DAMO Academy's Medical AI: Selected in International AI Report, Rated as "Highlighted Research" of the Year

On April 17, it was reported that Stanford University recently released the "AI Index Report 2024". Thanks to its innovative breakthrough in large-scale early screening of pancreatic cancer with AI, the medical AI of Alibaba DAMO Academy was selected as a highlighted research in the field of science and healthcare. This is also the only AI related highlighted research from a Chinese tech company in the report. Currently, this technology has been piloted in a public-welfare project in Lishui, Zhejiang, assisting doctors in screening for pancreatic cancer and other diseases.

Launched by Stanford University in 2017, the AI Index Report aims to comprehensively study and evaluate the AI industry and is one of the most authoritative research reports in the field. The report states that in the past year, AI has been advancing at an accelerated pace. Models such as GPT-4, Gemini, and Claude 3 have demonstrated impressive multimodal capabilities, often outperforming humans in benchmark performances in areas like image classification and visual reasoning. AI has also driven new discoveries in science and healthcare. Research shows that doctors can use AI to better diagnose breast cancer, interpret X-rays, and screen for fatal cancers. Therefore, this year, the report added a special chapter discussing the impact of AI on science and medicine, and included DAMO Academy's pancreatic cancer detection model, PANDA, in the highlighted research of the year.

It is known that in November 2023, the medical AI team of DAMO Academy, in collaboration with multiple global medical institutions, released the pancreatic cancer detection model PANDA. The related paper was published in the top-tier medical journal Nature Medicine. Relying on the method of "non-contrast CT + AI", PANDA achieved large-scale early screening of pancreatic cancer. While improving the detection rate, it does not impose additional radiation or economic burden on patients. In a retrospective trial of more than 20,000 real-world cases, the model detected 31 clinically missed lesions, and two early-stage pancreatic cancer patients have been cured through surgery.

Professor Jörg Kleeff, a German pancreatic cancer diagnosis and treatment expert and one of the reviewers of the paper, commented that "the results are quite remarkable and may significantly change our approach to pancreatic cancer screening." Professor Eric Topol from the Scripps Research Institute in the United States also highly praised it as "an impressive study." Nature Medicine also published a special commentary article stating: "Cancer screening based on medical imaging AI is about to enter the golden age."

In February 2024, relying on Alibaba's public-welfare project for early multi-cancer screening with medical AI, this model was first deployed at Lishui Central Hospital and Jingning County People's Hospital. It assists doctors in screening and diagnosing various diseases through the "non-contrast CT + AI" approach, and is expected to be promoted to more regions with unbalanced medical resources.

As introduced, DAMO Academy, as a research institute under Alibaba Group that focuses on basic science, innovative technologies, and applied technologies, focuses on frontier fields such as artificial intelligence and integrated circuits, helping to solve global social problems and improve human well-being. In the medical field, DAMO Academy is collaborating with multiple global medical institutions to explore the use of AI technology to achieve "8 + 5" multi-disease screening with non-contrast CT, covering eight fatal cancers such as pancreatic cancer, esophageal cancer, lung cancer, breast cancer, liver cancer, and gastric cancer, as well as five chronic diseases such as osteoporosis and cardiovascular diseases.