Étiquette : health (Page 1 of 3)

Predicting sex from retinal fundus photographs using automated deep learning

Figure 2

“Deep learning may transform health care, but model development has largely been dependent on availability of advanced technical expertise. Herein we present the development of a deep learning model by clinicians without coding, which predicts reported sex from retinal fundus photographs. A model was trained on 84,743 retinal fundus photos from the UK Biobank dataset. External validation was performed on 252 fundus photos from a tertiary ophthalmic referral center.”

Source : Predicting sex from retinal fundus photographs using automated deep learning | Scientific Reports

YouTube Bans Anti-Vaccine Misinformation

YouTube said it was banning the accounts of several prominent anti-vaccine activists from its platform, including Robert F. Kennedy Jr.’s.

“In a blog post, YouTube said it would remove videos claiming that vaccines do not reduce rates of transmission or contraction of disease, and content that includes misinformation on the makeup of the vaccines. Claims that approved vaccines cause autism, cancer or infertility, or that the vaccines contain trackers, will also be removed.
Prominent anti-vaccine activists have long been able to build huge audiences online, helped along by the algorithmic powers of social networks that prioritize videos and posts that are particularly successful at capturing people’s attention. A nonprofit, the Center for Countering Digital Hate, published research this year showing that a group of 12 people were responsible for sharing 65 percent of all anti-vaccine messaging on social media, calling the group the “Disinformation Dozen.” In July, the White House cited the research as it criticized tech companies for allowing misinformation about the coronavirus and vaccines to spread widely, sparking a tense back-and-forth between the administration and Facebook.”

Source : YouTube Bans Anti-Vaccine Misinformation – The New York Times

Using AI to help find answers to common skin conditions

“To make sure we’re building for everyone, our model accounts for factors like age, sex, race and skin types — from pale skin that does not tan to brown skin that rarely burns. We developed and fine-tuned our model with de-identified data encompassing around 65,000 images and case data of diagnosed skin conditions, millions of curated skin concern images and thousands of examples of healthy skin — all across different demographics.  Recently, the AI model that powers our tool successfully passed clinical validation, and the tool has been CE marked as a Class I medical device in the EU.”

Source : Using AI to help find answers to common skin conditions


“The policies are significantly stronger than Facebook’s typical rules against misinformation, under which false claims are marked as such and suppressed by the curation algorithms, but are not removed from Facebook and Instagram entirely. The distinction, the company said, was because misinformation about Covid and Covid vaccinations “could lead to imminent physical harm”.”
The topic has a personal connection for the Facebook founder and chief executive, Mark Zuckerberg. His philanthropic organisation, the Chan Zuckerberg Initiative, has embarked on a flagship effort to “cure all disease”, with a number of avenues of research being pursued, including a specific focus on vaccinations, through the Chan Zuckerberg Biohub.

Source : Facebook to remove false claims about Covid vaccines | Technology | The Guardian

StopCovid ou encore ? – Cédric O

Le choix, selon Cédric O !

“Le choix est donc très simple : tant que l’immunité collective n’est pas atteinte (ce qui est un horizon lointain), l’alternative se résume ainsi :
1. Tout faire pour couper les « départs de feu » le plus rapidement possible, y compris en utilisant des outils numériques comme StopCovid, dans des conditions très encadrées et proportionnées (et dans un contexte où l’ensemble des pays européens prévoient de déployer de tels outils) ;
2. Refuser ces outils pour des raisons philosophiques, mais dans ce cas accepter un risque significatif de malades et de morts supplémentaires.”

Source : StopCovid ou encore ? – Cédric O – Medium


“Les mesures de surveillance, via nos usages des technologies, que suggèrent nos gouvernants relèvent en réalité d’une stratégie pour détourner notre attention de la cause réelle du problème que constitue l’abandon de l’hôpital public. Ils tablent sur la culpabilisation des citoyens désireux d’agir pour faire adopter des outils toujours plus intrusifs et évitent soigneusement de mettre en lumière les multiples réseaux de solidarité qui se forment, les besoins criants des associations pour aider les plus précaires, les multiples critiques de notre mode de vie qui émergent même des plus libéraux. Plutôt que d’assumer les conséquences désastreuses d’une politique de santé défaillante, leur diversion consiste à inverser les rôles, à nous faire passer, nous, pour ceux qui refuseront d’aider les autres. Comme si nous devions être coupable de vouloir protéger notre vie privée, d’exprimer notre colère, ou simplement de suggérer des alternatives. ”

Source : Urgence partout, État nul part – La Quadrature du Net

Genomic epidemiology of novel coronavirus

“This phylogeny shows evolutionary relationships of hCoV-19 (or SARS-CoV-2) viruses from the ongoing novel coronavirus COVID-19 pandemic. This phylogeny shows an initial emergence in Wuhan, China, in Nov-Dec 2019 followed by sustained human-to-human transmission leading to sampled infections. Although the genetic relationships among sampled viruses are quite clear, there is considerable uncertainty surrounding estimates of transmission dates and in reconstruction of geographic spread. Please be aware that specific inferred transmission patterns are only a hypothesis.”

Source : Nextstrain / ncov

“We are not epidemiologists—we are the design, engineering, and support teams at a mapping company. Just as we don’t spend every moment thinking about how to track and slow the spread of infectious disease, most public health teams we work with are not experts in spatial data visualization. In the environment of a growing epidemic, maps have a way of spreading fast, too — making it imperative that they are accurate, informative, and thoughtfully designed. To answer common questions and help our partners make thoughtful design decisions while mapping this and other health crises, we have put together a number of best practices and common pitfalls to avoid.”

Source : 7 best practices for mapping a pandemic – Points of interest

« Older posts

© 2022 no-Flux

Theme by Anders NorenUp ↑