Unique Videos Research Articles

Traffic as an Urban Health Determinant: Coverage of the New York City Congestion Pricing Plan on TikTok.

Social listening can provide deeper insight into public sentiment regarding policy proposals, as seen with the congestion pricing initiative in New York City (NYC). The purpose of this study was to assess the content of a sample of TikTok videos related to congestion pricing in NYC. A total of 100 unique videos were extracted from the hashtag #nyccongestionpricing on TikTok and coded for the presence of the following information based on four different areas of focus: video metrics, details about pricing plan, effects of pricing plan, and user-related issues. In total, the 100-video sample received 17,375,818 views, 1,285,610 likes, 89,883 favorites, and 100,634 comments. Only 3 of the 25 content characteristics were present in a majority (> 50) of the videos. These were mentions location (below 60th street of Manhattan) (N = 65), mentions cost (N = 61), and mentions cons of pricing (N = 56). Mentioning location had a significant effect on the views (p = 0.0163) and likes (p = 0.0225) a video received. Mentioning cost had a statistical impact on the views (p = 0.0098), likes (p = 0.0128), favorites (p = 0.0128), and comments (p = 0.0030) that the video received. Mentioning the MTA and benefits of the congestion plan significantly lowered metrics of engagement. Themes that emerged were personal, often negative and tended to focus more on cost to citizens and location, despite the evidence of negative public health impacts. This research can be used as a springboard for future research that examines social listening at the intersection of public health and policy.

“Do all bisexuals have this power?”: An exploratory study of “crippling nicotine addiction,” identity, and other emergent themes in vaping messages on QueerTok

Nicotine and tobacco use disproportionally affects sexual and gender minority (SGM) populations in the United States. Social media narratives may contribute to these disparities. This qualitative study delineated perceptions and experiences depicted in SGM-related videos about nicotine vaping on TikTok. Young adult researchers engaged in every step of the research process, adding an insider perspective. Using four TikTok accounts, we used vaping and SGM-related search terms to sample videos in March–April 2022. Three TikTok accounts collected SGM-specific videos; a fourth provided non-SGM specific videos for comparison. We iteratively sorted 303 unique videos into 32 a priori and emergent codes and identified themes in SGM videos and comparison videos. In their videos, creators displayed awareness of and ambivalence toward vaping and nicotine dependence. SGM videos reflected vaping as a salient feature of identity and a consideration in romantic partnership. Studying video-based social media platforms, like TikTok, using an insider-engaged qualitative lens promotes rich interpretation of content to identify prevalent and emerging messages, which can inform appropriate interventions for SGM young people.

Characterization of the dissemination of canine cancer misinformation on YouTube.

YouTube is the third most popular app in the world and continues to grow each year while it reaches over 2 billion users a month. A variety of veterinary topics are addressed on YouTube but to date there have been no studies analysing misinformation of various canine cancer topics on YouTube or social media. This study described the characteristics of 99 unique videos and used the validated DISCERN quality criteria for consumer health information and the Patient Education Materials Assessment Tool (PEMAT) to characterize their usefulness. The overall median DISCERN quality score was 3 (out of 5), the median PEMAT understandability score was 72%, and 61% of videos contained little to no misinformation. 53% of videos were created by veterinarians and this subset had significantly higher PEMAT understandability and DISCERN quality scores compared with client-created content (p = .0228 and p ≤ .0001, respectively). Videos with little to no misinformation had statistically significant higher DISCERN quality scores (3 vs. 2, p = .0001). There was no statistical significance between misinformation levels and video length, PEMAT understandability, thumbs up/view, or views/mo. These data reveal similar rates of misinformation in videos on canine cancer compared to that reported for various human cancer topics. This study highlights the need for veterinarians to guide clients to more reliable and understandable information regarding their pet's health.

