Medicine

AI better at finding skin cancer than doctors

AI better at finding skin cancer than doctors

But in a new study of an artificial intelligence (AI)-based dermatology diagnostic system, the algorithms outperformed humans even without important demographic and contextual information about the images presented to them. In order to teach the artificial neural network how to identify skin cancer, the researchers fed it a dataset of over 100,000 images of malignant melanomas and benign moles.

After training the CNN to recognise the differing forms of skin cancer, the researchers pitted it against that of 58 global dermatologists.

Results of the study, published this week in the Annals of Oncology, show that the CNN missed fewer melanomas and misdiagnosed benign moles as malignant less often than the group of experienced dermatologists.

The dermatologists' performance improved when they were given more information about the patients and their skin lesions.

First author of the study, Professor Holger Haenssle, senior managing physician at the department of dermatology at the University of Heidelberg in Germany, said: 'The CNN works like the brain of a child. The flesh as well as blood dermatologists were able to detect 86.6 percent of the skin cancers accurately from the images while the CNN was able to detect 95 percent skin cancers accurately. The American Cancer Society estimates 9,320 people will die from melanoma in 2018 while 91,270 new cases will be diagnosed. Still, they were outperformed by the artificial intelligence system, which was working exclusively from the images. The study results explained that an AI system was more proficient than human dermatologists at detecting the cancer cells. The researchers then developed a pair of test sets (which were not involved in the system's training) and collected 58 dermatologists from 17 different countries.

More news: Germany: Huge fire at Europe's 2nd most popular theme park

It can be cured if detected early, but many cases are only diagnosed when the cancer is more advanced and harder to treat.

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).

"Only dermoscopic images were used, that is lesions that were imaged at a 10-fold magnification", Professor Haenssle said in a statement. Well, in diagnosing skin cancer perhaps it is.

The research team proclaimed that the AI can be used as a tool in faster and easier diagnosis of skin cancer. "Irrespective of any physicians' experience, they may benefit from assistance by a CNN's image classification".

Will Artificial Intelligence Take Over The World?