Early Detection of Lung Cancer Assisted by AI


It feels like artificial intelligence (AI) is becoming part of everyday conversation now. In many ways it’s the overnight success story that was years in the making. Use cases in deep learning and big data are quickly delivering breakthrough results for both consumers and businesses. With AI, society has the opportunity to address everything from scarce resource utilization and scientific exploration to city management. We’ll see unprecedented developments in medicine; fully autonomous cars and machines; and better and safer disaster response. And in the medical field in particular, AI use cases have the potential to save lives.

This week Intel, along with Alibaba Cloud*, the cloud computing arm of Alibaba Group and LinkDoc*, announced an AI competition that is poised to do just that. The Tianchi Healthcare AI Competition will be focused on medical imaging, an industry set to be transformed by AI. The competition seeks to assist doctors with the successful detection of lung cancer using AI technology to analyze medical images and screen for pulmonary nodules. LinkDoc will provide a large amount of high resolution CT scan data to participants through the Tianchi platform based on Alibaba Cloud running on Intel® Xeon® and Intel® Xeon Phi™ processors. Participants will train models using the CT image data to get the characteristics of cancer nodules, and then screen the images for nodules using their trained AI model.

Lung cancer is one of the most prevalent forms of cancer in China, with 4.3M new patients and more than 2.8M deaths in 2015 alone. The goal of the Tianchi Healthcare AI Competition is to eventually find a good algorithm to assist with the early detection of lung cancer, relieving doctor’s burdens, and enabling better treatment options for the millions affected.

As a key partner in this competition, Intel is contributing to Alibaba Cloud’s foundational infrastructure resources optimized to run AI workloads at higher performance, efficiency and scalability. Intel® Xeon and Intel Xeon® PhiTM processors provide efficient computing support for deep learning. The Intel® Math Kernel Library core algorithm and Intel® Python mathematic libraries target machine learning and deep learning applications. Intel is also providing deep learning framework software specifically designed for medical image analysis.

More than 500 participants from enterprises, universities, and academia have applied to compete for the best AI lung screening assistant. In addition to a one million RMB (approximately 145,000 US dollars) grand prize for the winner, the winning team’s algorithm is expected to be applied in the actual diagnosis and treatment of lung cancer. Final results of this first of three competitions will be announced in September of this year, with additional competitions to follow in 2018 and 2019.

Learn more or get involved with the Tianchi Healthcare AI Competition here. You can also learn more about Intel’s AI technology at our Website as well as by engaging with us on Twitter.


*Other names and brands are the property of their owners and are used with permission.