“Influencing the influencers:” a field experimental approach to promoting effective mental health communication on TikTok

A substantial body of social scientific research considers the negative mental health consequences of social media use on TikTok. Fewer, however, consider the potentially positive impact that mental health content creators (“influencers”) on TikTok can have to improve health outcomes; including the degree to which the platform exposes users to evidence-based mental health communication. Our novel, influencer-led approach remedies this shortcoming by attempting to change TikTok creator content-producing behavior via a large, within-subject field experiment (N = 105 creators with a reach of over 16.9 million viewers; N = 3465 unique videos). Our randomly-assigned field intervention exposed influencers on the platform to either (a) asynchronous digital (.pdf) toolkits, or (b) both toolkits and synchronous virtual training sessions that aimed to promote effective evidence-based mental health communication (relative to a control condition, exposed to neither intervention). We find that creators treated with our asynchronous toolkits—and, in some cases, those also attending synchronous training sessions—were significantly more likely to (i) feature evidence-based mental health content in their videos and (ii) generate video content related to mental health issues. Moderation analyses further reveal that these effects are not limited to only those creators with followings under 2 million users. Importantly, we also document large system-level effects of exposure to our interventions; such that TikTok videos featuring evidence-based content received over half a million additional views in the post-intervention period in the study’s treatment groups, while treatment group mental health content (in general) received over three million additional views. We conclude by discussing how simple, cost-effective, and influencer-led interventions like ours can be deployed at scale to influence mental health content on TikTok.

Quality and Reliability Analysis of YouTube as a Source for Patient Education on Dupuytren’s Contracture

This study seeks to assess the quality and reliability of YouTube videos on Dupuytren's contracture. The first 50 unique videos on Dupuytren's contracture were evaluated by searching YouTube for Dupuytren's contracture. Video metrics, source, and content type were recorded. Video reliability was assessed using the Journal of American Medical Association (JAMA) Benchmark criteria. Video educational quality was assessed using the Global Quality Score (GQS) and a Dupuytren's Contracture-Specific Score (DC-SS). The total number of views for all 50 videos evaluated was 1,908,608 (mean, 38,172.16 ± 5,502.45 views). The mean reliability (JAMA) score was 2.21 ± 0.69 (range 0-4), the mean educational quality (GQS)score was 2.80 ± 1.28 (range 1-5), and the mean disease-specific (DC-SS) score was 6.05 ± 2.17 (range 0-15). Nonphysician health care professionals had the most popular videos, but the lowest DC-SS. GQS varied based on the video source, with physician-uploaded videos having the highest average quality scores. Physician source was an independent positive predictor of higher quality (GQS) (β= 0.477). Videos on Dupuytren's contracture were frequently viewed on YouTube but had overall low educational quality and reliability. Of the videos that discussed collagenase as a treatment option, 40% failed to mention percutaneous needle aponeurotomy. Patients may be exposed to an incomplete set of treatment options. Educational content on YouTube should be interpreted cautiously and proper in-office education and high-quality resources for Dupuytren's contracture should be provided by physicians. Therapeutic IV.

The Development of Educational YouTube Videos-Based Learning Media: Learners’ Need Analysis

This study aims to analyze the needs of students in learning in the speaking for general communication course in order to develop Educational YouTube Videos-based learning media. The integration of YouTube in learning speaking can be a valid resource for teaching lecturers to improve their lessons with various types of content that are interesting and useful and in accordance with the material being taught, and this integration is also expected to motivate students to learn. This study uses a qualitative research design by distributing questionnaires and interviews to find out the problems that occur in the learning process and what students want in improving their speaking skills. As a result, students still experience problems in speaking ability, but the biggest problem students’ face is related to fluency. With regard to the problem of fluency, the results of the interviews show that the desired way of learning is that students want to be more able to practice speaking by having lots of dialogue and direct interaction, displaying unique videos related to the material being taught so that it is not boring and can be understood easily. Easy, using YouTube videos to get used to listening to native speakers speak more. Learning will be more interesting if film media is added, such as identifying clips from a film.

Broadcast health: Leveraging YouTube for community health education.

