Artificial Intelligence is demonstrating the potential to improve our dermatology care and services.
Currently this potential requires ongoing ‘real world’ dermatology clinic evaluation supervised by experienced dermatologists to ensure quality and safety.
A strong focus of artificial intelligence is the early diagnosis of melanoma. Various algorithms/scoring systems have been developed which are designed to support medical practitioners in detecting appearances or changes in appearance from images (photos) of skin lesions to assist with this goal. These artificial intelligence algorithms are promising, but require ongoing evaluation.
At Perth Dermatology, artificial intelligence is integrated into the imaging systems of our high risk melanoma surveillance clinic. Importantly, this artificial intelligence remains fully supervised by our experienced dermatologists, as we continue to evaluate the real world potential and work flows (augmented intelligence).
In Australia, artificial intelligence algorithms are considered as Software as a Medical Device (SaMD) by the Therapeutic Drugs Administration (TGA). TGA approval is required for dermatology algorithms to be used for diagnosis, prevention, monitoring, prediction, prognosis and treatment. These algorithms currently require further research and real world evidence before a thorough risk assessment process can be undertaken by the TGA for approval. This serves to highlight the current position of Perth Dermatology, that expert dermatologist supervision occurs alongside the evaluation of artificial intelligence, to ensure quality and safety.
