Quantum Entanglement Has Been Photographed Inside a Living Human Body for the First Time
A 48-year-old man with a neuroendocrine tumor walked into a PET scan in Kraków, Poland, and left with what sounds like science fiction. Not because of what the scan found in his pancreas, but because for the first time, his doctors could see quantum entanglement unfolding inside his living body.
Background & Context
Every time a positron meets an electron inside your body (and this happens constantly when patients are injected with standard PET tracers), the two particles annihilate, producing two high-energy photons that fly off in opposite directions. Quantum theory says those two photons should be entangled: measure the polarization of one, and the other’s polarization is instantly determined, even meters apart. Einstein called this “spooky action at a distance.” For decades it was something you demonstrated in optics labs with lasers and crystals, not something you looked for inside a patient’s liver.
It turns out that whether those annihilation photons stay entangled depends on where they are born. When a positron forms a short-lived atom called positronium before annihilating, the resulting photons are predicted to be maximally entangled. But when a positron smacks directly into a nearby molecule’s electron (the so-called pick-off process), the photons come out separable, with no quantum link at all. The ratio between these two pathways changes with the molecular environment of the tissue. That, as a team at Jagiellonian University realized, might be medically useful.
The J-PET collaboration had already shown they could image positronium lifetimes in the human brain in 2024, published in Science Advances. The next logical step: could they image entanglement itself?
What the Researchers Did
Standard PET scanners use crystal detectors that absorb gamma rays entirely. This makes them blind to the polarization information carried by annihilation photons. The J-PET scanner is different. It uses plastic scintillators, the same kind of material in some cheaper radiation detectors. In plastic, 511 keV photons interact mostly through Compton scattering rather than total absorption. When a photon Compton scatters, it leaves behind a trail: the scattered photon goes one way, the recoil electron goes another, and the direction of the scattering plane encodes information about the photon’s polarization.
This is the key insight. If you can detect both annihilation photons and their respective Compton-scattered daughter photons, you can reconstruct the angle between the two scattering planes. That angle distribution tells you the degree of quantum entanglement with a single number, C_QE, where 1.0 means maximally entangled and 0.5 means separable.
The team injected the patient with 145 MBq of 68Ga-DOTA-TATE, a standard neuroendocrine tumor agent, and waited 90 minutes for it to distribute. Then they scanned him for 20 minutes in the J-PET scanner, a lightweight 60 kg machine that can be rolled up to a patient’s bedside. The scanner recorded not just the usual annihilation pairs (two photons hitting the detector) but also the much rarer four-hit events where both annihilation photons Compton-scattered inside the plastic.
What They Found
After filtering through 30.6% random coincidences and correcting for scatter, the team had enough four-hit events from the patient’s liver and spleen to construct angular correlation functions. The results sit squarely between the two theoretical extremes:
| Organ | Degree of Entanglement (C_QE) |
|---|---|
| Liver | 0.79 ± 0.21 |
| Spleen | 0.76 ± 0.23 |
| Maximally entangled (theory) | 1.00 |
| Separable photons (theory) | 0.50 |
The measured entanglement is real, statistically distinguishable from separable photons, but weaker than the textbook prediction. This matches what the same group found two years ago in a porous polymer phantom, where C_QE came out around 0.72. The reason, they argue, is the pick-off process: roughly 30-40% of positrons in tissue annihilate directly with molecular electrons rather than forming positronium, and those pick-off annihilations produce separable photons that dilute the entanglement signal.
Here is the process in schematic form:
Positron from 68Ga decay
│
├── Forms positronium (Ps)
│ │
│ ├── para-Ps (singlet) → 2 photons, maximally entangled
│ └── ortho-Ps (triplet) → pick-off → 2 photons, separable
│
└── Direct annihilation with molecular electron
│
└── 2 photons, separable
The measured C_QE values correlate with the fraction of pick-off annihilations, which in turn depends on oxygen concentration and the molecular architecture of the tissue. Ex vivo studies have shown that the ortho-positronium lifetime varies between 1.4 and 2.9 nanoseconds across different tissue types (healthy vs. cancerous, adipose vs. fibrous). The entanglement degree should vary with it.
Why It Matters
A PET scan today tells you one thing about each voxel of tissue: how much radioactive tracer accumulated there. That is useful. It shows metabolically active tumors. But it is a single dimension of information. Quantum entanglement imaging could add a second dimension: what kind of molecular environment is the tracer sitting in?
Two tumors might both light up brightly on a standard PET scan. But if one has a different C_QE value because its tissue is more hypoxic (oxygen-starved), or because its cellular structure is more disordered, that difference could change the diagnosis. Hypoxia is a known marker of aggressive cancer and resistance to radiation therapy. Current methods to measure hypoxia require separate, specialized PET tracers. Entanglement imaging might extract that information from the same scan, using the same tracer, for free.
How It Could Change Our Lives
The practical path from here is long but visible. The J-PET team is already building a total-body scanner that they estimate will have roughly 3 cps/kBq sensitivity for entanglement imaging. This is comparable to the sensitivity of a standard short-axial-FOV PET scanner. Commercial whole-body PET scanners like the uEXPLORER reach 174 cps/kBq for standard imaging. If even a few percent of those events can be tagged as four-hit Compton events, entanglement imaging becomes clinically viable without building new hardware from scratch.
That matters because existing PET infrastructure is enormous. There are thousands of PET/CT scanners in hospitals worldwide. If the technique works on crystal-based detectors as well as on plastic ones (and early simulations say it might, given that double Compton scattering events account for a few percent of all detections in conventional systems), then adding entanglement imaging becomes a software upgrade, not a hardware replacement.
The Bigger Picture
This is part of a broader shift in medical imaging. The J-PET group has been pushing positronium imaging, using the lifetime and decay modes of positronium atoms as biomarkers, for years. A human brain study was published in 2024. Entanglement imaging is the natural extension: if you can measure how long positronium survives before annihilating, and you can also measure how entangled its annihilation photons are, you get two orthogonal probes of the same molecular microenvironment.
The theoretical foundation paper that predicted the C_QE values for different tissues was published earlier this year in Bio-Algorithms and Med-Systems. It calculated that adipose tissue should give C_QE ≈ 0.89, water ≈ 0.87, and isooctane ≈ 0.78. The liver measurement of 0.79 falls right in that predicted range. The model is holding up.
Limitations & What’s Next
The statistics are thin. Twenty minutes of scan time. Fifty-seven MBq of remaining activity. Thirty percent random coincidences. The uncertainties on the C_QE values are large enough that you could not distinguish liver from spleen on entanglement alone. The event selection purity is decent (97% for liver, 94% for spleen), but the total number of useful four-hit events from a 20-minute scan is small.
The next step is the total-body J-PET, which should collect events much faster. The group is also developing iterative reconstruction algorithms (adapted from their positronium lifetime imaging work) that could produce voxel-by-voxel entanglement maps rather than organ averages. Meanwhile, other groups are working on cadmium-zinc-telluride detectors and electron-tracking Compton cameras that could bring the technique to conventional PET hardware.
What makes this study remarkable is not the precision of the numbers. It is the fact that the numbers exist at all. Quantum entanglement, the phenomenon that makes quantum computing possible, that underlies tests of Bell’s theorem, that most physicists assumed was too fragile to survive inside a warm, wet, noisy human body, turned out to be measurable with a $60,000 plastic scanner and a standard clinical tracer. If that holds up, it changes what we think medical imaging can see.
📄 Source: P. Moskal et al., “First-in-human quantum entanglement imaging,” arXiv:2606.29421v1 (2026). https://arxiv.org/abs/2606.29421