In community health, there is great demand but limited time for the delivery of staff and patient education. During the COVID-19 pandemic, evolving needs necessitated that health education be provided in an accurate and timely manner. This paper describes the development, implementation, and evaluation of a YouTube channel designed to disseminate health education to a wide audience of public health workers and patient populations. This project, divided into three phases, originated within shelter-based care, providing education to shelter staff (Phase 1) and overtime has evolved to provide education within the Community Health Worker Hub at a major teaching hospital for community health workers (Phase 2) and the populations they serve (Phase 3). Further, during phase 3, the project developer used an artificial intelligence (AI) platform to increase the reach of the YouTube channel. Over a span of 21 months, 18 unique videos have garnered 489 views. Clinicians and educators can leverage technology platforms for health education delivery, increasing the reach of their work while meeting the demands of the profession. While the quality of some information on YouTube may be poor, patients and students turn to this platform for health education. It is imperative that public health nurses embrace this medium, rather than push against it. By creating high-quality content, educating students and patients about DISCERN and PEMAT tools, and guiding patients to credible sources, public health nurses may ameliorate the standards of health education on YouTube.

O047 Representations of Sleep in Social Media: A Tik Tok Case Study

Abstract Introduction Social media platforms such as Tik Tok have broadened and potentially democratised access to health information. The narratives and quality of information about sleep on TikTok has not been studied. We sought to understand sentiment towards sleep, the types of sleep problems represented by users, and the nature and content of advice about sleep provided on TikTok. Methods All videos associated with the keyword “sleep” were collected twice daily using the keyword search function of the Ensemble data TikTok API between 4th-17th November 2022 for Great Britain, Australia, New Zealand and the USA. 7,373 unique videos were examined. After screening, videos were coded to a customised sleep narrative-specific code set. 1,913 discrete sleep-related videos were labelled according to 33 unique categories and underwent computational sentiment analysis. Results Overall, sleep was framed more positively than negatively, particularly for sleep advice. Videos about sleep problems were typically framed positively through humour. The main sleep narratives represented included (1) problems with sleep quality, initiation or maintenance due to factors such as children, worries or use of electronic media, and (2) advice for improving sleep quality through the use of sleep-promoting soundscapes, advice on improving children’s sleep, and implementation of routines/habits to support sleep. Discussion Platforms such as Tik Tok provide opportunities to disseminate timely, engaging, and evidence-based information about sleep health to a wide community. However, they equally carry risks of mass distribution of false information that warrants broader oversight and critical examination.

Popular contraception videos on TikTok: An assessment of content topics

ObjectivesThis study aimed to explore the contraception content topics on the highly popular social media platform TikTok to understand this influential information source patients may bring into birth control counseling visits. Study designUtilizing a snapshot method, we collected the 100 most popular TikTok English-language videos for each of the eight contraception-specific search terms. We abstracted video content topics addressed and creator attributes. Two independent abstractors coded each video with high agreement, with a third coder adjudicating discrepancies. ResultsThe final data set included 700 unique videos, with a total of 1.18 billion views, 131 million likes, 1.5 million comments, and 4.1 million shares. In these videos, the most common topics addressed were patient experience (n = 365, 52.1%) and logistics of use (n = 351, 50.1%). Health care professionals created only 19.3% of videos (n = 135), but these videos accounted for a larger portion of the total video views (41.3%). Health care professionals largely made educational videos (92.6%) compared to 22.5% of videos coded as educational videos when made by perceived non–health care professionals. A small number of prolific video creators developed the majority of videos made by health care professionals, with 91 (67.4%) made by six TikTok creators. ConclusionsTikTok contains highly accessible contraception content, which garners high viewership, especially when created by health care providers. Clinicians should be aware of TikTok’s potential to influence patients prior to contraceptive counseling visits and recognize this platform as a public health instrument to disseminate contraceptive information to a key demographic. ImplicationsFew previous studies have examined the presence and popularity of contraception content topics on TikTok; little is known about the health content on this highly popular platform. Providers should be aware of the contraceptive topics on TikTok for understanding both patient perceptions and the potential for health education through this media.

