Mode
Text Size
Log in / Sign up

AI decision support in nurse-led teledermoscopy associated with 1.73 odds ratio for malignancy detectionAI tools help nurses identify more skin cancer cases

AI-generated summary of the cited source, checked by automated accuracy review. How we work

Key Takeaway
Note that AI decision support was associated with higher malignancy detection rates in nurse-led teledermoscopy.

This multi-site evaluation analyzed 1,102,382 lesions from 98,422 patients across MoleMap clinics in New Zealand and Australia. The study compared nurse-led teledermoscopy with standard nurse-led teledermoscopy without AI decision support. The primary outcome was the malignancy detection rate.

In the AI-assisted group, the malignancy detection rate was higher than the standard group, with an odds ratio of 1.73 (95% CI 1.68-1.77). Specifically, the AI-assisted group identified 25.5 malignancies per 1,000 lesions compared to 15.7 in the standard group. Regarding secondary outcomes, the AI-assisted group showed 21 additional intervention recommendations and 13 additional safety-netting recommendations per 1,000 lesions. Conversely, there were 34 fewer no-action recommendations per 1,000 lesions in the AI-assisted group.

No safety or tolerability data were reported. The study was not randomized, and the findings are associational. Clinical application of these results should be interpreted with caution as prospective, outcome-based confirmation is required to establish a causal link between AI decision support and improved clinical outcomes.

When a patient shows a suspicious mole, the first step is often a screening to see if it needs a biopsy. In a large study across Australia and New Zealand, researchers looked at how artificial intelligence (AI) helped nurses during these screenings. They compared nurses using AI decision support against those using standard methods to see if the technology changed how they managed skin lesions.

The study looked at over one million skin lesions from more than 98,000 patients. The results showed that nurses using AI were more likely to detect malignancies, which are cancerous growths. Specifically, the AI group found 25.5 malignancies per 1,000 lesions, while the group without AI found 15.7. The AI also led to more recommendations for immediate action or safety-netting, which means extra steps to ensure a patient is monitored closely.

While the results are promising, it is important to note that this was an observational study, not a randomized trial. This means the researchers could not prove that the AI caused the better results, only that the two methods were associated with different outcomes. More prospective studies are needed to confirm these findings before they can be fully relied upon in every clinic.

What this means for you:
AI tools were associated with higher detection rates of skin cancer and more frequent treatment recommendations.

Common questions

How did the AI help nurses identify skin cancer?

The AI acted as a decision support tool. In the study, nurses using this AI were associated with a higher malignancy detection rate, finding 25.5 cases per 1,000 lesions compared to 15.7 in the group without AI. It also led to more frequent recommendations for intervention and safety-netting.

Is this AI tool a replacement for doctors?

The study focused on nurse-led teledermoscopy, which is a way to examine skin lesions remotely. While the AI showed a higher detection rate, the study was observational and not randomized, meaning more research is needed to confirm how these tools should be used in clinical practice.

What are the limitations of this study?

Because the study was not randomized, the results are associational rather than causal. This means we cannot say for certain that the AI caused the higher detection rates. More prospective, outcome-based studies are needed to confirm these findings before they can be fully adopted.

Study Details

Study typeRct
Sample sizen = 98,422
EvidenceLevel 2
PublishedOct 2026
View Original Abstract ↓
Nurse-led skin cancer screening extends specialist reach but depends on the nurse selecting which lesions to forward for diagnosis, and performance varies with experience. We evaluated whether real-time decision support using artificial intelligence (AI) improves malignancy detection in routine nurse-led teledermoscopy screening across MoleMap clinics in New Zealand and Australia (January 2024-July 2025). In this real-world study, clinics using AI decision support were compared with standard clinics. Nurses examined patients, captured dermoscopic images, and forwarded selected lesions to teledermatologists, who provided the reference diagnosis. The analytic cohort comprised 1,102,382 lesions from 98,422 patients across 577 sites. AI-assisted screening was associated with a higher malignancy detection rate than standard screening (25.5 vs 15.7 malignancies per 1,000 lesions; odds ratio adjusted for nurse experience 1.73, 95% CI 1.68-1.77), a finding consistent across all nurse-experience tiers and the three major malignant subtypes. AI assistance was also associated with a shift in recommended management: 21 additional intervention recommendations and 13 additional safety-netting recommendations (self-monitoring, short-term follow-up, or specialist referral) per 1,000 lesions, approximately balanced by 34 fewer no-action recommendations. The reference standard was teledermatologist diagnosis and allocation was not randomised; findings are therefore associational and require prospective, outcome-based confirmation.
Free Newsletter

Clinical research that matters. Delivered to your inbox.

Join thousands of clinicians and researchers. No spam, unsubscribe anytime.