Recent years have seen the rapid integration of artificial intelligence (AI) into the field of radiology. The most recent potential one is DeepSeek, which is redefining the benchmarks for open-weight, computational efficiency, and problem-solving capabilities.1 The emergence of DeepSeek has ignited enthusiasm as well as consternation, causing stock market volatility with its economic implications.2 More importantly, its technical capabilities have generated extensive discussions in the engineering and biomedical domains.3 Here, we briefly discuss what sets DeepSeek apart and why it could be the next tool in the AI transformation of radiology
למאמר המלא
Yifan Peng, Qingyu Chen, George Shih, DeepSeek is open-access and the next AI disrupter for radiology, Radiology Advances, Volume 2, Issue 1, January 2025, umaf009, https://doi.org/10.1093/radadv/umaf009