Biomedical text readability after hypernym substitution with fine-tuned large language models.

The advent of patient access to complex medical information online has highlighted the need for simplification of biomedical text to improve patient understanding and engagement in taking ownership of their health. However, comprehension of biomedical text remains a difficult task due to the need for domain-specific expertise. We aimed to study the simplification of biomedical text via large language models (LLMs) commonly used for general natural language processing tasks involve text comprehension, summarization, generation, and prediction of new text from prompts. Specifically, we finetuned three variants of large language models to perform substitutions of complex words and word phrases in biomedical text with a related hypernym. The output of the text substitution process using LLMs was evaluated by comparing the pre- and post-substitution texts using four readability metrics and two measures of sentence complexity. A sample of 1,000 biomedical definitions in the National Library of Medicine's Unified Medical Language System (UMLS) was processed with three LLM approaches, and each showed an improvement in readability and sentence complexity after hypernym substitution. Readability scores were translated from a pre-processed collegiate reading level to a post-processed US high-school level. Comparison between the three LLMs showed that the GPT-J-6b approach had the best improvement in measures of sentence complexity. This study demonstrates the merit of hypernym substitution to improve readability of complex biomedical text for the public and highlights the use case for fine-tuning open-access large language models for biomedical natural language processing.

Entropy removal of medical diagnostics.

Shannon entropy is a core concept in machine learning and information theory, particularly in decision tree modeling. To date, no studies have extensively and quantitatively applied Shannon entropy in a systematic way to quantify the entropy of clinical situations using diagnostic variables (true and false positives and negatives, respectively). Decision tree representations of medical decision-making tools can be generated using diagnostic variables found in literature and entropy removal can be calculated for these tools. This concept of clinical entropy removal has significant potential for further use to bring forth healthcare innovation, such as quantifying the impact of clinical guidelines and value of care and applications to Emergency Medicine scenarios where diagnostic accuracy in a limited time window is paramount. This analysis was done for 623 diagnostic tools and provided unique insights into their utility. For studies that provided detailed data on medical decision-making algorithms, bootstrapped datasets were generated from source data to perform comprehensive machine learning analysis on these algorithms and their constituent steps, which revealed a novel and thorough evaluation of medical diagnostic algorithms.

Impact of medical technologies may be predicted using constructed graph bibliometrics.

Scientific research is driven by allocation of funding to different research projects based in part on the predicted scientific impact of the work. Data-driven algorithms can inform decision-making of scarce funding resources by identifying likely high-impact studies using bibliometrics. Compared to standardized citation-based metrics alone, we utilize a machine learning pipeline that analyzes high-dimensional relationships among a range of bibliometric features to improve the accuracy of predicting high-impact research. Random forest classification models were trained using 28 bibliometric features calculated from a dataset of 1,485,958 publications in medicine to retrospectively predict whether a publication would become high-impact. For each random forest model, the balanced accuracy score was above 0.95 and the area under the receiver operating characteristic curve was above 0.99. The high performance of high impact research prediction using our proposed models show that machine learning technologies are promising algorithms that can support funding decision-making for medical research.

Predictors of Mortality in Extracorporeal Membrane Oxygenation Support Patients Following Major Trauma.

