Imagine sitting in your GP surgery, wondering if that new psoriasis treatment will work. Now imagine your doctor showing you a picture, not just of someone else, but of you, weeks from now, with your skin looking clearer.
Sounds a bit like magic, doesn’t it? But a team of UK researchers is bringing that idea closer to reality using AI (artificial intelligence).
They trained a clever computer system on real photos, which included 375 images of people with psoriasis. The system learned what mild, moderate, and severe psoriasis looks like, and even how the patches change over time. Then, using what’s called a "latent w-space" (basically, a fancy digital control panel), it could create new images showing skin getting better, less red, smaller patches, clearer skin.
It’s not a treatment, but it could be a powerful tool, helping people see what to expect, and helping doctors talk through treatment options with a clear visual.
With more testing, it could one day be used in clinics, giving people a hopeful, personal preview of progress. And that kind of hope? That’s something science can’t measure, but it matters.
Comment
While this technology isn’t in everyday clinics just yet, it shows just how powerful AI could be in the future of healthcare, not just for doctors, but for patients too. Giving people a visual preview of how their skin might respond to treatment, can help take some of the fear and guesswork out of managing psoriasis. It’s not a cure, but it’s a step toward making treatment feel more personal, more hopeful, and more informed. And sometimes, seeing even a small glimpse of improvement, even just on a screen, can make a big difference in how we feel about the road ahead.
Reference:
Scott, J., Grant-Jacob, J.A., Praeger, M. et al. Modifying the severity and appearance of psoriasis using deep learning to simulate anticipated improvements during treatment. Sci Rep 15, 7412 (2025). Full article