A natural language processing system for national COVID-19 surveillance in the US Department of Veterans Affairs

Timely and accurate accounting of positive cases has been an important part of the response to the COVID-19 pandemic. While most positive cases within Veterans Affairs (VA) are identified through structured laboratory results, some patients are tested or diagnosed outside VA so their clinical status is documented only in free-text narratives. We developed a Natural Language Processing pipeline for identifying positively diagnosed COVID-19 patients and deployed this system to accelerate chart review. As part of the VA national response to COVID-19, this process identified 6,360 positive cases which did not have corresponding laboratory data. These cases accounted for 36.1% of total confirmed positive cases in VA to date. With available data, performance of the system is estimated as 82.4% precision and 94.2% recall. A public-facing implementation is released as open source and available to the community. © ACL 2020.All right reserved.

Hyflex for Successful Student Veteran Engineering Education: Say it Like You Mean It

There has been increased attention on producing engineers that are technically proficient while having many professional skills such as organization, time management, communication, and leadership. Across organization types, especially academia, veterans are admired by their peers for their professionalism and communication skills. Student veterans have trained and taken online classes in diverse and remote environments. They are accustomed to learning under ideal and less than ideal circumstances. The combined traits of increased professionalization, prior experience with online learning, and persistence position student veterans to perform as well or better than their traditional college-aged peers during the COVID-19 crisis. In a study of the effectiveness of Hyflex (Hybrid Flexible) learning conducted in the School of Engineering at The Citadel, forced-choice and free text survey responses showed that student veterans match with and differ from traditional college-aged students in important ways. Results from this study can be used to guide best practices in the Hyflex educational model, in order to better serve the student veteran demographic and all students. In particular, student veteran responses coalesce around a focus on effectiveness and time management concerns, as many have families and other external obligations. As a result, student veterans simultaneously want more Hyflex educational options going forward, however they want Hyflex implementation strategies to be refined and executed better in the future with more long-term planning. Active duty and student veterans can serve vital roles in the engineering classroom, modeling appropriate communication strategies for traditional students as well as connecting their global knowledge with the course content, enriching all students' understanding. Faculty and traditional students can benefit from this unique demographic if they are aware of their skills and experiences. This paper presents some of the issues and concerns of active duty and veterans pursuing an engineering degree compared to their traditional student counterparts when institutions pivot to alternative instructional delivery, specifically Hyflex. © American Society for Engineering Education, 2022

Using natural language processing on free-Text clinical notes to identify patients with long-Term COVID effects

As of May 15th, 2022, the novel coronavirus SARS-COV-2 has infected 517 million people and resulted in more than 6.2 million deaths around the world. About 40% to 87% of patients suffer from persistent symptoms weeks or months after their original infection. Despite remarkable progress in preventing and treating acute COVID-19 conditions, the clinical diagnosis of long-Term COVID remains difficult. In this work, we use free-Text clinical notes and natural language processing (NLP) techniques to explore long-Term COVID effects. We first obtain free-Text clinical notes from 719 outpatient encounters representing patients treated by physicians at Emory Clinic to detect patterns in patients with long-Term COVID symptoms. We apply state-of-The-Art NLP frameworks to automatically identify patients with long-Term COVID effects, achieving 0.881 recall (sensitivity) score for note-level prediction. We further interpret the prediction outcomes and discuss potential phenotypes. Our work aims to provide a data-driven solution to identify patients who have developed persistent symptoms after acute COVID infection. With this work, clinicians may be able to identify patients who have long-Term COVID symptoms to optimize treatment. © 2022 Owner/Author.

Changes, Longings, and Potentials for Future Pedagogical Practices: Investigating University Teachers’ Experiences During the Spring 2020 Lockdown

This paper takes a quantitative ethnographic approach to understand how the lockdown in the spring of 2020 affected the teachers’ work situation at a large university in Denmark. Based on free-text responses from survey data, we create epistemic networks to explore how teachers articulate changes, longings, and potentials for their own future digital teaching practices based on their experiences of the first lockdown. The findings illustrate that interaction and human contact, which play a pivotal role in teachers’ didactical skillsets and efforts to create good learning environments, were missing during the lockdown. The study also highlights that teachers’ perceptions of the inability of digital, technological solutions appear to relate to on-campus teaching ideals. As such, this paper argues that future crisis-handling, as well as developments and refinements of digital teaching formats, at the university, should be attentive to support and foster better areas of contact and interaction between students and teachers. © 2022, Springer Nature Switzerland AG.