At the Alibaba Cloud Summit Shenzhen Summit held today, Alibaba Cloud and Intel and ZeroOne Technology jointly announced the launch of the Tianchi Medical AI Contest. The first quarter of the contest will challenge the diagnosis of early lung cancer. It is said that Tianchi is the world's largest intelligent platform and brings together more than 60,000 AI algorithm scientists. Specifically, the medical big data platform Zhanji Technology will provide decontamination processing for the contest and high-definition chest CT scan image data marked by experts. The player needs to obtain the nodule feature through the original CT image training model algorithm, and finally realizes the intelligent judgment of the nodule area of ​​the image picture. The current malady of cancer treatment: misdiagnosis and missed diagnosis A nodule is a descriptive term in imaging. It can be further confirmed as benign or malignant only after finding a nodule. Therefore, the accurate detection of nodules is the first step in the diagnosis of lung cancer screening. According to reports, after a rigorously trained doctor with many years of clinical experience, an average of 200 CT scans is required to diagnose a case. The diagnosis time is more than 20 minutes. The computer nodule detection system can rapidly improve the diagnostic ability in a short period of time by learning a large number of samples labeled by experienced physicians, and assist the doctors of primary medical institutions to reduce misdiagnosis. It can be said that the use of science and technology to promote medical advancement has become urgent. Aliyun cloud machine vision scientist Hua Xiansheng claimed that Aliyun hopes to find a sophisticated algorithm for the diagnosis of early lung cancer through Tianchi, implant it in the ET medical brain, and finally provide recommendations to assist doctors, improve doctors' work efficiency, improve detection accuracy, and save more. More lives. Algorithm challenges cancer diagnosis The ET medical brain is Alibaba Cloud's AI platform for medical applications. With the medical brain storing medical knowledge base and medical experience, ET can have many medical capabilities and assume the role of doctor's assistant. Previously, the First Affiliated Hospital of Zhejiang University used ET to achieve rapid analysis of thyroid B ultrasound. ET can circle the nodule area on the film and give a judgment of benignity and malignancy, which greatly saves the doctor's diagnosis time and the accuracy rate is 21% higher than that of the top three hospitals. Lung cancer diagnosis is the next goal for the medical brain to "catch up." As a co-host of the competition, Intel Corporation provided Xeon PhiTM processors, Xeon® processors and other products and technologies that provide efficient computing support for deep learning, and Intel® for machine learning and deep learning. Math Kernel Library (Intel MKL) core algorithm library and Intel Python math library, and deep learning framework software specially designed for medical image analysis. The total prize pool in the first season of the contest totaled 1 million. Tianchi official website open registration on the same day, is expected to be completed on September 25 to resolve the championship team. The champion team can not only get bonuses, the algorithm is also expected to be applied in the actual diagnosis and treatment. The competition will also be composed of dozens of experts from the top three hospitals to form a jury to provide academic support for the event, ensure the quality of training data, and participate in the key match points in the race, providing professional medical and imaging guidance and evaluation. . In addition, Tianchi is also conducting the KDD Cup data mining competition in the same period. The KDD Cup is a global top data mining competition, and Tianchi has previously won exclusive rights to host the event.