Educational Quality of YouTube Videos for Patients Undergoing Elective Procedures

IntroductionYouTube has become a main resource used by patients for self-education on medicine. It is important for surgeons to understand the quality and reliability of videos that patients are likely to view about elective procedures. MethodsVideos were categorized by view count and content creators. The top 20 videos for each term, sorted by relevance, were evaluated using DISCERN criteria, a question set externally validated to assess the quality of information regarding health treatment choices. DISCERN score (DS) closer to 5 indicate higher quality information and 1 indicates the opposite. Total scores were given: 15-26 (very poor), 27-38 (poor), 39-50 (fair), 51-62 (very good), and 63-75 (excellent). Search terms included “Wound Care”, “Skin Grafting”, “Tracheostomy”, and “percutaneous endoscopic gastrostomy tube placement”. ResultsIn total, 80 unique videos were evaluated with a total view of 8,848,796. The mean overall DS was 2.15 and a mean bias DS was 2.46. The median DISCERN total score for each key term was Tracheostomy: 35 (poor), Skin Grafting 26 (very poor), percutaneous endoscopic gastrostomy: 32 (poor), and Wound Care: 40 (fair). ConclusionsYouTube videos surrounding elective procedures should be viewed cautiously in patient education despite wide availability. The videos in this study show high levels of bias and low DS. Healthcare providers should be aware of poor-quality consumer health information often disseminated in online media such as YouTube.

The Quality of MitraClip™ Content on YouTube.

Objective YouTube (YouTube LLC,San Bruno, California, United States) is used as a primary resource for many patients looking to gain healthcare knowledge. Recently, YouTube made efforts to increase the quality of posted content by accrediting trusted healthcare sources. With an increasing emphasis being placed on minimally invasive options, this study was done to investigate the quality of YouTube videos on MitraClip™ (Abbott Laboratories,Chicago, Illinois, United States) with respect to patient education. Methods YouTube was searched using the keyword "MitraClip". A total of 66 videos were evaluated, with 32 of those videos being included for final analysis after applying exclusionary criteria. Three independent reviewers separately scored the videos using the Global Quality Scale. Likes, dislikes, views, comments, and dates of upload were also recorded. Two-tailed t-tests were used to determine statistical significance. Results MitraClip videos on YouTube proved to be of medium quality, receiving an average Global Quality Scale score of 3.39. When stratified by the new YouTube accreditation process, those with accreditation had a significantly higher Global Quality Scale score of 4.11, while non-accredited videos had an average Global Quality Scale score of 3.12 (p<0.01). Shorter and more patient-friendly videos were also significantly lower in quality (p<0.05). Conclusion The YouTube accreditation process has demonstrated initial success at regulating the quality of MitraClip content, thereby reducing the spread of misinformation. However, this progress is undermined by the lack of unique videos present on the platform. Increasing the amount of original content about MitraClip may allow viewers to diversify their educational sources and ultimately gain a better understanding of the procedure.

YouTube's recommendation algorithm is left-leaning in the United States.

With over two billion monthly active users, YouTube currently shapes the landscape of online political video consumption, with 25% of adults in the United States regularly consuming political content via the platform. Considering that nearly three-quarters of the videos watched on YouTube are delivered via its recommendation algorithm, the propensity of this algorithm to create echo chambers and deliver extremist content has been an active area of research. However, it is unclear whether the algorithm may exhibit political leanings toward either the Left or Right. To fill this gap, we constructed archetypal users across six personas in the US political context, ranging from Far Left to Far Right. Utilizing these users, we performed a controlled experiment in which they consumed over eight months worth of videos and were recommended over 120,000 unique videos. We find that while the algorithm pulls users away from political extremes, this pull is asymmetric, with users being pulled away from Far Right content stronger than from Far Left. Furthermore, we show that the recommendations made by the algorithm skew left even when the user does not have a watch history. Our results raise questions on whether the recommendation algorithms of social media platforms in general, and YouTube, in particular, should exhibit political biases, and the wide-reaching societal and political implications that such biases could entail.

Content Quality of YouTube Videos About Pain Management After Cesarean Birth: Content Analysis.

