The professor is 47. The AI model that analyzed his data pinned his biological age at around 32. Between those two numbers sits nearly two years of self-tracking, extreme fasting, daily strength work and a Mediterranean-style diet.
The study, called DELTA, was published in August 2026 in PLOS One and has one participant. He is also the senior author.
Dean Ho is Provost's Chair Professor of Biomedical Engineering and Pharmacology at the National University of Singapore's Yong Loo Lin School of Medicine, where he leads the Institute for Digital Medicine.
Ho's argument for enrolling himself is straightforward. Annual checkups miss what happens in between, and a single blood draw cannot capture how the body handles a stressor, a bad night's sleep or a long fast.
Also read: One plate rule, 10 dinners under 600 calories that still leave you full
What he actually did
The protocol layered several interventions. A four-hour daily eating window, roughly 20 hours of fasting, with periodic 48-hour water-only fasts on top. Daily strength training paired with Norwegian-style high-intensity intervals two to three times a week.
A Mediterranean-inspired diet added leafy greens, seeds, olive oil, lean protein and specific supplements. Bedtime shifted earlier to about 9:15 pm.
Everything was measured. The research team tracked apolipoprotein B and A, high-sensitivity C-reactive protein, homocysteine, blood glucose and blood ketones, alongside wearable data on sleep architecture and gut microbiome composition. A custom AI tool crunched the readouts.
What the data showed
Sleep saw the largest shift. Total sleep rose from about 372 minutes a night at retrospective baseline to 493 minutes, and REM sleep climbed from 65 to 108 minutes.
Also read: This unexpected symptom can appear after a solar eclipse
Metabolic switching, the body's ability to move from burning glucose to burning ketones, sped up across repeated fasts. Apolipoprotein B rose during the fasted state but returned to baseline afterward, which the paper reads as efficient lipoprotein clearance.
The custom AI tool estimated Ho's biological age at 31.8 years against his chronological age of 45 at the time of the readout. He is now 47, which is where the 15-year gap in the headline comes from.
This is not the first time diet and daily habits have moved biological-age markers. A separate study reported measurable shifts in those markers after just four weeks of dietary change.
The limits of one man's data
The authors are direct about the caveats. The paper acknowledges the N=1 design and states that no claims are made about the DELTA protocol's efficacy beyond within-subject comparisons.
Also read: Can you stand on one leg for 10 seconds? The answer predicts your risk of dying in the next 7 years
There was no control group. Multiple interventions ran at once, so no single change can be credited with the shift.
Biological-age tools carry their own limits. A February 2026 review in eBioMedicine noted that current clocks capture broad trends but often diverge in their estimates.
Most derive their predictions from correlation, not causation. That leaves open what such a change actually means for a person's health.
For readers, DELTA is less a template than a proof of concept. Continuous tracking of the same body across time surfaces patterns that annual screenings cannot see. Whether it makes anyone measurably younger is a separate question worth raising with a doctor.
Also read: Woman suffers stroke in her sleep and shares the first warning signs
Also read: What eating one avocado a day can do to your cholesterol, skin and weight
This article is made and published by Jesper Bengtson, who may have used AI in the preparation.