AI-Powered fMRI Dream Machine can ‘read your dreams’
Reviewed By: Kevin Pocock
Table of Contents
Would you let a machine see your dreams? That may soon be a possibility, with a study involving an AI-assisted MRI machine reporting 60% accurate visual recreations of a consenting patients dreams. The study, performed by Dr. Kamitani Yukiyasu, resulted in high-quality video of human dreams, thanks to what they’re calling the “Dream Machine“.
Can an MRI machine record dreams?
A recent study in Japan, lead by by Dr. Kamitani Yukiyasu, saw up to 60% accurate visual reconstruction across 200 willing patients. The fMRI (functional magnetic resonance imaging) machine, dubbed the “Dream Machine” takes brain scans and reconstructs dreams using AI image model Stable Diffusion. Astoundingly, the result is a high-quality video of the participants dream.
In short, yes. Brain signals have been readable and recordable since MRI machines were invented (or at least first patented in the US) by American-Armenian Dr. Raymond Damadian in the 1970’s. With the additional layer of AI, acting as a general classifier and generative AI image (or video) production mechanism, deep image reconstruction is possible today.
Essential AI Tools
Razer DeathStalker V2 Pro Wireless Gaming Keyboard (Renewed)
Samsung Galaxy Watch 4 40mm Smart Watch Bluetooth – Pink Gold (Renewed)
ASUS ROG Rapture WiFi 6E Gaming Router (GT-AXE16000)
Amazon Fire HD 10 tablet
AMD Ryzen 5 5600X 6-core, 12-Thread Unlocked Desktop Processor with Wraith Stealth Cooler
This dream-recording MRI technology has in fact been replicated by multiple scientists across the globe. Dream researcher Daniel Oldis, working at the Cognitive Neuroscience Lab at the University of Texas has long been on this path. Along the way, learning how to “decode brain activity, nerve impulses, and even the intricate details of speech and movement within dreams.”
“We’re going into the dream space,” claims Oldis, comparing the prospect of brain recordings to the early years of the space race.
Deals season is here folks, and Amazon has already kickstarted its early Black Friday deals! We'll be covering all the best deals in more details over in our deals hub, but if you haven't got time to read through those, why not see our top picks below.
- ASUS TUF NVIDIA RTX 5080 Was $1599 Now $1199
- ASUS TUF RTX 5070 Ti Was $999 Now $849
- Samsung Odyssey OLED G6 Was $899 Now $649
- TCL 43S250R Roku TV 2023 Was $279 Now $199
- iBUYPOWER Y40 Gaming PC Was $2,299 Now $1,819
- Samsung Odyssey G9 (G95C) Was $1,299 Now $777
- Alienware Area-51 gaming laptop Was $3,499 Now $2,799
- Samsung 77-inch OLED S95F Was $4,297 Now $3,497
- ASUS ROG Strix G16 Was $1,499 Now $1,199
*Prices and savings subject to change. Click through to get the current prices.
Researchers at the National University of Singapore, and the Chinese University of Hong Kong reported similar findings in May of this year.
Artificial intelligence is proving effective in enhancing the ability of MRI and EGM (electromyogram) machines to study imagery, brain activity and muscle movement. Shinji Nishimoto, a neuroscientist at Osaka University, muses that these processes may even be used to capture thoughts, as well as dreams.
Is there a machine that records dreams?
If there is, it’s an MRI machine. Ultimately, viewing or recording a person’s dreams requires an interface with the brain — After all, that’s where dreams ‘happen’. There’s no better interface on the planet than a device custom-designed to record human brain activity.
There are limitations, however. The process of fMRI itself pays attention to blood oxygenation level-dependent (BOLD) signals, capturing images of brain activity once every few seconds. The result would be a very low frame rate video — essentially, not particularly high-quality. The goal would be 30, not 0.3, frames per second (FPS).
Cracking this limitation of neuroscience is no easy feat (of course). Thankfully, research leader Jiaxin Qing, along with Zijiao Chen and Juan Helen Zhou have also performed similar research using Stable Diffusion and their own Mind-Video model. Self-described as “a two-module pipeline designed to bridge the gap between image and video brain decoding,” its fMRI decoder can learn progressively, fine tuning an image database that will in time aid our understanding of the human brain.
Jiaxin Qing and colleagues were able to create high-quality videos with non-invasive tools. The resulting 85% accuracy rate also included motion and dynamic scenes, not merely still images.
Can artificial intelligence read minds?
Which AI visualizes dreams?
There is no AI that, by itself, can read your mind. I would boldly claim that AI, no matter how advanced, will always need a physical interface with your brain to do so. Could Elon Musks’ Neuralink enable this on a mass scale? There’s potential there, of course. It will absolutely be a consideration as society progresses further into technomorphism.
About the Author