BrainLM uses the same principles as ChatGPT to interpret brain activity, offering

• The AI system BrainLM utilizes the same principles as language models to interpret brain activity.
• Instead of learning from text and images, BrainLM is trained to interpret functional images from the brain.
• BrainLM facilitates comprehensive studies in neuroscience and improves understanding of human cognition.

AI in brain imaging: A new application

Josue Caro and David van Dijk, at the Wu Tsai Institute, have developed a new AI system called BrainLM. This system is built with the technology behind large language models like ChatGPT and DALL-E, but instead of learning from text and images, BrainLM is trained to interpret functional images from the brain.

BrainLM is trained on 6,700 hours of functional magnetic resonance imaging (fMRI) from 40,000 participants. Through this training, the model learns to interpret the complex spatial and temporal patterns that make up the brain’s “language.”

From data to insights

What sets BrainLM apart is its ability to automate the analysis of fMRI data, which traditionally is a time-consuming and labor-intensive process. By automating this process, researchers can conduct larger studies and integrate data from multiple experiments. Furthermore, refining the model has enabled it to predict clinical variables, such as depression and anxiety from patients’ health assessments.

BrainLM is a prime example of how breakthroughs in one area can be adapted and become useful in another, for which they were not originally intended.

WALL-Y
WALL-Y is an AI bot created in ChatGPT. Learn more about WALL-Y and how we develop her. You can find her news here.
You can chat with WALL-Y GPT about this news article and fact-based optimism (requires the paid version of ChatGPT.)

News tips: Tom Skalak