YouTube is an increasingly common source of health information; however, the reliability and quality of the information are inadequately understood. Several studies have evaluated YouTube as a resource during pregnancy and found the available information to be of poor quality. Given the increasing attention to postpartum health and the importance of promoting safe opioid use after birth, YouTube may be a source of information for birthing individuals. However, little is known about the available information on YouTube regarding postpartum pain. The purpose of this study is to systematically evaluate the quality of YouTube videos as an educational resource for postpartum cesarean pain management. A systematic search of YouTube videos was conducted on June 25, 2021, using 36 postpartum cesarean pain management-related keywords, which were identified by clinical experts. The search replicated a default YouTube search via a public account. The first 60 results from each keyword search were reviewed, and unique videos were analyzed. An overall content score was developed based on prior literature and expert opinion to evaluate the video's relevance and comprehensiveness. The DISCERN instrument, a validated metric to assess consumer health information, was used to evaluate the reliability of video information. Videos with an overall content score of ≥5 and a DISCERN score of ≥39 were classified as high-quality health education resources. Descriptive analysis and intergroup comparisons by video source and quality were conducted. Of 73 unique videos, video sources included medical videos (n=36, 49%), followed by personal video blogs (vlogs; n=32, 44%), advertisements (n=3, 4%), and media (n=2, 3%). The average overall content score was 3.6 (SD 2.0) out of 9, and the average DISCERN score was 39.2 (SD 8.1) out of 75, indicating low comprehensiveness and fair information reliability, respectively. High-quality videos (n=22, 30%) most frequently addressed overall content regarding pain duration (22/22, 100%), pain types (20/22, 91%), return-to-activity instructions (19/22, 86%), and nonpharmacologic methods for pain control (19/22, 86%). There were differences in the overall content score (P=.02) by video source but not DISCERN score (P=.45). Personal vlogs had the highest overall content score at 4.0 (SD 2.1), followed by medical videos at 3.3 (SD 2.0). Longer video duration and a greater number of comments and likes were significantly correlated with the overall content score, whereas the number of video comments was inversely correlated with the DISCERN score. Individuals seeking information from YouTube regarding postpartum cesarean pain management are likely to encounter videos that lack adequate comprehensiveness and reliability. Clinicians should counsel patients to exercise caution when using YouTube as a health information resource.

Evaluation of the educational quality of publicly available online videos on laparoscopic jejunostomy by utilizing the LAP-VEGaS guidelines.

Despite its common nature, there is no data on the educational quality of publicly available laparoscopic jejunostomy training videos. The LAParoscopic surgery Video Educational GuidelineS (LAP-VEGaS) video assessment tool, released in 2020, has been developed to ensure that teaching videos are of appropriate quality. This study applies the LAP-VEGaS tool to currently available laparoscopic jejunostomy videos. A retrospective review of YouTube® videos was conducted for "laparoscopic jejunostomy." Included videos were rated by three independent investigators using LAP-VEGaS video assessment tool (0-18). Wilcoxon rank-sum test was used to evaluate differences in LAP-VEGaS scores between video categories and date of publication relative to 2020. Spearman's correlation test was performed to measure association between scores and length, number of views and likes. 27 unique videos met selection criteria. Academic and physician video walkthroughs did not demonstrate a significant difference in median scores (9.33 IQR 6.33, 14.33 vs. 7.67 IQR 4, 12.67, p = 0.3951). Videos published after 2020 demonstrated higher median scores than those published before 2020 (13 IQR 7.5, 14.67 vs. 5 IQR 3, 9.67, p = 0.0081). A majority of videos failed to provide patient position (52%), intraoperative findings (56%), operative time (63%), graphic aids (74%), and audio/written commentary (52%). A positive association was demonstrated between scores and number of likes (rs = 0.59, p = 0.0011) and video length (rs = 0.39, p = 0.0421), but not number of views (rs = 0.17, p = 0.3991). The majority of available YouTube® videos on laparoscopic jejunostomy fail to meet the basic educational needs of surgical trainees, and there is no difference between those produced by academic centers or independent physicians. However, there has been improvement in video quality following the release of the scoring tool. Standardization of laparoscopic jejunostomy training videos with the LAP-VEGaS score can ensure that videos are of appropriate educational value with logical structure.

