com, 2010 Decennial Census, 2010-2014 American Community Survey, Supplemental Nutrition Assistance Program, TDLinex, Nielson directories, U.S. Department of Education, National Center for Education Statistics, and Common Core of Data. Most (67%) of Oregon public high schools had open campus policies. Logistic regression analyses modeled open campus policy holding as a function of built, food, social, and organizational environment influences. With health and policy implications, results indicate that schools’ walkability, food access, and extent of neighboring open campus policy-schools are significantly associated with open campus policy holding in Oregon.