Artificial intelligence may have treated a woman’s acne after suffering from the condition for more than two years. Remarkable transformational images show how Rachel, 27, from Zaandam in the Netherlands, used machine learning to overcome hormonal breakouts. Developed in Denmark, the method creates and recommends tailored solutions that could impact billions of people around the world struggling with their skin.
Daniel Jensen, CEO and co-founder of dermatology brand NØIE, which translates to “meticulous” from Danish, hopes the algorithm his company has developed will help people suffering from skincare issues around the world.
NØIE’s technology, which is free to use, seeks to personalize skin care through the use of data, science and technology to offer products that have a statistical probability of helping each user.
Mr. Jensen said: “Every time someone uses their personalized solution and gives us feedback, they are helping someone else. Since launching our data model, we have collected a massive amount of data from approximately 80,000 people with different skin types and conditions, and currently 86 percent report an improvement in their skin using our solution.
“Our goal is to create a large community of people with skin problems who, through their own improvement, will help others to find their solution.”
READ MORE: NHS uses robots to cut waiting lists
NØIE’s algorithm is built on network effects, meaning the more user feedback it gets, the better it can recommend the right skincare regimen for future users.
Skin conditions are the fourth leading cause of all human diseases, according to the British Journal of Dermatology, and affect almost a third of the world’s population – meaning the technology Rachel uses could impact billions worldwide.
It uses Bayesian modeling, a statistical method that allows the model to incorporate prior knowledge and data into the analysis, which the company says has particular relevance to skin care based on many individual factors, such as skin type and symptoms, and its effectiveness of treatments.
Whenever a patient receives a solution predicted by the data model, they are asked to provide feedback on their progress after 30 days, whether negative or positive.
The real-life feedback data is then integrated into the model, teaching it about the effectiveness of treatments.
Commenting on how he developed the innovative method, Mr. Jensen said: “We wanted to see if we could solve some of the challenges in the skin care industry and offer better solutions through innovation.
“With a team that had expertise in dermatology and data science, we spent the first three years researching existing over-the-counter products and collecting feedback from users to reverse engineer our own personalized products for people.”
dr Ivy Lee, a practicing dermatologist in the US and vice chair of the American Academy of Dermatology’s Augmented Intelligence Committee, advocates the AI-driven approach to patients.
On the need for personalization in dermatology, she said, “Patients are so fed up. All other aspects of her life have been transformed by technology, but her healthcare experience is still embarrassingly awful.”
“This is where personalization goes for the rest of our lives.”
Mr. Jensen explained how the story of a young woman suffering from eczema was the driving force behind his innovative skin care solution.
He said: “In her frustration, she studied the latest publications on ingredients in dermatology – hoping to find information or research on new cures that might help her.
“I just couldn’t understand why she had to go that far.”
Mr. Jensen wanted to leverage his expertise from previous positions he held in digital and reverse engineering of products to better serve consumer needs.
With the support of Danish pharmaceutical giant LEO Pharma, Mr. Jensen and his business partner Chris Christiansen initiated an experiment with the aim of offering better solutions to people with skin problems.
In their research, they screened 2,938 over-the-counter cosmetics, finding that the ingredients in 81 percent of the products examined were classified as potentially harmful to problem skin, and only 1 percent of the products had been tested for efficacy.
Based on these insights, NØIE aims to develop personalized solutions based on individuals’ unique data to offer what they describe as “an alternative approach to the mass-produced one-size-fits-all products available on the market”.
Since its inception in 2019, the company claims its success rate has increased from 30 percent to 86 percent, but acknowledges that more research is needed before its technology can be used for treatments at scale.
The NØIE team is now offering to share their findings with the dermatological world to help validate their findings.