INTRODUCTION: The usage of extracorporeal membrane oxygenation (ECMO) in trauma patients has increased significantly within the past decade. Despite increased research on ECMO application in trauma patients, there remains limited data on factors predicting morbidity and mortality outcome. Therefore, the primary objective of this study is to describe patient characteristics that are independently associated with mortality in ECMO therapy in trauma patients, to further guide future research. METHODS: This retrospective study was conducted using the Trauma Quality Improvement Program database from 2010 to 2019. All adult (age ≥ 16 y) trauma patients that utilized ECMO were included. A Significant differences (P < 0.05) in demographic and clinical characteristics between groups were calculated using an independent t-test for normal distributed continuous values, a Mann-Whitney U test for non-normal distributed values, and a Pearson chi-square test for categorical values. A multivariable regression model was used to identify independent predictors for mortality. A survival flow chart was constructed by using the strongest predictive value for mortality and using the optimal cut-off point calculated by the Youden index. RESULTS: Five hundred forty-two patients were included of whom 205 died. Multivariable analysis demonstrated that the female gender, ECMO within 4 h after presentation, a decreased Glasgow Coma Scale, increased age, units of blood in the first 4 h, and abbreviated injury score for external injuries were independently associated with mortality in ECMO trauma patients. It was found that an external abbreviated injury score of ≥3 had the strongest predictive value for mortality, as patients with this criterion had an overall 29.5% increased risk of death. CONCLUSIONS: There is an ongoing increasing trend in the usage of ECMO in trauma patients. This study has identified multiple factors that are individually associated with mortality. However, more research must be done on the association between mortality and noninjury characteristics like Pao2/Fio2 ratio, acute respiratory distress syndrome classification, etc. that reflect the internal state of the patient.

Does temperature favour the spread of Raphidiopsis raciborskii, an invasive bloom-forming cyanobacterium, by altering cellular trade-offs?

As a tropical filamentous cyanobacterium, Raphidiopsis raciborskii has attracted much attention due to its expansion and toxin production. However, the mechanisms of its expansion to temperate regions have not been studied in detail. To address the potential strategies, the physiological and metabolomic profiles of R. raciborskii FACHB 1096 isolated from a temperate lake in China were determined and measured at different temperatures (10 °C, 15 °C, 20 °C, 25 °C, and 32 °C). The results demonstrated that temperature significantly changed cell viability, chlorophyll a content, specific growth rate, Chl a fluorescence, and filamentous shape of R. raciborskii. Low temperature decreased cell viability, specific growth rate, and photosynthetic efficiency, while the proportion of akinete and carbon fixation per unit cell were significantly increased compared with high temperature (32 °C). A constructed unimodal model indicated that filament length, cell volume, and cell length/width of R. raciborskii were significantly reduced in both high and low temperature environments. Under low-temperature conditions, R. raciborskii suffered different degrees of oxidative damage and produced corresponding antioxidant substances to resist oxidative stress, suggesting that low temperature changes the metabolic level of the cells, causing the cells to gradually switch from development to defense. Metabolomic data further confirmed that temperature change induced shifts in metabolic pathways in R. raciborskii, including starch and sucrose metabolic pathways, glutathione metabolic pathways, and the pentose phosphate pathways (PPP), as well as metabolic pathways related to the tricarboxylic acid (TCA) cycle. Our results indicated that the trade-offs of R. raciborskii cells among the growth, cell size, and metabolites can be significantly regulated by temperature, with broad implications for its global expansion in temperate waterbodies.

PlanNet: A Generative Model for Component-Based Plan Synthesis.

We propose a novel generative model named as PlanNet for component-based plan synthesis. The proposed model consists of three modules, a wave function collapse algorithm to create large-scale wireframe patterns as the embryonic forms of floor plans, and two deep neural networks to outline the plausible boundary from each squared pattern, and meanwhile estimate the potential semantic labels for the components. In this manner, we use PlanNet to generate a large-scale component-based plan dataset with 10 K examples. Given an input boundary, our method retrieves dataset plan examples with similar configurations to the input, and then transfers the space layout from a user-selected plan example to the input. Benefiting from our interactive workflow, users can recursively subdivide individual components of the plans to enrich the plan contents, thus designing more complex plans for larger scenes. Moreover, our method also adopts a random selection algorithm to make the variations on semantic labels of the plan components, aiming at enriching the 3D scenes that the output plans are suited for. To demonstrate the quality and versatility of our generative model, we conduct intensive experiments, including the analysis of plan examples and their evaluations, plan synthesis with both hard and soft boundary constraints, and 3D scenes designed with the plan subdivision on different scales. We also compare our results with the state-of-the-art floor plan synthesis methods to validate the feasibility and efficacy of the proposed generative model.

Network Analysis of Academic Medical Center Websites in the United States.

