Recently, the DAMO Academy of Alibaba and Zhejiang Cancer Hospital collaborated to release the world's first AI model for early gastric cancer imaging screening - DAMO GRAPE. The launch of this model marks a major breakthrough in medical imaging technology. It can effectively identify early gastric cancer lesions using plain CT images, providing new possibilities for early diagnosis.
Image source note: The image is AI-generated, and the image licensing service provider is Midjourney.
According to data from the National Cancer Center, there are as many as 360,000 new cases of gastric cancer each year in China, with more than 260,000 deaths, making gastric cancer the third leading cause of cancer-related deaths. If detected and intervened early, patients' five-year survival rate can be significantly increased from 30% to over 90%.
The research team behind the DAMO GRAPE model conducted a large-scale clinical study involving nearly 100,000 people through cooperation with 20 hospitals across the country, significantly improving the detection rate of gastric cancer. The research results were published in the top international journal "Nature Medicine," marking the widespread recognition of the scientific validity and effectiveness of this technology.
Traditionally, plain CT scans were considered unsuitable for examining hollow organs such as the gastrointestinal tract. However, this team successfully overcame challenges such as variations in stomach morphology and the identification of early lesions by building the largest international dataset of plain CT images of gastric cancer. Research data show that the sensitivity and specificity of the DAMO GRAPE model reached 85.1% and 96.8%, respectively, surpassing the performance of radiologists by 21.8% in sensitivity and 14.0% in specificity. This makes the combination of plain CT and AI an effective method for early gastric cancer screening.
In simulated screening of high-risk populations, the model's gastric cancer detection rates reached 24.5% and 17.7%, with about four out of ten being asymptomatic patients. This achievement not only provides strong support for early gastric cancer screening but also opens up new paths for multi-cancer screening in the future.
The release of the DAMO GRAPE model represents a significant advancement in the medical field, offering a new solution to the challenges of gastric cancer prevention and treatment. Studies show that using AI technology for initial screening of abdominal CT combined with gastroscopy will effectively improve the detection rate of early gastric cancer.
Key points:
- 🎯 Alibaba and Zhejiang Cancer Hospital jointly launched the world's first AI model for gastric cancer imaging screening, DAMO GRAPE.
- 📈 In a study involving nearly 100,000 people, the model significantly improved the detection rate of gastric cancer, with sensitivity and specificity reaching 85.1% and 96.8%, respectively.
- 🏥 The application of the model has made the combination of plain CT and AI technology an effective means of early gastric cancer screening, promoting further development of medical imaging technology.