Airdoc founder Ray Zhang: Artificial Intelligence will Revolutionize Industries

On June 15, the 2017 Forum for Future Medical Technology and Artificial Intelligence conference hosted by Bio-Valley was held in Shanghai. The goal of the conference was to share and discuss the development and application of artificial intelligence in the medical field.

The panel of speakers at the conference included Prof. Jianwei Zhang, Director of the Institute of the Multi-Modal Technology Systems and professor at the University of Hamburg in Germany, Mr. Ray Zhang, founder and CEO of Airdoc, Mr. Fabao Zhang, Chairman of Shanghai Metz Pharmaceutical Technology Co., Mr. Xubo Hu, the managing partner at Qiming Venture Partners.

Ray Zhang, panelist at 2017 Forum for Future Medical Technology and AI

Since the discovery of X-ray in 1895, X-ray has been widely used to examine the human body, to assist in the diagnosis of diseases, laying the basis for radiological and medical imaging, medical imaging has now become the most common diagnostic diagnostic tool.

Over the past few decades, medical imaging technology in China has developed rapidly, however imaging specialists have been in short supply, and mainly concentrated in large hospitals of large cities. Many small and medium-size cities do not have adequate imaging diagnostics resources; patients in smaller cities have found it necessary to travel to big cities in order to seek proper medical treatment.

Airdoc aims to solve the problem of inadequate medical imaging resources by leveraging scalable technology - artificial intelligence. Airdoc equips many smaller community-level health care institutions with AI medical image recognition capability previously only available at the best hospitals.

In the 60 years since the conception of artificial intelligence, lack of computing power and immature algorithms impeded its development. In 2006 the advent of “deep learning” brought significant revival to artificial intelligence. In 2012 the emergence of AlexNet brought about a turning point in AI. At the Large Scale Visual Recognition Challenge 2012 (ILSVRC2012), AlexNet bested the previous year’s top-5 error by 10 percentage points. Ever since, artificial intelligence in the field of image recognition has constantly been making and breaking records.

In recent years, articles appearing in Nature, JAMA, Science and other authoritative medical journals have begun writing about using artificial intelligence to solve medical problems. For example, “Artificial Intelligence identifies skin cancer” appeared on the cover of Nature. Science Magazine reported on computers predicting heart attack at higher accuracy rate than that of human doctors, etc. Meanwhile major Chinese hospitals have also begun to seriously study clinical applications of artificial intelligence.

Ray Zhang asserts that artificial intelligence has virtually limitless possibilities in the medical field. A few examples besides medical imaging include virtual nurse assistant, health management, medical risk analysis, drug extraction, auxiliary diagnosis and medical research. But medical artificial intelligence is still in its infancy, but its role in medical imaging recognition is well on its way.

In medical image recognition, the development of artificial intelligence requires massive numbers of medical images to generate algorithm models. Data volume and data quality are critical; high data volume increases the model’s inclusion rate, while accurate data and annotations ensure high accuracy of the model’s training and test sets.

Ray Zhang Dalei contends that within the next 15 years, artificial intelligence will play a critical role in the medical field. Artificial intelligence will be the core driving force behind the next revolution in medicine. AI can thoroughly analyze and aggregate medical knowledge to help provide higher quality clinical advice.

With its comprehensive inter-disciplinary integration, AI is set to facilitate the evolution of economic patterns by transforming across, commercial, financial and medical industries.