Healthcare resources are published annually in repositories such as the AHA Annual Survey DatabaseTM. However, these data repositories are created via manual surveying techniques which are cumbersome in collection and not updated as frequently as website information of the respective hospital systems represented. Also, this resource is not widely available to patients in an easy-to-use format. Network analysis techniques have the potential to create topological maps which serve to aid in pathfinding for patients in their search for healthcare services. This study explores the topological structure of forty United States academic health center websites. Network analysis is utilized to analyze and visualize 48,686 webpages. Several elements of network structure are examined including basic network properties, and centrality measures distributions. The Louvain community detection algorithm is used to examine the extent to which these techniques allow identification of healthcare resources within networks. The results indicate that websites with related healthcare services tend to form observable clusters useful in mapping key resources within a hospital system.

Digital Impact Factor: A Quality Index for Educational Blogs and Podcasts in Emergency Medicine and Critical Care.

STUDY OBJECTIVE: Given the popularity of educational blogs and podcasts in medicine, learners and educators need tools to identify trusted and impactful sites. The Social Media Index was a multi-sourced formula to rank the effect of emergency medicine and critical care blogs. In 2022, a key data point for the Social Media Index became unavailable. This bibliometric study aimed to develop a new measure, the Digital Impact Factor, as a replacement. METHODS: The Digital Impact Factor incorporated modern measures of website authority and reach. This formula was applied to a cross-sectional study of active emergency medicine and critical care blogs and podcasts. For each website, we generated a Digital Impact Factor score based on Ahrefs Domain Rating and the follower count of the websites' pages from 8 social media platforms. A series of Spearman correlations provided evidence of association by comparing a rank-ordered list to rank lists derived from the Social Media Index over the last 5 years. The Bland-Altman analysis assessed for agreement. RESULTS: The authors identified 88 relevant websites with a median Ahrefs Domain Rating of 28 (range 0 to 71, maximum 100) and total social media followership count across 8 platforms of 1,828,557. The Domain Rating and individual social media followership scores were normalized based on the highest recorded values to yield the Digital Impact Factor (median 4.57; range 0.02 to 9.50, maximum 10). The correlation between the 2022 Digital Impact Factor and the 2021 Social Media Index was 0.94 (95% confidence interval 0.89 to 0.97; p<.001; n=41 rankings correlated), suggesting that they measure similar constructs. The Bland-Altman plot also demonstrated fair agreement between the 2 scores. CONCLUSION: The Digital Impact Factor is a measure of the relative effect of educational blogs and podcasts within emergency medicine and critical care.

Influence of ICU Surge and Capacity on COVID Mortality Across U.S. States and Regions During the COVID-19 Pandemic.

BACKGROUND: To describe the influence of COVID-19 caseload surges and overall capacity in the intensive care unit (ICU) on mortality among US population and census divisions. METHODS: A retrospective analysis of the national COVID ActNow database between January 1, 2021 until March 1, 2022. The main outcome used was COVID-19 weekly mortality rates, which were calculated and incorporated into several generalized estimation of effects models with predictor variables that included ICU bed capacity, as well as ICU capacity used by COVID cases while adjusting for ratios of vaccinations in populations, case density, and percentage of the population over the age of 65. RESULTS: Each 1% increase in general ICU capacity is correlated with approximately 5 more weekly deaths from COVID-19 per 100,000 population and each percentage increase in the number of patients with COVID-19 admitted to the ICU resulted in approximately 10 more COVID-19 deaths per week per 100,000 population. Significant differences in ability to handle caseload surges were observed across US census divisions. CONCLUSIONS: A strong association was observed between COVID-19 ICU surges, overall ICU surge, and increased mortality. Further research is needed to reveal best practices and public health measures to prevent ICU overcrowding amidst future pandemics and disaster responses.

Correction: The Story of #GetMePPE and GetUsPPE.org to Mobilize Health Care Response to COVID-19 : Rapidly Deploying Digital Tools for Better Health Care.

[This corrects the article DOI: 10.2196/20469.].

