Skip to main content
https://www.highperformancecpmgate.com/rgeesizw1?key=a9d7b2ab045c91688419e8e18a006621

MIT AI tool can predict breast cancer up to 5 years early, works equally well for white and black patients

MIT’s Computer Science and Artificial Intelligence Lab has developed a new deep learning-based AI prediction model that can anticipate the development of breast cancer up to five years in advance. Researchers working on the product also recognized that other similar projects have often had inherent bias because they were based overwhelmingly on white patient populations, and specifically designed their own model so that it is informed by “more equitable” data that ensures it’s “equally accurate for white and black women.”

That’s key, MIT notes in a blog post, because black women are more than 42 percent more likely than white women to die from breast cancer, and one contributing factor could be that they aren’t as well-served by current early detection techniques. MIT says that its work in developing this technique was aimed specifically at making the assessment of health risks of this nature more accurate for minorities, who are often not well represented in development of deep learning models. The issue of algorithmic bias is a focus of a lot of industry research and even newer products forthcoming from technology companies working on deploying AI in the field.

This MIT tool, which is trained on mammograms and patient outcomes (eventual development of cancer being the key one) from over 60,000 patients (with over 90,000 mammograms total) from the Massachusetts General Hospital, starts from the data and uses deep learning to identify patters that would not be apparent or even observable by human clinicians. Because it’s not based on existing assumptions or received knowledge about risk factors, which are at best a suggestive framework, the results have so far shown to be far more accurate, especially at predictive, pre-diagnosis discovery.

Overall, the project is intended to help healthcare professionals put together the right screening program for individuals in their care and eliminate the heartbreaking and all-too common outcome of late diagnosis. MIT hopes the technique can also be used to improve detection of other diseases that have similar problems with existing risk models with far too many gaps and lower degrees of accuracy.

Comments

Post a Comment

Popular posts from this blog

Uber co-founder Garrett Camp steps back from board director role

Uber co-founder Garrett Camp is relinquishing his role as a board director and switching to board observer — where he says he’ll focus on product strategy for the ride hailing giant. Camp made the announcement in a short Medium post in which he writes of his decade at Uber: “I’ve learned a lot, and realized that I’m most helpful when focused on product strategy & design, and this is where I’d like to focus going forward.” “I will continue to work with Dara [Khosrowshahi, Uber CEO] and the product and technology leadership teams to brainstorm new ideas, iterate on plans and designs, and continue to innovate at scale,” he adds. “We have a strong and diverse team in place, and I’m confident everyone will navigate well during these turbulent times.” The Canadian billionaire entrepreneur signs off by saying he’s looking forward to helping Uber “brainstorm the next big idea”. Camp hasn’t been short of ideas over his career in tech. He’s the co-founder of the web 2.0 recommendatio
Post a Comment
Read more

Drone crash near kids leads Swiss Post and Matternet to suspend autonomous deliveries

A serious crash by a delivery drone in Switzerland have grounded the fleet and put a partnership on ice. Within a stone’s throw of a school, the incident raised grim possibilities for the possibilities of catastrophic failure of payload-bearing autonomous aerial vehicles. The drones were operated by Matternet as part of a partnership with the Swiss Post (i.e. the postal service), which was using the craft to dispatch lab samples from one medical center for priority cases. As far as potential applications of drone delivery, it’s a home run — but twice now the craft have crashed, first with a soft landing and the second time a very hard one. The first incident, in January, was the result of a GPS hardware error; the drone entered a planned failback state and deployed its emergency parachute, falling slowly to the ground. Measures were taken to improve the GPS systems. The second failure in May, however, led to the drone attempting to deploy its parachute again, only to sever the line
Post a Comment
Read more

ProtonMail logged IP address of French activist after order by Swiss authorities

ProtonMail , a hosted email service with a focus on end-to-end encrypted communications, has been facing criticism after a police report showed that French authorities managed to obtain the IP address of a French activist who was using the online service. The company has communicated widely about the incident, stating that it doesn’t log IP addresses by default and it only complies with local regulation — in that case Swiss law. While ProtonMail didn’t cooperate with French authorities, French police sent a request to Swiss police via Europol to force the company to obtain the IP address of one of its users. For the past year, a group of people have taken over a handful of commercial premises and apartments near Place Sainte Marthe in Paris. They want to fight against gentrification, real estate speculation, Airbnb and high-end restaurants. While it started as a local conflict, it quickly became a symbolic campaign. They attracted newspaper headlines when they started occupying prem
Post a Comment
Read more