Read time: 6 minutes
Here are four pieces of research and health news that piqued my curiosity this week. One made my radiographer brain stop scrolling. Another caught my attention because I’m 40 weeks pregnant and would currently read anything promising better sleep.
Here’s what caught my eye.
Midjourney just announced a full-body scanner. And yes, that Midjourney.
I spent the better part of an afternoon reading everything I could find about this one.
Midjourney… the company most people know as the AI image generator, has announced Midjourney Medical, a new division with an audacious first product: a full-body ultrasonic scanner designed to image the whole body in under 60 seconds with no radiation and no magnetic fields.
The technology is called Fullbody Ultrasonic CT, without the ionising radiation. The system uses sound waves and water: you step onto a platform, descend into a shallow pool, and pass through a ring of underwater ultrasonic sensors. The underlying hardware is based on Butterfly Network’s ultrasound-on-chip technology under licence. Midjourney’s image-generation models are not involved.
The company claims the scanner can map more than 25 organs and anatomical structures with detail approaching MRI and CT accuracy. Midjourney has committed over $74 million to the project, with a first San Francisco location planned for late 2027.
The regulatory situation is what you’d expect for a prototype this early: no FDA approval for diagnostic use. The company plans to launch initially offering body composition maps rather than clinical diagnoses. A deliberate step around the regulatory pathway while the technology matures.
As someone who has spent years working in radiology and medical imaging, here’s my take:
The underlying science is legitimate. Acoustic reconstruction and AI processing is a promising and active research area, but “rivals MRI and CT accuracy across 25+ organs in 60 seconds” is a claim that needs peer-reviewed clinical validation, not a press release. MRI and CT achieve their resolution through fundamentally different physics. That said, if even a fraction of it eventuates, affordable, radiation-free, full-body screening without a referral or a six-week wait is a direction worth taking seriously.
This one is worth watching. Just not worth staking a clinical decision on yet.
TLDR: Midjourney has announced a full-body ultrasound CT scanner. 60 seconds, no radiation, no magnetic fields, using sound waves and water. No FDA approval for diagnostic use. Ambitious, scientifically plausible, and a long way from clinical validation. But if it delivers, the implications for accessible preventive imaging are genuinely significant..
Further reading: Midjourney Medical announcement · Engadget coverage
A blood test for endometriosis just got a lot more serious
Endometriosis affects roughly one in seven women, yet diagnosis still takes somewhere between seven and ten years on average. That number keeps appearing in studies, and it keeps being appalling.
Part of the problem is that ultrasound and MRI can miss the disease, while definitive confirmation has traditionally required laparoscopic surgery.
A 2026 multicentre validation study may be a meaningful step toward changing that.
Researchers tested HerResolve, an AI-assisted blood test combining several biomarkers with age and BMI, in a multicentre study involving 298 women across 11 clinical sites in the US, Europe and Hong Kong.
The results were striking: 80% sensitivity, 97.5% specificity and, most remarkably, it identified 61.5% of confirmed endometriosis cases that ultrasound and MRI had missed.
The blood test was finding disease in women whose imaging had come back clear.
It is not yet part of standard clinical care, and one study is not the end of the story. But this is the kind of result that moves a test from interesting idea to something clinicians should be watching closely.
For women stuck in the diagnostic waiting room, it feels like the first credible sign that a better path may be coming.
TLDR: A non-invasive blood test for endometriosis achieved 80% sensitivity and 97.5% specificity across 11 international clinical sites and caught 61.5% of cases already missed by standard imaging. Still early but significant.
Further reading: Wong et al. Journal of Minimally Invasive Gynecology, 2026
Scientists induced the restorative effects of deep sleep in awake mice
I’ll be honest: I am 40 weeks pregnant and this one is self serving. Anything involving the words “sleep” and “breakthrough” in the same sentence has my full, undivided, slightly desperate attention.
Researchers at the University of Wisconsin–Madison have published a study in Nature Neuroscience showing that some restorative functions of deep sleep can be reproduced in the brains of awake mice.
During deep sleep, neurons cycle between brief periods of activity and silence. These ON/OFF patterns are thought to help reduce sleep pressure, recalibrate connections between neurons and support memory.
Using optogenetics, essentially light-based tools for controlling specific neurons, the researchers recreated these sleep-like patterns in targeted areas of the cortex while the mice were still awake.
The result was surprisingly sleep-like: local sleep pressure fell, markers of synaptic strength were reset and memory consolidation improved in sleep-deprived mice.
In other words, the researchers reproduced some of sleep’s restorative work without putting the animals to sleep.
There is an enormous caveat. This was a mouse study using genetic modification and implanted light-based technology. Nobody is about to replace a full night’s sleep with a quick brain tune-up.
TLDR: The restorative effects of deep sleep were replicated in awake mice by inducing sleep-like brain patterns in specific cortical regions. Cognitive performance matched fully rested controls. Mouse study, early stage, but a genuinely fascinating proof of concept.
Further reading: Driessen K et al. Induction of cortical ON/OFF periods in awake mice fulfills sleep functions. Nature Neuroscience, 2026
Astaxanthin: the antioxidant in your salmon that’s been quietly cracking on
Astaxanthin is the compound that makes salmon, shrimp and flamingos pink. It belongs to the carotenoid family alongside beta-carotene and has one of the strongest antioxidant profiles in the group.
The skin-ageing claims have circulated for years, so I was interested when a systematic review and meta-analysis examined the human evidence.
The review, published in Nutrients, pooled eleven studies: nine randomised controlled trials of oral astaxanthin and two combining oral and topical use. Supplementation was associated with improved skin moisture and elasticity, along with a reduction in wrinkle appearance compared with placebo.
The evidence is nowhere near the level of prescription retinoids. The studies were small and varied in dose and duration. But the results were consistent enough to suggest there may be something behind the marketing.
The proposed mechanism is reduced oxidative damage from UV exposure and better protection of collagen-producing cells.
Worth knowing, but better viewed as one small part of a broader skin-ageing strategy than a standalone solution.
TLDR: A systematic review of 11 human studies found that astaxanthin was associated with improved skin moisture, elasticity and reduced wrinkles. Small studies, modest effects, but a consistent and plausible signal.
Further reading: Zhou et al. Systematic Review and Meta-Analysis on the Effects of Astaxanthin on Human Skin Ageing. Nutrients, 2021
Next Up
Your gut. It turns out the trillions of bacteria living in your digestive tract aren't just handling digestion. A specific community of them is quietly deciding how much estrogen circulates through your body, which means they're influencing your skin, your cycle, your mood, and the way you'll age decades from now.
What the research is uncovering challenges a lot of what we've been told about "gut health." We'll separate what's proven from what's just hype, and as always, land on the small, everyday things you can actually start doing.