Your heart and your brain give off magnetic fields about a billionth the strength of the Earth's. Reading them today needs a room full of liquid helium. A team in Tokyo just did it with a diamond, at room temperature, two millimetres from the skin.
There is a machine that can watch your brain think without touching you. It is called a MEG scanner, and it is not really a machine. It is a room.
Inside sits a helmet of detectors cooled with liquid helium to a few degrees above absolute zero, surrounded by heavy magnetic shielding to keep the rest of the world out. It costs millions. There are only a few hundred of them on Earth. The reason for all that expense is simple: when your neurons fire, the magnetic field they produce is roughly a billion times weaker than the planet's own. Almost nothing can hear a whisper that faint, and the things that can need to be very cold.
Diamond was supposed to fix this. It has not, for one stubborn and slightly absurd reason. The laser you need to read a diamond sensor gets hot enough to burn skin.
Here is what happened
A sensor that runs cool enough to touch you. On 21 August, researchers at the Institute of Science Tokyo described a diamond quantum magnetometer that operates on a laser of just 210 milliwatts, roughly a tenth of the power such sensors normally need. The paper was published in Applied Physics Letters on 22 July and selected as a Featured Article.
The heat problem, solved by a number that matters. Temperature rise in the sensor was held to 13 kelvin, keeping the diamond at about 38 degrees Celsius. Human tissue begins to register pain at about 42. This is a quantum device engineered around the threshold at which skin hurts.
Close enough to hear. A compact printed-circuit-board antenna let the whole assembly sit 2 millimetres from the target. It reached a sensitivity of roughly 3 picotesla per root hertz across the 100 to 400 hertz band.
It read a brain signal. Tested against a dry phantom that mimics the magnetic signals of a brain, the sensor detected a field of 77.7 picotesla at a distance of about 2.5 millimetres.
An industrial name on an academic paper. The author list runs across Science Tokyo, the National Institute for Materials Science, the National Institutes for Quantum Science and Technology, and DENSO, one of the world's largest automotive component makers.
How it works
The sensor is a deliberate flaw. Take a diamond and replace one carbon atom with a nitrogen atom, leaving the neighbouring lattice site empty. That defect is called a nitrogen-vacancy centre, and it behaves like an exquisitely small compass needle. Put it in a magnetic field and its quantum state shifts in a way you can measure with light. Unlike almost every other quantum sensor, it does this at room temperature.
Reading it takes a laser, and that was the problem. You shine light in to set the state and to read it back out. The fainter the field you want to detect, the more light you need, and conventional biomagnetic work uses watt-level lasers. That much light in a small crystal makes heat, and heat means you have to back the sensor away from anything living.
Distance is the whole game. Biomagnetic fields fall away sharply with distance. Every millimetre you retreat costs you signal, which is why a sensor that must sit far from the chest is a sensor that cannot hear the heart properly.
The fix was optical, not quantum. The team built a light-trapping diamond waveguide that bounces the laser through the crystal by total internal reflection, so the same beam passes many more sensing points instead of passing through once. They collected about 20 milliwatts of fluorescence from 210 milliwatts of input, a conversion efficiency of 9.5 percent, which is what allows a weak laser to do the work of a strong one.
The measurement itself is a stopwatch. The technique, Ramsey interferometry, sends two microwave pulses and lets the quantum state evolve in the gap between them. How far it drifts in that gap tells you the strength of the magnetic field it was sitting in.
Why it matters
It attacks access, not accuracy. Magnetic imaging of the heart and brain is among the most information-rich diagnostics we have and among the least used, purely because the sensors need cryogenics. A room-temperature sensor that can touch the skin changes who can own one.
Magnetocardiography is the nearer prize. Reading the heart's magnetic field can reveal electrical problems that a standard ECG misses, without any wires on the chest. The heart's signal is stronger than the brain's, so it is the first realistic target.
Watch the company, not just the physics. DENSO is a mass manufacturer of precision components. An automotive supplier co-authoring a biomedical quantum paper is what industrialisation intent looks like before anybody announces a product.
It is quantum technology with a patient in it. Most quantum news is about qubit counts and error rates, which are real but distant. This is the same underlying physics arriving somewhere a person can be helped by it, and it is worth training yourself to notice the difference.
The honest catch
This is a component result, not a scanner, and the gap between the two is measured in years.
A phantom is not a person. The 77.7 picotesla measurement came from a dry model built to imitate brain signals. No living human has been measured with this device.
SQUIDs are still more sensitive. The cryogenic detectors in today's MEG rooms remain the better instrument in raw terms. The argument for diamond is not that it hears better, it is that it can get closer and does not need helium.
One sensor is not an array. A clinical scanner needs many channels working together. Nothing here shows that sensitivity survives being multiplied.
13 kelvin is still a rise. The margin to the pain threshold is real but it is a margin, not an absence of heat.
EDITOR'S TAKE
The interesting thing about this paper is that the breakthrough is not quantum at all. The nitrogen-vacancy centre has worked for years. What changed is an optics trick, bouncing light around inside a crystal so a weak laser does the job of a strong one, and a circuit board small enough to get out of the way. That pattern repeats across the frontier more often than anyone admits: the physics arrives first and then sits still for a decade while somebody solves a heat problem or a packaging problem. Watch for two things. A measurement on a living chest, which is the next honest milestone. And DENSO, because when a components manufacturer with automotive volumes puts its name on a medical sensor paper, somebody has already run the numbers on making a lot of them.
Quick questions
What is a nitrogen-vacancy centre, in plain terms?
It is a flaw in a diamond, put there on purpose. One carbon atom in the crystal is swapped for a nitrogen atom, and the lattice position next to it is left empty. That pairing traps electrons whose quantum state responds to any magnetic field around them, and the state can be set and read using light and microwaves. The useful part is that it does all this at room temperature, unlike most quantum sensors, which need to be cooled to within a fraction of a degree of absolute zero. The diamond is not exotic or enormous. The sensing happens in a small crystal you could hold between two fingers.
Why does two millimetres matter so much?
Because magnetic fields from the body weaken very quickly with distance. The signal from a beating heart or a firing cluster of neurons is already around a billionth of the Earth's field at the skin, and it drops away sharply from there. A sensor that has to sit centimetres back because it would otherwise burn is a sensor that has thrown away most of what it came to measure. Getting to two millimetres is not a refinement, it is the difference between hearing something and not. That is why the heat number and the distance number in this paper are really the same result stated twice.
Does this replace MRI?
No, and they answer different questions. An MRI scanner shows you structure, meaning what tissue is where, using a very strong magnetic field applied from outside. Magnetocardiography and magnetoencephalography are passive: they listen to the faint magnetic fields the body generates by itself, which reflect electrical activity rather than anatomy. That makes them better suited to questions about function and timing, such as how an electrical signal travels through heart muscle. If this technology matures, it adds a cheap, cryogen-free way to watch the body's electrics. It does not take a picture of what is inside you.
Sources
Araki et al., Applied Physics Letters: A Highly Sensitive Diamond NV Magnetometer Using Ramsey Interferometry with a Short Sensor-to-Sample Distance, published 22 July 2026 and selected as a Featured Article.
Institute of Science Tokyo: the 21 August 2026 announcement, with the laser power, temperature rise, sensor distance and phantom measurement figures.
Scilight: the American Institute of Physics summary of the result.
Frontier Signal, 13 July 2026: our explainer on MEG and reading the brain without an implant.
Frontier Signal, 7 August 2026: the same class of quantum magnetometer, pointed at the planet instead of the body.
Frontier Signal explains frontier technology in plain English. Company and laboratory figures should be independently verified. This is general information, not medical advice.
