基本信息

姓名：余程嵘

职称：助理研究员

系所：人工智能系

研究领域：智能医学，医学图像分析，智能放疗

电子邮箱：yuchengrong@ynu.edu.cn

学术主页：https://scholar.google.com/citations?user=SG5R43EAAAAJ

个人主页：https://ychengrong.github.io/

个人简介

余程嵘，中共党员，工学博士，英国365集团官网助理研究员。中国人工智能学会（CAAI）会员，中国计算机学会（CCF）会员，四川省人工智能学会会员。2024年博士毕业自四川大学，获计算机科学与技术博士学位，师从IEEE Fellow章毅教授。主要研究方向为智能医学，包括多模态医学图像分析、放疗计划智能设计等。先后参与主研科技部科技创新2030-“新一代人工智能”重大项目，国家基金委面上项目。近五年，先后发表论文10余篇，包括《Knowledge-Based System》，《Neurocomputing》以及《International Journal of Neural Systems》等期刊，授权发明专利5项，并担任CVPR，NeurIPS等多个国际顶级会议审稿人。

教育经历

2019/09-2024/12，四川大学，计算机科学与技术，博士

2016/09-2019/06，昆明理工大学，计算机系统结构，硕士

2012/09-2016/06，南昌大学，软件工程，本科

工作经历

2025.09 —至今,云南大学,英国365集团官网,助理研究员

代表性学术论著

[1] Chengrong Yu, Junjie Hu, Guangjun Li, Shenqian Zhu, Sen Bai, Zhang Yi. Segmentation for regions of interest in radiotherapy by self-supervised learning. Knowledge-Based Systems[J], 2022, 256: 109370.

[2] Chengrong Yu, Ying Song, Qiang Wang, Shengqian Zhu, Zhang Yi, Junjie Hu. Leveraging denoising diffusion probabilistic model to improve the multi-thickness CT segmentation. Neurocomputing[J], 2024, 128573.

[3] Chengrong Yu, Ying Song, Qiang Wang, Zhonglian Wei, Zhang Yi, Guangjun Li, Junjie Hu. Enhancing Exploration and Exploitation in Radiotherapy Treatment Planning through Action-guided Deep Reinforcement Learning. International Journal of Neural Systems[J], 2026,36.07: 2650020.

[4] Junjie Hu, Chengrong Yu, Zhang Yi, Haixian Zhang. Enhancing Robustness of Medical Image Segmentation Model with Neural Memory Ordinary Differential Equation. International Journal of Neural Systems[J], 2023, 33.12: 2350060:1-2350060:13.

[5] Shengqian Zhu, Chengrong Yu, Junjie Hu. Regularizing deep neural networks for medical image analysis with augmented batch normalization. Applied Soft Computing[J], 2024, 154: 111337.

[6] Shengqian Zhu, Chengrong Yu, Zhang Yi, Junjie Hu. Visual prompt-driven universal model for medical image segmentation in radiotherapy.Knowledge-Based Systems[J], 2025, 326:114006.

[7] Shengqian Zhu, Chengrong Yu, Wenbo Qi, Jiafei Wu, Ying Song, Guangjun Li, Zhang Yi, Xiaogang Xu, Junjie Hu PRIME: Prototype-Driven Class Incremental Learning for Medical Image Segmentation. Proceedings of the 33rd ACM International Conference on Multimedia(ACMM), 2025. p. 4688-4697.

[8] Shengqian Zhu, Chengrong Yu, Qiang Wang, Ying Song, Guangjun Li, Jiafei Wu, Xiaogang Xu, Zhang Yi, Junjie Hu.Class incremental medical image segmentation via prototype-guided calibration and dual-aligned distillation. Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 40. No. 16. 2026.

[9] Junjie Hu, Chengrong Yu, Shengqian Zhu, Haixian Zhang. Incorporating Adaptive Sparse Graph Convolutional Neural Networks for Segmentation of Organs at Risk in Radiotherapy, International Journal of Intelligent Systems[J], 2024, 2024.1: 1728801.

[10] Yuan Yao, Chengrong Yu, Junjie Hu, Yan Zhang. Enhanced multi-objective evolutionary framework for fluence map optimization in intensity-modulated radiation therapy. 2023 International Annual Conference on Complex Systems and Intelligent Science (CSIS-IAC). IEEE, 2023. p. 576-582.

[11] Xiangjie Tan, Ying Song, Qiang Wang, Chengrong Yu, Zhang Yi, Junjie Hu. Cascaded neural memory ODEs for predicting fluence maps in rectal cancer IMRT. Pattern Recognition[J], 2026, 171: 112301.

[12] Ying Song, Junjie Hu, Qiang Wang, Chengrong Yu, Jiachong Su, Lin Chen, Xiaorui Jiang, Bo Chen, Lei Zhang, Qian Yu, Ping Li, Feng Wang, Sen Bai, Yong Luo, Zhang Yi. Young oncologists benefit more than experts from deep learning-based organs-at-risk contouring modeling in nasopharyngeal carcinoma radiotherapy: A multi-institution clinical study exploring working experience and institute group style factor. Clinical and Translational Radiation Oncology, 2023, 41: 100635.

[13] 余程嵘,王威,戴伟,邓辉;王锋;卫守林. 基于Docker的射电干涉阵软件系统敏捷封装与部署[J].天文研究与技术,2019,16(01):123-130.

部分专利

[1] 章毅; 柏森; 余程嵘; 宋莹; 胡俊杰; 王强; 张海仙; 郭际香; 郭泉。一种基于自监督学习的放疗靶区自动分割方法（已授权，专利号:CN202110274005.3）

[2] 章毅; 祝生乾; 胡俊杰; 余程嵘; 李贵元。一种面向复杂场景的多计算模型管理方法（已授权，专利号:CN202210221092.0）

[3] 柏森; 章毅; 胡俊杰; 宋莹; 余程嵘。基于深度神经网络的鼻咽癌放疗靶区自动分割方法（已授权，专利号:CN202011059919.X）

[4] 章毅; 柏森; 胡俊杰; 宋莹; 王强; 余程嵘。基于深度神经网络的放疗剂量自动预测方法（已授权，专利号:CN202110207866.X）

主要科研项目

[1] 新一代认知神经网络模型, 国家科技部重大专项，2019.7-2023.7，参与.

受邀担任审稿人

CVPR

NeurIPS