AI Shows Potential In Early Autism Detection - 1

Image by MahmudAl, from Pixabay

AI Shows Potential In Early Autism Detection

  • Written by Kiara Fabbri Former Tech News Writer
  • Fact-Checked by Justyn Newman Former Lead Cybersecurity Editor

A research paper published yesterday presented some promising results of a machine learning model designed to identify children at risk of autism spectrum disorder (ASD) at an early age. The model, named AutMedAI, achieved an accuracy rate of 80%, offering hope for early detection.

Developed by researchers at Karolinska Institutet, AutMedAI analyzed data from approximately 30,000 individuals to identify patterns linked to autism. The data was based on 28 parameters that can be easily obtained before a child turns two, such as the age of the first smile, the first short sentence, and the presence of eating difficulties.

In a statement , the study author Shyam Rajagopalan emphasized the significance of these findings: “The results of the study are significant because they show that it is possible to identify individuals who are likely to have autism from relatively limited and readily available information.”

The researchers highlight the potential of this study to screen children at an early age, which could lead to the implementation of timely interventions, helping children with autism develop optimally.

However, the researchers caution that while the findings are promising, the model is not a substitute for comprehensive clinical evaluation. Further research and validation are needed to fully assess the model’s potential for clinical use.

It’s important to note that AI tools can sometimes lead to misdiagnosis with potentially harmful consequences . A recent study found that AI struggled to accurately diagnose pediatric cases, with incorrect diagnoses in 83% of the cases it analyzed.

“Generative AI technologies have the potential to improve health care, but only if those who develop, regulate and use these technologies identify and fully account for the associated risks,” said Jeremy Farrar, the WHO’s chief scientist, as reported by Nature .

Build Your Own AI-Powered Robot: Startup Launches Affordable Kits for Beginners - 2

Photo by Nik on Unsplash

Build Your Own AI-Powered Robot: Startup Launches Affordable Kits for Beginners

  • Written by Andrea Miliani Former Tech News Expert
  • Fact-Checked by

LeRobot, the unicorn startup Hugging Face’s new open-source robotics project, just launched a new tutorial to help regular people build their own AI-powered robots. Remi Cadene, a former Tesla scientist now leading the initiative, announced the update, and all instructions and tutorials are available on GitHub.

“The wait is finally over! We just dropped an in-depth tutorial on how to build your own robot!” wrote Cadene on X, “Teach it new skills by showing it a few moves with just a laptop. Then watch your homemade robot act autonomously.”

The wait is finally over!!! 😁 We just dropped an in-depth tutorial on how to build your own robot! Teach it new skills by showing it a few moves with just a laptop. Then watch your homemade robot act autonomously 🤯 1/🧵👇 pic.twitter.com/ReeDvNlrg9 — Remi Cadene (@RemiCadene) August 19, 2024

Hugging Face, an AI and machine learning open-source platform originally founded by French entrepreneurs and now headquartered in the United States announced the new project LeRobot in May. According to Venture Beat , its mission is to “democratize AI robotics and inspire a new generation of roboticists.”

Now, LeRobot has launched its first tutorial, Getting Started with Real-World Robots, available for everyone through GitHub. The new instructions have been designed to teach users to order and build their own robot, connect and configure it, record and visualize the database, prepare policy for evaluation, and visualize results.

The kit with the robot parts—including instructions to print 3D parts— and the links to buy each piece are available on GitHub for customers in the United States, the European Union, and the United Kingdom. The total price for all pieces is $278 or 360€, but Cadene said on X that they are already working on a less expensive version, Moss v1, that won’t require 3D-printed parts and shouldn’t surpass $150.

Users on social media have already expressed excitement. “Finally, a robotics tutorial that’s accessible to everyone! Can’t wait to try it out and see what kind of skills I can teach my robot. Thanks for making robotics more inclusive!” wrote one user on X .

Amateur robot builders could soon try to compete against other mechanisms like Google’s new ping-pong robot or at least build a similar structure.