Mobile App Cuts Skin Cancer Referrals in Half, British Trial Shows

A mobile app designed to help detect potential skin cancer has reduced hospital referrals by more than 50% in a trial at three primary care clinics in Britain's National Health Service.

The Map My Mole app, which launched in March 2024, has reviewed more than 1,500 moles and lesions, streamlining the diagnostic process for both patients and providers, according to the company.

The system works by giving patients a small lens attachment for their phones to capture high-quality images of concerning skin spots. These images are then uploaded for assessment by consultants, bypassing traditional referral pathways for many benign conditions.

"In the three sites participating, we have reduced GP-to-hospital referrals by over 50%," said Dr. Toby Nelson, a dermatologist involved in the trial, to the BBC. "Previously, a GP would spend about 15 minutes assessing a mole. Now, reviewing the images online takes just two to four minutes."

Another location that participated in the trial reported a significant drop in further referrals — only 10% of the 400 patients evaluated were sent on for additional review.

Non-melanoma skin cancer is the most common type of cancer in the United Kingdom, though many cases are not included in national cancer statistics due to differences in how they’re recorded and treated, according to Cancer Research UK. Melanoma, the deadlier form, is the fifth most common cancer overall, with around 17,500 new cases each year.

While most cases are treatable when detected early, delays in diagnosis can lead to more complex care and poorer outcomes. The growing burden on dermatology services has fueled interest in technologies that can support earlier identification and reduce unnecessary specialist referrals.

Related: AI Model Shows Promise in Reducing Skin Cancer Detection Bias

Artificial intelligence is improving skin cancer detection, with studies showing a 15% to 20% improvement in prediction accuracy. In the United States, the FDA last year cleared the first AI-powered device designed to help primary care doctors evaluate skin lesions for cancer.

In a pivotal study led by the Mayo Clinic, the device, called DermaSensor, showed a 96% sensitivity rate across all skin cancer types and cut the rate of missed cancers in half. In a companion clinical utility study with 108 physicians, the DermaSensor device was found to decrease the number of missed skin cancers by half (from 18% to only 9%).

In the United States, one out of five Americans will have had some type of skin cancer by the age of 70. The annual cost of treating skin cancers in the United States is estimated at $8.1 billion, with an estimated 5.5 million new cases each year.

Such technologies have broader implications for dermatology services, where specialists currently dedicate significant time reviewing referrals that often prove benign. By reducing unnecessary specialist visits, consultants could potentially redirect resources to patients with chronic skin conditions.

“If we can reduce unnecessary referrals, consultants will have more time to see patients with eczema, acne, and psoriasis, who currently face waits of six to 18 months,” Dr. Nelson said.