English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Cameras are a common in public spaces. London England is estimated to have 500,000 and Beijing China over 800,000. Smartphone penetration exceeds 60% of the population in 20 countries worldwide. Hundreds of sudden cardiac arrests are captured on video annually. This study searches publically available cardiac arrest videos with two objectives i) describe sudden cardiac arrest behaviour and ii) explore potential opportunities for machine learning.  Methods:  The search terms: “sudden death”, “heart attack”, “cardiac arrest” and “public death” were used. English sources included: Youtube.com, Dailymotion, vimeo.com, vidamax.com, LiveLeak.com and documentingreality.com Whereas, iqiyi.com, youku.com, le.com, fun.tv, pptv.com and tudou.com were searched using simplified Chinese. Inclusion criteria required that the subject in the video be completely visible five seconds prior to the event and at least ten seconds after and the quality of the video be adequate to visualize movement. Exclusion criteria included trauma or precipitating event (substance misuse, toxic exposure or asphyxiation). Each video source was searched until 30 consecutive irrelevant videos were obtained.  Results:  Four hundred and eighty eight videos met inclusion criteria. Of those videos, 112 could be confirmed as a “cardiac arrest” by at least two sources (news, or family social media account). In 53 (47%) of these videos the person touches their face or head within five seconds of collapse. Of the 98 videos where the person is upright, in 41 (37%) instances they hip-flexes and with their hands on their upper legs prior to collapse. This pattern of behaviour is combined in 36 (32%) instances. After collapse, 68 (61%) appeared to exhibit extension posturing activity. Agonal breathing was visible in 39 instances (35%).  Conclusion:  Sudden out of hospital cardiac arrest has a recognizable pattern. This represents an opportunity for machine learning, using shape tracking and edge detection, to recognize this event and activate the emergency response system.

P031: An online video analysis study of out of hospital cardiac arrest: patterns in presentation and opportunities for machine learning