AI and Registry Data May Help Spot People at Higher Melanoma Risk, Study Finds

GOTHENBURG, Sweden — In a significant step toward precision medicine, a massive new study from Sweden suggests that artificial intelligence (AI) can analyze routine health registry data to identify individuals at a high risk of developing melanoma years before a diagnosis occurs.

By analyzing the medical histories of more than 6 million adults, researchers at the University of Gothenburg found that AI models could distinguish those likely to develop the deadliest form of skin cancer with significantly higher accuracy than traditional methods. The findings, published this week, offer a potential roadmap for shifting from broad, population-wide screenings to a targeted approach that prioritizes those most in need of clinical intervention.

Targeted Prediction: How AI Sifts Through the Data

Melanoma remains a major global health challenge. In the United States alone, the American Cancer Society estimates approximately 112,000 new invasive melanoma cases and 8,510 deaths in 2026. While early detection is the “gold standard” for survival, identifying exactly who should be screened—and how often—has long been a logistical hurdle for healthcare systems.

The Swedish study, led by researchers at the Sahlgrenska Academy, leveraged the country’s comprehensive health registries to see if “invisible” patterns in medical records could predict future cancer. The study included:

• 6,036,186 individuals (the total adult population of Sweden).

• 38,582 confirmed melanoma cases (0.64% of the cohort) over a five-year period.

• Data points including age, sex, previous diagnoses, medication use, and socioeconomic factors.

The results were striking. While using only age and sex to predict melanoma yielded an accuracy of about 64%, the most advanced AI model increased that figure to 73%. Most notably, the AI was able to isolate small, high-risk subgroups where the probability of developing melanoma within five years was as high as 33%.

From “One Size Fits All” to Precision Screening

Current screening methods often rely on a patient’s own initiative or broad age-based recommendations. However, this study suggests that the data already sitting in electronic health records (EHRs) could act as an early warning system.

“The work shows that data already present in healthcare systems can identify people at higher risk,” said Martin Gillstedt, a doctoral student at the University of Gothenburg and a statistician at Sahlgrenska University Hospital. He noted, however, that while the results are promising, this is not yet ready to be used as a routine decision-support tool in the clinic.

Dr. Sam Polesie, Associate Professor of Dermatology and Venereology at the University of Gothenburg, believes the value lies in efficiency. “Selective screening of small high-risk groups could improve monitoring and use healthcare resources more efficiently,” Polesie explained. He emphasized that the AI is intended to supplement clinical judgment, not replace the expert eye of a dermatologist.

The Evolution of AI in Dermatology

This research builds upon a 2022 proof-of-concept study published in Acta Dermato-Venereologica, which first tested the feasibility of using neural networks to predict melanoma from registry data. The new 2026 analysis expands that scope significantly, moving from a theoretical model to a massive, real-world population study.

In the dermatology community, there is a growing consensus that “risk stratification”—ranking patients by their likelihood of disease—is the future of cancer prevention. Known risk factors such as heavy sun exposure, indoor tanning, family history, and a high mole count are well-documented by the American Academy of Dermatology (AAD). The AI model excels by aggregating these factors with more subtle indicators, such as socioeconomic status and medication history, to create a more holistic risk profile.

Limitations and Practical Realities

Despite the impressive statistics, experts urge a balanced interpretation of the data.

1. Geographic Specificity: The study relied exclusively on Swedish registries. Sweden has a relatively homogenous population and a specific healthcare infrastructure. It remains unclear if the model would perform as well in more diverse populations or in countries like the U.S., where healthcare data is often fragmented across different providers.

2. Prediction vs. Prevention: A high-risk score from an AI does not mean a patient has cancer, nor does it prevent it. “Even a model with strong performance can miss some cases and flag some people who will never develop melanoma,” the researchers cautioned.

3. Clinical Integration: We are still in the early stages of determining how a doctor would use this information. If an AI flags a patient as “high risk,” does that trigger a biopsy, a more frequent skin check, or simply a lifestyle counseling session?

What This Means for You

While the healthcare industry works to integrate these high-tech tools, the fundamental advice for the public remains unchanged. Early detection starts with personal awareness and traditional protection.

“AI may soon help doctors identify melanoma risk more precisely, but everyday prevention still matters,” the AAD notes.

Steps for Health-Conscious Consumers:

• Know Your History: Be aware of personal or family histories of skin cancer.

• Monitor Your Skin: Perform regular self-exams for new or changing spots.

• Sun Safety: Continue using broad-spectrum sunscreen, wearing protective clothing, and avoiding indoor tanning beds.

• Consult a Professional: If you have fair skin, a high number of moles, or significant past sun damage, talk to your doctor about an individualized screening schedule.

As AI continues to mature, it will likely become a silent partner for physicians, helping them find the “needle in the haystack” and ensuring that those at the highest risk for melanoma receive the life-saving attention they need before it’s too late.

References

• https://www.news-medical.net/news/20260415/Registry-data-and-AI-can-identify-high-risk-populations-for-skin-cancer.aspx

Medical Disclaimer: This article is for informational purposes only and should not be considered medical advice. Always consult with qualified healthcare professionals before making any health-related decisions or changes to your treatment plan. The information presented here is based on current research and expert opinions, which may evolve as new evidence emerges.