Primary Perspectives in Meme Utilization as a Digital Driver for Medical Community Engagement and Education Mobilization: Pre-Post Study.

BACKGROUND: Memes have gone "viral," gaining increasing prominence as an effective communications strategy based on their unique ability to engage, educate, and mobilize target audiences in a call to action through a cost-efficient and culturally relevant approach. Within the medical community in particular, visual media has evolved as a means to influence clinical knowledge transfer. To this end, the GetWaivered (GW) project has leveraged memes as part of a behavioral economics toolkit to address one of the most critical public health emergencies of our time-the 20-year opioid epidemic. As part of a multidimensional digital awareness campaign to increase Drug Enforcement Administration (DEA)-X waiver course registration, GW investigated the results of meme usage in terms of impressions, website traffic, and ultimately user acquisition, as determined by web-based training enrollment and attendance outcomes. OBJECTIVE: The objective of this study was to determine the efficacy of implementing humor-based promotional content versus the traditional educational model, and how the translation of the increase in engagement would increase the participant count and website traffic for GW's remote DEA-X waiver training. METHODS: The approach to this study was based on 2 time frames (pre- and postcampaign). During April-July 2021, we developed a campaign via advertisements on Facebook, Twitter, Instagram, and the GW website to expand outreach. These memes targeted medical professionals with the ability to prescribe buprenorphine. The time frame of this campaign measured engagement metrics and compared values to preceding months (January-March 2021) for our GetWaivered website and social media pages, which translated to registrants for our remote DEA-X waiver training. RESULTS: By the end of July 2021, a total of 9598 individuals had visited the GW website. There was an average of 79.3 visitors per day, with the lowest number of daily visitors being 0 and the highest being 575. CONCLUSIONS: The use of memes may provide a medium for social media engagement (likes, comments, and shares) while influencing viewers to pursue a proposed action, such as e-training registration.

Chinese medicine Yangmo decoction ameliorates intrauterine adhesion prognosis following hysteroscopic adhesiolysis. / ä¸­è¯å »è†œæ–¹æ”¹å–„å®«è ”é•œä¸‹å®«è ”ç²˜è¿žåˆ†ç¦»æœ¯åŽå®«è ”ç²˜è¿žçš„é¢„åŽ.

OBJECTIVES: Hysteroscopic adhesiolysis (HA) remains the mainstay on treatment for intrauterine adhesions (IUA). The fertility outcome of patients with moderate and severe intrauterine adhesions after HA is still far from satisfactory. Estrogen combined with progesterone is the most common treatment; however, they do not help in improving the fertility rate to the maximum because of the limitations. This retrospective, non-randomized controlled study will assess the effects of traditional Chinese medicine Yangmo decoction after HA in restoration of the endometrium and improvement of the fertility rate. METHODS: A total of 427 patients, who met the inclusion criteria, aged between 20 and 45 years and diagnosed with moderate or severe IUA underwent HA at the Third Xiangya Hospital from January to August 2021, were enrolled for this study. Participants were assigned into 2 groups: A Yangmo decoction group (n=213, patients were given Yangmo decoction consisting of Ginseng flower, Sanchi flower, Daidai flower, Snow lotus, Licorice and so on after HA), and an estrogen and progesterone group (n=214, patients were given estrogen and progesterone after HA). The following basic information was collected retrospectively for both groups, including age, parity, history of abortion, menstrual status, and times of hysteroscopic interventions. American Fertility Society (AFS) score was used by a senior surgeon and the density of opening of endometrial glands was evaluated during HA. The parameters were obtained from three-dimensional transvaginal ultrasound (3D-TVUS) preoperatively and postoperatively, to evaluate the efficacy of Yangmo decoction, estrogen, and progesterone. All patients were followed up on telephone to determine the fertility rate until 6 months from the last HA. RESULTS: Based on the basic information collected preoperatively, there were no significant differences between the groups (all P>0.05). Postoperatively, patients in the Yangmo decoction group had a better surgical success rate with a more significant AFS reduction (P<0.001), better density of opening of endometrial glands in the uterine cavity (P<0.000 1) after HA, and a better fertility rate (40.4%) in the time of 6 months after the last HA than those of the estrogen and progesterone group. CONCLUSIONS: Yangmo decoction has better therapeutic efficacy in the treatment of intrauterine adhesion after HA than the combined effect of estrogen and progesterone. Yangmo decoction helps restore the endometrium and improve the fertility rate, therefore, it can be adopted as a routine practice for IUA patients who have fertility requirements.

