By analyzing descriptions of magnetic resonance imaging, it can diagnose over a hundred diseases, with an average time of just 0.8 seconds to generate a diagnostic opinion for a case. 

It is "Doctor Xiao Jun", the world's first Chinese digital radiologist, which has recently made its debut in Beijing. 

The development of "Doctor Xiao Jun" is based on RadGPT, the world's first large language model (LLM) designed and built specifically for medical imaging diagnosis. RadGPT has been jointly launched by Professor Liu Yaou of the Radiology Department of Beijing Tiantan Hospital affiliated with Capital Medical University, and Professor Zhang Meihui and Professor Ye Chuyang of the Beijing Institute of Technology (BIT).

The creation of RadGPT is based on over one million pieces of high-quality medical imaging data and decades of accumulated professional imaging diagnostic experience. 

It has undergone a long-term artificial intelligence training by research teams led by Zhang Meihui and Ye Chuyang of BIT. 

The model is also based on Chinese natural language processing, adopting various large model fine-tuning methods tailored for medical data privacy protection, which effectively improves the model's performance on specific downstream tasks within the vertical domain.

The core function of RadGPT is to quickly generate diagnostic opinions by analyzing descriptions of MRI images. 

Based on this model, "Doctor Xiao Jun" can provide diagnostic opinions for over a hundred diseases including cerebrovascular diseases, tumors in over a dozen areas such as the brain, neck and chest, as well as infectious diseases. 

After verification of nearly a thousand cases, its accuracy rate exceeds 95% and the average time to generate a diagnostic opinion for a case is only 0.8 seconds. 

In addition, "Doctor Xiao Jun" can work around the clock every day. This means that patients' imaging diagnostic needs can be handled more promptly and the diagnostic pressure on busy radiologists can be relieved.

As a pioneering digital radiologist, the various functions of "Doctor Xiao Jun" are still being continuously improved, with the goal of mastering the interpretation and diagnosis of CT and X-ray images, as well as further developing various professional skills in radiology such as generating imaging diagnoses through images.

In the future, "Doctor Xiao Jun" will continue to learn, gradually growing into an excellent and tireless radiologist, alleviating the problems of radiologist shortages and heavy workloads, constantly improving diagnostic efficiency and accuracy, and providing patients with faster and higher-quality services.