Liu Shi, female, PhD, Assistant Researcher at Nanchang University (tenured). She holds a Bachelor’s degree in Information and Computational Science from the Department of Mathematics at Chongqing University, and a PhD in Particle Physics and Nuclear Physics from the Institute of High Energy Physics, Chinese Academy of Sciences.

Research Direction: Primarily engaged in research on intelligent image processing methods for cone-beam CT. Specifically focuses on: sparse angle, low-dose, limited-angle CT reconstruction; image noise removal, metal artifact removal; image segmentation; computer tomography (CL); helical cone-beam CT scanning techniques; medical image post-processing; explainable AI/large models.

Application Fields: Intelligent medical auxiliary diagnosis, intelligent industrial non-destructive testing, etc.

Research Background: Computerized tomography plays a crucial role in both medical diagnostics and industrial inspection. Sparse angle, low-dose, limited-angle CT scanning methods help reduce patient dosage and improve sampling efficiency. Improving the reconstruction quality is currently a key research issue, as is removing noise and artifacts from images during practical application. Recently, combining artificial intelligence models with CT imaging, especially utilizing popular diffusion models and large model technologies, has become a trend in research.

Project Collaboration Units: Institute of High Energy Physics, Chinese Academy of Sciences; Imperial College London; City University of Hong Kong; The First Affiliated Hospital of Nanchang University, The Second Affiliated Hospital of Nanchang University, Stomatological Hospital of Nanchang University; Yifang Medical Technology Co., Ltd.; Jinan Zhongke Nuclear Research Institute and Ruiying Detection Technology Company (CAS team); China General Nuclear Begonia Technology Co., Ltd.; Beijing Lang Vision Instrument Co., Ltd.

Platforms Provided: Computing power (the group has purchased 4 x 4090 graphics cards, and 4 x ada5880 graphics cards for three-dimensional data processing); the supervisor possesses a solid research foundation and rich scientific experience, providing comprehensive guidance from idea formation to experimental design and article writing/revision.

Student Recruitment: The lab regularly recruits undergraduate interns and outstanding graduate students to conduct research. Research projects start from practical problems, aiming to help every student who loves exploration and dares to challenge unlock their potential in scientific work. The lab has good cooperative relationships with many top universities, leading hospitals in Jiangxi Province, and relevant companies, which can provide certain assistance for further studies and employment. Interested students are welcome to contact me (email: shiliu@ncu.edu.cn).

Courses Taught: "Digital Image Processing," "Machine Learning and Pattern Recognition."