A research team from the Chinese University of Hong Kong has made a major breakthrough and successfully developed a new artificial intelligence model called VisionFM. This model focuses on the diagnosis and prediction of eye diseases, and is particularly outstanding in predicting the progression of glaucoma. Its accuracy even surpasses the previously leading RETFound model. This result has been published in the authoritative journal "NEJM AI", marking another milestone in the field of artificial intelligence in medical image analysis. The emergence of VisionFM is expected to significantly improve the efficiency and accuracy of diagnosis of ophthalmic diseases and bring a better medical experience to patients.
A research team at the Chinese University of Hong Kong (CUHK) has successfully developed a new basic artificial intelligence model "VisionFM" that focuses on the diagnosis and prediction of eye diseases, especially in disease screening and glaucoma progression prediction. Relevant research was published in the journal NEJM AI last month.
Picture source note: The picture is generated by AI, and the picture authorization service provider Midjourney
VisionFM’s capabilities are impressive. Research shows that the model's accuracy in diagnosing 12 eye diseases is not only comparable to that of mid-level ophthalmologists, but even performs better in some tasks, especially in predicting the progression of glaucoma, surpassing the previous basic model RETFound in the field of ophthalmology. . The research team believes that with the increase in the amount of training data, the application scenarios of VisionFM will be further expanded, and it may play a key role in more ophthalmic clinical tasks in the future.
The launch of this breakthrough technology comes at a time when the application of generative artificial intelligence in the medical field is accelerating. More and more researchers and medical experts are beginning to explore how AI technology can improve medical services. From surgical planning to diagnostic support, generative AI is reshaping medical practice.
For example, the Hong Kong Center for Artificial Intelligence and Robotics (CAIR) launched an AI model developed based on the Meta Platforms open source model Llama2 earlier this year, which has been put into use at the First Affiliated Hospital of Sun Yat-sen University and the Prince of Wales Hospital in Hong Kong to assist surgical operations. Planning and diagnostic report generation.
With the emergence of advanced technologies such as VisionFM, the potential of artificial intelligence in the medical field is gradually unfolding. This not only helps improve diagnostic accuracy, but also provides a new way to optimize medical services and improve patient experience.
The successful application of VisionFM heralds the broad prospects of artificial intelligence in the medical field. In the future, as technology continues to advance and data accumulation increases, AI models like VisionFM will provide strong support for more accurate disease diagnosis and more effective medical services, benefiting more patients. Its breakthroughs in the field of ophthalmic disease diagnosis also provide valuable experience and reference for other medical image analysis fields.