Computer scientists at Stanford University have created a deep learning algorithm that can diagnose skin cancer as accurately as a dermatologist, according to a new paper published in the journal Nature.
The algorithm was trained using a database of nearly 130,000 skin lesions using deep learning, a type of machine learning technique to visually diagnose a potential cancer. It was then tested against 21 dermatologists and successfully matched the performance of the dermatologists.
“We made a very powerful machine learning algorithm that learns from data,” said co-lead author of the paper and graduate student at Stanford, Andre Esteva. “Instead of writing into computer code exactly what to look for; you let the algorithm figure it out.”
The researchers trained an already-existing algorithm created by Google for image classification. The results showed that the algorithm was roughly 91% accurate to human doctors.
“There’s no huge dataset of skin cancer that we can just train our algorithms on, so we had to make our own,” said co-lead author of the paper and a graduate student at Stanford, Brett Kuprel.
“We gathered images from the internet and worked with the medical school to create a nice taxonomy out of data that was very messy – the labels alone were in several languages, including German, Arabic and Latin,” continued Kuprel.
In the future, the team would like to implement the algorithm on a mobile device; however, it still needs further testing in a real-world clinical setting.
“Advances in computer-aided classification of benign versus malignant skin lesions could greatly assist dermatologists in improved diagnosis for challenging lesions and provide better management options for patients,” said Professor of Dermatology and Director of the Pigmented Lesion and Melanoma Program at the Stanford Cancer Institute, and co-author of the paper, Prof Susan Swetter.
“However, rigorous prospective validation of the algorithm is necessary before it can be implemented in clinical practice, by practitioners and patients alike,” concluded Swetter.