Emoji Education: How Students Can Help Increase Health Awareness by Making Emojis.

Emojis can improve health communication, especially when incorporating emojis into traditionally word-only texts. Beyond improving communication, emojis also offer greater access to health care, especially for vulnerable and marginalized populations with limited health literacy. A recent study found that 94% of patients with limited health literacy preferred health reports with emojis. Moreover, health officials are considering adding emojis to cardiopulmonary resuscitation guidelines and public health guidelines for handwashing. As the world evolves with new technology and new methods of communication, we must also evolve the language and method we use to communicate health information to patients. In this viewpoint, we aim to discuss the methods health care professionals can use to develop novel communication methods using emojis and the benefits of their incorporation into health care communication.

Analyzing opioid-use disorder websites in the United States: An optimized website usability study.

Background: As the United States continues to tackle the opioid epidemic, it is imperative for digital healthcare organizations to provide Internet users with accurate and accessible online resources so that they can make informed decisions with regards to their health. Objective: The primary objectives were to adapt and modify a previously established usability methodology from literature, apply this modified methodology in order to perform usability analysis of opioid-use-disorder (OUD)-related websites, and make important recommendations that OUD-related digital health organizations may utilize to improve their online presence. Methods: A list of 208 websites (later refined) was generated for usability testing using a modified Google Search methodology. Four keywords were chosen and used in the search: "DEA-X Waiver Training", "opioid-use-disorder (OUD) Initiatives", "Buprenorphine Assisted Treatment", and "Opioid-Use Disorder Websites". Usability analysis was performed concurrently with optimization of the methodology. OUD websites were analyzed and scored on several usability categories established by previous literature. Results: "DEA-X Waiver Training" yielded websites that scored the highest average in "Accessibility" (0.84), while "Opioid-Use Disorder Websites" yielded websites that scored the highest average in "Content Quality" (0.67). "Buprenorphine Assisted Treatment" yielded websites that scored the highest average across "Marketing" (0.52), "Technology" (0.89), "General Usability" (0.69), and "Overall Usability" (0.68). "Technology" and "Marketing" were the highest and lowest scoring usability categories, respectively. T-test analysis revealed that each usability, except "Marketing" had a pair of one or more keywords that were significantly different with a p-value that was equal to or less than 0.05. Conclusions: Based on the study findings, we recommend that digital organizations in the OUD space should improve their "General Usability" score by making their websites easier to find online. Doing so, may allow users, especially individuals in the OUD space, to discover accurate information that they are seeking. Based on the study findings, we also made important recommendations that OUD-related digital organizations may utilize in order to improve website usability as well as overall reach.

Website Usability Analysis of U.S. Military Residency Programs.

