Artificial intelligence (AI) applications for healthcare has made groundbreaking advancements with new research showing a 100% detection rate for melanoma, the most severe form of skin cancer, using specialized AI software.

Presented at a medical conference in Berlin, the study is considered a pivotal moment in the convergence of technology and medicine.

Spanning 2.5 years and involving more than 22,000 patients with suspected skin cancers, the research detected all 59 cases of melanoma presented.

Moreover, the AI’s detection rate was astonishingly precise across the board, with a 99.5% accuracy rate for identifying all types of skin cancer—missing just one case out of 190—and a 92.5% effectiveness in identifying pre-cancerous lesions.

“This study has demonstrated how AI is rapidly improving and learning, with the high accuracy directly attributable to improvements in AI training techniques and the quality of data used to train the AI,” said lead author Dr. Kashini Andrew, specialist registrar at University Hospitals Birmingham NHS Foundation Trust in the United Kingdom, in a news release from the European Academy of Dermatology and Venereology.

The research highlights a quantum leap in AI’s capabilities compared to previous versions. In 2021, the first version of this specific AI software detected nearly 86% of melanomas, 84% of all skin cancers, and 54% of pre-cancerous lesions.

The current, third iteration of the software also brings practical benefits to the healthcare system. “The latest version of the software has saved over 1,000 face-to-face consultations in the secondary care setting between April 2022 and January 2023, freeing up more time for patients that need urgent attention,” Andrew added.

However, despite this remarkable leap in detection capabilities, the researchers are unequivocal about the limits of the technology. AI should not serve as a standalone detection tool and must operate in concert with dermatologists for clinical oversight.

Among the cases of basal cell carcinoma, the AI failed to detect one, which was subsequently caught by a dermatologist serving as a “safety net.”

“We would like to stress that AI should not be used as a stand-alone tool in skin cancer detection and that AI is not a substitute for consultant dermatologists,” emphasized study co-author Dr. Irshad Zaki, also a consultant dermatologist at University Hospitals Birmingham NHS Foundation Trust.

Dr. Andrew pointed out that the role of AI in dermatology remains a subject of debate. “Further research with appropriate clinical oversight may allow the deployment of AI as a triage tool. However, any pathway must demonstrate cost-effectiveness, and AI is currently not a stand-alone tool in dermatology. Our data shows the great promise of AI in future provision of health care,” she said.

The study findings were unveiled at a meeting of the European Academy of Dermatology and Venereology, held in Berlin. Importantly, these findings should be considered preliminary until subjected to peer review and published in a respected medical journal.