Computer better at finding skin cancer than doctors

Adjust Comment Print

A new study has found that artificial intelligence is better than humans at detecting skin cancer.

Researchers from France and Germany trained the CNN by showing it more than 100,000 photos of malignant melanomas as well as and benign moles and lesions.

For level two, the dermatologists improved by diagnosing 88.9 percent of malignant melanomas and 75.7 percent benign when given additional clinical information and close-up images, but again were outperformed by the CNN.

Lead researcher Professor Holger Haenssle, from the University of Heidelberg, said: "It works like the brain of a child".

On average, flesh and blood dermatologists accurately detected 86.6 per cent of skin cancers from the images, compared to 95 per cent for the CNN.

More news: Where is Melania Trump? First lady keeps low profile after hospital stay

There are about 232,000 new cases of melanoma, and 55,500 deaths, in the world each year, they added.

A form of artificial intelligence was taught to recognize signs of the disease after being shown more than 100,000 images of melanoma. In a second test, only the dermatologists were presented with further information such as age, sex, and the position of the lesion alongside the same images.

From just dermoscopic images, the clinicians were asked first to diagnose either a benign or malignant condition and asked to assign a care plan (level 1).

Although the CNN algorithm will not replace human doctors, the researchers believe that it can be used to aid doctors to diagnose skin cancer faster and better.

"Only dermoscopic images were used, that is lesions that were imaged at a 10-fold magnification", Professor Haenssle said in a statement. "Irrespective of any physicians' experience, they may benefit from assistance by a CNN's image classification".

More news: Singapore Airlines announces 19-hour ultra-marathon to NY

However, they write that the study shows "artificial intelligence (AI) promises a more standardized level of diagnostic accuracy, such that all people, regardless of where they live or which doctor they see, will be able to access reliable diagnostic assessment". These include the difficulty of imaging some melanomas on certain locations of the body, such as on the fingers, toes and scalp.

A type of machine learning called a convolutional neural network was trained...

"Currently, there is no substitute for a thorough clinical examination". Others came from countries such as Switzerland, Australia, Japan, and Argentina.

The cancer-sniffing deep learning artificial intelligence will work side-by-side with human doctors.

More news: After Giro win, Froome quickly changes focus to Tour