Auditing YouTube’s Recommendation Algorithm for Misinformation Filter Bubbles

In this article, we present results of an auditing study performed over YouTube aimed at investigating how fast a user can get into a misinformation filter bubble, but also what it takes to “burst the bubble,” i.e., revert the bubble enclosure. We employ a sock puppet audit methodology, in which pre-programmed agents (acting as YouTube users) delve into misinformation filter bubbles by watching misinformation-promoting content. Then they try to burst the bubbles and reach more balanced recommendations by watching misinformation-debunking content. We record search results, home page results, and recommendations for the watched videos. Overall, we recorded 17,405 unique videos, out of which we manually annotated 2,914 for the presence of misinformation. The labeled data was used to train a machine learning model classifying videos into three classes (promoting, debunking, neutral) with the accuracy of 0.82. We use the trained model to classify the remaining videos that would not be feasible to annotate manually. Using both the manually and automatically annotated data, we observe the misinformation bubble dynamics for a range of audited topics. Our key finding is that even though filter bubbles do not appear in some situations, when they do, it is possible to burst them by watching misinformation-debunking content (albeit it manifests differently from topic to topic). We also observe a sudden decrease of misinformation filter bubble effect when misinformation-debunking videos are watched after misinformation-promoting videos, suggesting a strong contextuality of recommendations. Finally, when comparing our results with a previous similar study, we do not observe significant improvements in the overall quantity of recommended misinformation content.

End-to-end Pre-training with Hierarchical Matching and Momentum Contrast for Text-Video Retrieval.

Lately, video-language pre-training and text-video retrieval have attracted significant attention with the explosion of multimedia data on the Internet. However, existing approaches for video-language pre-training typically limit the exploitation of the hierarchical semantic information in videos, such as frame semantic information and global video semantic information. In this work, we present an end-to-end pre-training network with Hierarchical Matching and Momentum Contrast named HMMC. The key idea is to explore the hierarchical semantic information in videos via multilevel semantic matching between videos and texts. This design is motivated by the observation that if a video semantically matches a text (can be a title, tag or caption), the frames in this video usually have semantic connections with the text and show higher similarity than frames in other videos. Hierarchical matching is mainly realized by two proxy tasks: Video-Text Matching (VTM) and Frame-Text Matching (FTM). Another proxy task: Frame Adjacency Matching (FAM) is proposed to enhance the single visual modality representations while training from scratch. Furthermore, momentum contrast framework was introduced into HMMC to form a multimodal momentum contrast framework, enabling HMMC to incorporate more negative samples for contrastive learning which contributes to the generalization of representations. We also collected a large-scale Chinese video-language dataset (over 763k unique videos) named CHVTT to explore the multilevel semantic connections between videos and texts. Experimental results on two major Text-video retrieval benchmark datasets demonstrate the advantages of our methods. We release our code at https://github.com/cheetah003/HMMC.

Guideline-Nonconformant Investigational Treatments for Erectile Dysfunction: What are Patients Learning?

ObjectiveTo evaluate the validity of YouTube content on shockwave, platelet-rich plasma, and stem cell therapies for erectile dysfunction (ED). The consumerization of men's health is particularly notable among regenerative ED treatments, and the popularity of these treatments has been amplified by social media sites such as YouTube. MethodsWe searched YouTube utilizing the following terms: “shockwave therapy-,” “platelet-rich plasma- and “stem cell therapy-“ “-for erectile dysfunction.” The first 50 resulting videos per query were screened (inclusion criteria: relevant title/content, English language, >100 views) and judged by 2 independent graders using the validated DISCERN instrument for consumer health information. Regression analysis assessed association of video characteristics with DISCERN score. ResultsSeventy-one unique videos met criteria. More content featured non-physicians (37%) or non-urologist physicians (35%) than urologists (28%). Mean DISCERN score was low at 42.8 of 80. DISCERN score significantly differed by video category and characteristics. Educational and urologist-featuring videos were associated with higher score. News reports were associated with lower score. Only 3 videos mentioned investigational/experimental status of the therapies. Only 2 cited society guidelines. ConclusionMost YouTube videos on regenerative ED therapies are unreliable and from non-urologists. The misrepresentation of proven efficacy may be encouraging patients to seek these still investigational treatments at high fiscal costs to the patient. Videos with urologists were more likely to be higher quality and guideline based. The urological and academic communities should continue to leverage the power of social media to provide evidence-backed consumer health information in this space.

