罗桂波

职称:助理教授
电话:
办公室:
Email:luogb[at]pku dot edu dot cn
实验室网站:
研究方向:1、数据智能分析;2、多方隐私计算
职称 助理教授 电话
办公室 Email luogb[at]pku dot edu dot cn
研究方向 1、数据智能分析;2、多方隐私计算 实验室网站

导师与研究领域、方向:

北京大学信息工程学院助理教授/研究员、博士生导师。分别在华南理工大学获得学士学位,北京大学获得硕士和博士学位,博士毕业后在哈佛大学医学院/麻省总医院从事博士后研究工作。从事研究领域包括:AI for science, 特别是利用智能分析从多家医院数据中提取科学规律与量化指标,实现疾病定量诊疗;多方隐私计算,在多方参与计算场景下保障隐私不泄露并实现高性能模型训练。参与了多项NIH项目,积极推动智能算法赋能于临床医学的应用与发展,直接辅助医生更准确及高效率诊断疾病。以第一作者身份发表了包括IEEE T-PAMIRadiology: AI, IEEE TCSVTCVPRMICCAI等国际顶级期刊和会议,并担任IEEE Transactions on Image ProcessingIEEE Transactions on Broadcasting等国际期刊审稿人,多次受邀在放射学顶级大会RSNA做口头报告。

本实验室欢迎优秀的本科生/硕士生推免或报考硕士/博士研究生,同时也欢迎实习生和博士后申请者。

2024AI for science考研招生项目:

“基于图神经网络的智能电池材料创制研究”,合作导师:潘峰教授,网站链接:https://sam.pkusz.edu.cn/info/1017/1353.htm

讲授的课程:

“高级程序设计与实践”:该课程以应用实践为目标,紧密跟踪计算机科学、软件工程及人工智能的发展与需求变化。课程在系统地论述软件工程、编程语言和程序设计的基础上,重点讲述现代程序设计的并发处理、编程优化实践、代码编写安全方法、异常处理,以及软件开发的新技术和新工具等。最后讲述程序设计实践,包括跨语言跨平台编程、数据挖掘与科学统计、人工智能编程、深度学习模型的效率优化。

近期研究工作:

1. 医学影像分割大模型训练:“A Segmentation Foundation Model for Diverse-type Tumors

2. 基于联邦学习的医学影像分割大模型:“FedFMS: Exploring Federated Foundation Models for Medical Image Segmentation

3. 基于生成式人工智能的新型多方隐私计算范式:“MPCPA: Multi-Center Privacy Computing with Predictions Aggregation based on Denoising Diffusion Probabilistic Model

近年来取得的主要成果:

1. Ke Zhang, Chaoran Liu, Jielin Pan, Yunfei Zhu, Ximeng Li, Jing Zheng, Yingying Zhan, Wenjuan Li, Shaolin Li*, Guibo Luo*, Guobin Hong*. “Use of MRI-based deep learning radiomics to diagnose sacroiliitis related to axial spondyloarthritis”, European Journal of Radiology, European Journal of Radiology, 172, Feb. 2024.

2. Guibo Luo, Tianyu Liu, Jinghui Lu, Xin Chen, Lequan Yu, Jian Wu, Danny Z. Chen, Wenli Cai. “Influence of data distribution on federated learning performance in tumor segmentation”, Radiology: Artificial Intelligence. vol. 5, no. 3, e220082, May 2023.

3. Ke Zhang#, Guibo Luo#, Wenjuan Li#, Yunfei Zhu#, Jielin Pan, Ximeng Li, Chaoran Liu, Jianchao Liang, Yingying Zhan, Jing Zheng, Shaolin Li, Wenli Cai, Guobin Hong. “Automatic image segmentation and grading diagnosis of sacroiliitis associated with AS using a deep convolutional neural network on CT images”, Journal of Digital Imaging, vol 36, 2023: 2025-2034.

4. Guibo Luo, Tianyu Liu, Bin Li, Wenli Cai, “Deep-Cleansing: Deep-learning based electronic cleansing in dual-energy CT Colonography”, Medical Image Computing and Computer Assisted Intervention (MICCAI), 2021, pp. 43-53.

5. Wenli Cai, Tianyu Liu, Xing Xue, Guibo Luo, Xiaoli Wang, et al., “CT quantification and machine-learning models for assessment of disease severity and prognosis of COVID-19 patients”, Academic Radiology, vol. 27, no. 12, pp. 1665-1678, Dec. 2020.

6. Guibo Luo, Yuesheng Zhu, Zhenyu Weng, Zhaotian Li, “A disocclusion inpainting framework for depth-based view synthesis”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol 42, no. 6, pp. 1289-1302, Jun. 2020.

7. Guibo Luo, Yuesheng Zhu, “Foreground removal approach for hole filling in 3D video and FVV synthesis”, IEEE Transactions on Circuits and Systems for Video Technology, vol. 27, pp. 2118 - 2131, Oct, 2017.

8. Yifan Pan, Guibo Luo, Bairong Li, Yuesheng Zhu, “Enhanced unsupervised domain adaptation with dual-attention between classification and domain alignment”, IEEE International Conference on Acoustics, Speech, and Signal Processing, 2024.

会议报告

1. Guibo Luo, Zixuan Huang, Bin Li, Tianyu Liu, Mingyue Luo, Wenli Cai, “Deep convolutional neural networks for electronic cleansing in Non-cathartic CT Colonography”, RSNA 2022.

2. Guibo Luo, Wenli Cai, K. Ina, Ly, Eva Dombi, Brigitte Widemann, Justin T. Jordan, Tianyu Liu, Scott R. Plotkin, Gordon Harris, “Deep-NF: Deep convolutional neural networks for volumetric segmentation of plexiform neurofibromas on MRI”, CTF 2022.

3. Guibo Luo, Wenli Cai, Tianyu Liu, Xin Chen, Lequan Yu, “Federated vs centralized deep-learning models for liver and tumor segmentation in multi-center hepatic CT datasets”, RSNA 2021.

4. Guibo Luo, Wenli Cai, K. Ina Ly, Eva Dombi, Brigitte Widemann, Tianyu Liu, Gordon Harris, “3D random walk for volumetric tumor segmentation of plexiform neurofibromas on MRI”, CTF 2021.

专利:

1. “基于去噪扩散模型数据生成的多中心隐私计算方法及系统”,申请号:CN202410274548.9。

2. “生物体存活个数检测方法及装置”,专利号:ZL201410136763.9,授权。

3. “虚拟视点视频、图像的空洞填充方法、装置和终端”,申请号:PCT/CN2016/083746。

4. “一种自动对生物行为进行采集和跟踪以及分析的系统”, 申请号:CN202010552186.7。

对计划招收研究生的基本要求:

1. 专业范围:计算机科学技术、电子信息科学与技术;

2. 外语能力:英文六级或托福、雅思;

3. 具有独立思考能力和真诚可靠品质,积极主动,动手实践能力强。