Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog

Category: robotics/AI – Page 290

How Journey Foods is leveraging AI to streamline the CPG industry

As a simple illustration, let’s say someone wanted to create a tomato sauce recipe, optimizing vitamin C and using sustainable tomatoes within a certain cost range. Journey Foods then taps into its database to generate an optimal recipe, and will continually push recommendations of top suppliers.

“Essentially, when people go to ChatGPT or something, and they’re asking them, ‘write this paper for me, or give me a social media post, speak to this audience,’ or whatever, right? It’s the same thing with our generative recipe recommendations,” Lynn said.

Except Lynn doesn’t use ChatGPT. Systems such as ChaptGPT gather data from the open internet, but Journey Foods gets its data from research institutions, academic journals, suppliers and manufacturers. Lynn said her business uses a lot of private, hard data that’s unstructured, with her company then giving it structure and doing so globally.

Mayo Clinic researchers develop new AI tools to reveal seizure hotspots, improve patient care

Mayo Clinic researchers have developed new artificial intelligence (AI)-based tools to pinpoint specific regions of the brain with seizure hotspots more quickly and accurately in patients with drug-resistant epilepsy. Their study, published in Nature Communications Medicine, highlights the potential of AI to revolutionize epilepsy treatment by interpreting brain waves during electrode implantation surgery. This transformative approach could significantly reduce the time patients spend in the hospital, accelerating the identification and removal of seizure-generating brain regions.

“This innovative approach could enable more rapid and accurate identification of seizure-generating areas during stereo-electroencephalography (EEG) implantation surgery, potentially reducing the cost and risks of prolonged monitoring,” says Nuri Ince, Ph.D., senior author of the study and a consultant in the Mayo Clinic Department of Neurologic Surgery.

Drug-resistant epilepsy often requires surgical removal of the seizure-causing brain tissue. A first step in that treatment is typically a surgery that involves implanting electrodes in the brain and monitoring neural activity for several days or weeks to identify the location of the seizures.

Algorithms based on deep learning can improve medical image analysis

Artificial intelligence has the potential to improve the analysis of medical image data. For example, algorithms based on deep learning can determine the location and size of tumors. This is the result of AutoPET, an international competition in medical image analysis, where researchers of Karlsruhe Institute of Technology (KIT) were ranked fifth.

The seven best autoPET teams report in the journal Nature Machine Intelligence on how algorithms can detect lesions in (PET) and computed tomography (CT).

Imaging techniques play a key role in the diagnosis of cancer. Precisely determining the location, size, and type of tumor is essential for choosing the right therapy. The most important imaging techniques include positron emission tomography (PET) and computer tomography (CT).

An AI Chemist Made A Catalyst to Make Oxygen On Mars Using Local Materials

Breaking oxygen out of a water molecule is a relatively simple process, at least chemically. Even so, it does require components, one of the most important of which is a catalyst. Catalysts enable reactions and are linearly scalable, so if you want more reactions quickly, you need a bigger catalyst. In space exploration, bigger means heavier, which translates into more expensive. So, when humanity is looking for a catalyst to split water into oxygen and hydrogen on Mars, creating one from local Martian materials would be worthwhile. That is precisely what a team from Hefei, China, did by using what they called an “AI Chemist.”

Unfortunately, the name “AIChemist” didn’t stick, though that joke might vary depending on the font you read it in. Whatever its name, the team’s work was some serious science. It specifically applied machine learning algorithms that have become all the rage lately to selecting an effective catalyst for an “oxygen evolution reaction” by utilizing materials native to Mars.

To say it only chose the catalyst isn’t giving the system the full credit it’s due, though. It accomplished a series of steps, including developing a catalyst formula, pretreating the ore to create the catalyst, synthesizing it, and testing it once it was complete. The authors estimate that the automated process saved over 2,000 years of human labor by completing all of these tasks and point to the exceptional results of the testing to prove it.

How to Pop a Microscopic Cork

Researchers used machine learning to optimize the process by which a tiny cage is opened to release a molecule.

Researchers have designed a tiny structure that could help deliver drugs inside the body [1]. The theoretical and computational work required machine learning to optimize the parameters for the structure, which could stick to a closed shell containing a small molecule and cause the shell to open. The results demonstrate the potential for machine learning to assist in the development of artificial systems that can perform complex biomolecular processes.

Researchers are developing artificial molecular-scale structures that could perform functions such as drug delivery or gene editing. Creating such artificial systems, however, usually entails a frustrating tradeoff. If the components are simple enough to be computationally tractable, they are unlikely to yield complex interactions. But if the components are too complex, they become harder to combine and coordinate. Machine learning can reduce the computational cost of designing useful artificial systems, according to graduate student Ryan Krueger of Harvard University.

An efficient small language model that could perform better on smartphones

There are over 30,000 weather stations in the world, measuring temperature, precipitation and other indicators often on a daily basis. That’s a massive amount of data for climate researchers to compile and analyze to produce the monthly and annual global and regional temperatures (especially) that make the news.

Now researchers have unleashed artificial intelligence (AI) on these datasets to analyze in Europe, finding excellent agreement compared to existing results that used traditional methods, and as well have uncovered climate extremes not previously known. Their work has been published in Nature Communications.

With the world’s climate changing rapidly, it is important to know how temperature and precipitation extremes are changing, so planners can adapt to the extremes here now and to what’s coming.