Saturday December 7, 2019

AI Outwits Doctors at Detecting Skin Cancer

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

0
//
Previous research has shown that obesity and high-fat diets both together and independently increase the risk of pancreatic cancer.
The actress was diagnosed with cancer earlier this year. Pixabay

An Artificial Intelligence (AI) system has been found to detect skin cancer more accurately than a group of experienced dermatologists from 17 countries around the world, a study said on Tuesday.

In the experiment, the team of researchers from Germany, France and the US trained a form of AI or Machine Learning known as a deep learning convolutional neural network (CNN) to identify skin cancer by showing it more than 100,000 images of malignant melanomas — the most lethal form of skin cancer — as well as harmless moles.

When its performance was compared with that of 58 international dermatologists, the CNN missed fewer melanomas and misdiagnosed benign moles less often as malignant than the group of dermatologists, showed the findings published in the journal Annals of Oncology.

“The CNN works like the brain of a child. To train it, we showed the CNN more than 100,000 images of malignant and benign skin cancers and moles and indicated the diagnosis for each image,” said first author of the study Professor Holger Haenssle from the University of Heidelberg in Germany.

“Only dermoscopic images were used, that is lesions that were imaged at a 10-fold magnification. With each training image, the CNN improved its ability to differentiate between benign and malignant lesions,” Haenssle added.

Representational image (AI)
Representational image (AI). Pixabay

A CNN is an artificial neural network inspired by the biological processes at work when nerve cells (neurons) in the brain are connected to each other and respond to what the eye sees.

The CNN is capable of learning fast from images that it “sees” and teaching itself from what it has learned to improve its performance — a process known as Machine Learning.

“These findings show that deep learning convolutional neural networks are capable of out-performing dermatologists, including extensively trained experts, in the task of detecting melanomas,” Haenssle said.

The incidence of malignant melanoma is increasing, with an estimated 232,000 new cases worldwide and around 55,500 deaths from the disease each year.

Also Read: Satya Nadella: Robots Won’t Make People Jobless

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

But despite the promising results from the experiment, the researchers do not envisage that the CNN would take over from dermatologists in diagnosing skin cancers, but that it could be used as an additional aid.

“This CNN may serve physicians involved in skin cancer screening as an aid in their decision whether to biopsy a lesion or not,” Haenssle said. (IANS)

Next Story

US Chipmaker Intel Eyes AI on ‘Edge Computing’

It is designed to accelerate AI tasks, particularly ones like image processing, delivering performance up to six times the power efficiency of existing processors

0
Intel on Wednesday unveiled eight additional 10th Gen Intel Core processors for modern laptop computing.
Intel on Wednesday unveiled eight additional 10th Gen Intel Core processors for modern laptop computing. Pixabay

US chipmaker Intel Corp. has said that it will focus on “edge computing” that could hold the key to the success of artificial intelligence (AI) in the future.

Edge computing refers to the practice of storing data on computers located near cell towers and other network equipment to improve network response times. It is different from today’s Cloud-based system, where information is sent to a distant data centre, Yonhap news agency reported on Wednesday.

“Forty-three per cent of AI tasks will be handled by edge computing in 2023,” Kwon Myung-sook, CEO of Intel Korea, said in a statement during a forum in Seoul.

“AI devices empowered with edge function will jump 15-fold.”

The expansion of computing at the edge is an important growth opportunity for the chip giant — an estimated $65 billion market by 2023, Intel said.

Huawei, Atlas 900, World
The future of computing is a massive market worth more than two trillion US dollars. Pixabay

More AI is being incorporated into edge devices, from Internet of Things (IoT) devices to smartphones, as AI algorithms improve, according to the company.

“Innovation in edge computing has become necessary where data is most produced,” Kwon said. “It is why Intel is preparing a platform solution that can cover both hardware and software.”

Intel said AI will support and provide new services in eight key industries, including smart cities, robots and gaming.

Also Read: India Ranked 5th Worst Country For Use of Biometric Data: Comparitech Report

In order to do so, Intel said it will launch the next-generation Movidius Vision Processing Unit next year.

It is designed to accelerate AI tasks, particularly ones like image processing, delivering performance up to six times the power efficiency of existing processors. (IANS)