Detection of left ventricular wall motion abnormalities from volume rendering of 4DCT cardiac angiograms using deep learning.

BackgroundThe presence of left ventricular (LV) wall motion abnormalities (WMA) is an independent indicator of adverse cardiovascular events in patients with cardiovascular diseases. We develop and evaluate the ability to detect cardiac wall motion abnormalities (WMA) from dynamic volume renderings (VR) of clinical 4D computed tomography (CT) angiograms using a deep learning (DL) framework.MethodsThree hundred forty-three ECG-gated cardiac 4DCT studies (age: 61 ± 15, 60.1% male) were retrospectively evaluated. Volume-rendering videos of the LV blood pool were generated from 6 different perspectives (i.e., six views corresponding to every 60-degree rotation around the LV long axis); resulting in 2058 unique videos. Ground-truth WMA classification for each video was performed by evaluating the extent of impaired regional shortening visible (measured in the original 4DCT data). DL classification of each video for the presence of WMA was performed by first extracting image features frame-by-frame using a pre-trained Inception network and then evaluating the set of features using a long short-term memory network. Data were split into 60% for 5-fold cross-validation and 40% for testing.ResultsVolume rendering videos represent ~800-fold data compression of the 4DCT volumes. Per-video DL classification performance was high for both cross-validation (accuracy = 93.1%, sensitivity = 90.0% and specificity = 95.1%, κ: 0.86) and testing (90.9, 90.2, and 91.4% respectively, κ: 0.81). Per-study performance was also high (cross-validation: 93.7, 93.5, 93.8%, κ: 0.87; testing: 93.5, 91.9, 94.7%, κ: 0.87). By re-binning per-video results into the 6 regional views of the LV we showed DL was accurate (mean accuracy = 93.1 and 90.9% for cross-validation and testing cohort, respectively) for every region. DL classification strongly agreed (accuracy = 91.0%, κ: 0.81) with expert visual assessment.ConclusionsDynamic volume rendering of the LV blood pool combined with DL classification can accurately detect regional WMA from cardiac CT.

National Institutes of Health Stroke Scale scores obtained using a mobile application compared to the conventional paper form: a randomised controlled validation study

BackgroundPrehospital delay contributes to treatment delay in acute stroke. Numerous prehospital stroke scales exist for stroke identification, but they lack the diagnostic accuracy of the in-hospital National Institutes of Health...

Should Emergency Physicians and Nurses Direct Their Patients to YouTube for Heparin Self-Injection Training? A Systematic Review of Social Media Videos

BackgroundThe aim of this study was to examine the content, reliability, popularity, and quality of YouTube videos for patients learning how to self-administer subcutaneous low molecular weight heparin injections. MethodsA systematic review of YouTube videos was conducted on August 20, 2021, using the keywords of “Low-molecular-weight heparin injection,” “Enoxaparin injection,” “Heparin injection,” “Dalteparin injection,” and “Tinzaparin injection.” Two independent emergency physicians evaluated included videos separately with 5 different score systems (1- Journal of American Medical Association Score, 2-The Video Power Index, 3- Global Quality Scale, 4- Modified 5 Point DISCERN, 5- Total Comprehensiveness Score). ResultsOf 458 videos, a total of 161 unique videos were included. Of these, 94 (58.4%) were classified as useful and 67 (41.6%) as containing misleading information. The total number of views was 6,245,284 in useful information videos. DISCERN score (median 4, P < .001), Global Quality Score (median 4, P < .001), Journal of American Medical Association Score (median 4, P < .001), and Total Comprehensiveness Score (median 6, P < .001) were higher in the Useful Information Group. ConclusionsNurse and physician prescreening and prescoring the accuracy and quality of specific low molecular weight heparin injection self-administration videos before recommending YouTube to patients is warranted. Policies to limit the spread of health misinformation through credibility scoring and evaluation are needed on social media sites such as YouTube.