INTRODUCTION: The Military Match is the residency matching system for medical students attending the Uniformed Services University of Health Sciences, and the students were funded by the Health Professions Scholarship Program through the U.S. Army, Air Force, and Navy. To evaluate and compare military residency programs, students use residency program websites. Often, the residency program's website serves as a key source, or the only point of reference, when considering residency options, especially during times when face-to-face interactions are limited.This report aims to provide a systematic evaluation of military residency programs and their websites. MATERIALS AND METHODS: Utilizing a previously published website usability scoring system, military residency programs were categorized to objectively and quantitatively analyze their websites. Usability was divided into four categories for quantifiable analysis: accessibility, marketing, content quality, and technology. The methodology for this analysis was replicated from published reports that have examined healthcare website usability. Each website was analyzed and scored in four categories: accessibility, content quality, marketing, and technology. A "General Usability" score was calculated for each website using a composite of the key factors within the four categories. An overall score was generated utilizing the weighted percentage across all four categories. To address deficiencies of the original methodology, a secondary analysis was performed on the listed websites utilizing an automated methodology for website usability. RESULTS: A comprehensive list of 125 Accreditation Council for Graduate Medical Education U.S. Military residency program websites was compiled. Of these, 96 programs and 106 websites were evaluated. The primary analysis employing usability methodology identified technology as the highest ranked category with a score of 0.749 (SD ± 0.039) (SE 0.005) (P < .05). Marketing and content quality were the lowest scoring categories with mean scores of 0.414 (SD ± 0.054) (SE 0.006) and 0.428 (SD ± 0.229) (SE 0.027), respectively (P < .05). There was no significant difference in overall usability rankings or scores among the 96 residency program websites across the three branches (P < .05).Secondary analysis with the new usability methodology demonstrated military residency websites to exhibit more external backlinking compared to internal backlinking (P < 0.05) and no social media backlinking to any of the 106 analyzed websites. When comparing the three services, the Army had significantly lower external backlinking ranking 43.4 (P < .05) and overall backlinking ranking 56.4 (P < 0.05) when compared to the Navy (mean 48.8 and 71.7, and 43.4). There were no other differences in backlinking rankings across the three branches. CONCLUSIONS: Residency websites have become a primary way to communicate information to applicants. By assessing the overall usability of the various military residency websites, we determined the effectiveness of these websites to relay information to prospective students interested in applying for military residency. We predict that by improving website accessibility, residency programs increase their effectiveness at communicating information to potential applicants and increase interest in military residency programs.

American Anesthesiology Residency Programs: Website Usability Analysis.

BACKGROUND: The Association of American Medical Colleges has recently issued recommendations for the upcoming 2022-2023 application cycle that residency programs should conduct all interviews for this upcoming application cycle over the web. In light of these recommendations, many students will have limited exposure to anesthesiology programs and will rely on information gleaned digitally. This change means that the aspects of program websites used to provide information, such as size, structure, location, requirements, and contact information, will be crucial in helping prospective residents decide where and how to apply in the future. An evaluation of website usability, which includes initial appearance along with factors that influence its ease of navigation and convenience of use, can thus be applied to anesthesiology residency websites. Areas of need can be targeted to increase web presence and provide effective pathways to exhibit the different attributes of their programs to future applicants. OBJECTIVE: This study aimed to compile a list of US anesthesiology residency programs and their websites while objectively analyzing the websites using a formally published usability scoring system, as well as to identify positive and negative trends to offer areas of improvement among anesthesiology residency websites. METHODS: We included only 114 US anesthesiology residency program websites in our sample set, since some websites we analyzed showed errors or inconclusive. Website usability was separated into 4 distinct categories for analysis based on methodology outlined in previous literature on both health care website usability and residency website usability. The 4 categories were Accessibility, Marketing, Content Quality, and Technology. Each website was then analyzed and scored based on key components highlighted within the 4 categories. The multiple factors were then graded using a percentage system to create a comprehensive score for each program. RESULTS: The highest scoring category was Content Quality (mean 4.7, SD 2.48, SE 0.23). The lowest scoring category was Technology (mean 0.9, SD 0.38, SE 0.04). CONCLUSIONS: Through the application of a health care website usability framework, multiple anesthesiology residency programs were analyzed and scored in the areas of Accessibility, Marketing, Content Quality, and Technology, which allowed us to determine the effectiveness of the usability of these websites to convey information to their end user. Websites must communicate vital information, with usability at the forefront, to continue to grow, especially as the United States faces challenges due to the COVID-19 pandemic. Our recommendation is that anesthesiology programs should strive to improve website usability to increase the ease by which applicants can collect vital information about anesthesiology programs. A few proposed solutions include making changes such as decreasing error pages on websites, migrating away from using in-line cascading style sheets, and improving web page loading speeds to improve the Technology category.