迟剑宁  (副教授)

Trustworthy-AI for Reliable, Universial and Transformative Healthcare (TRUTH) 课题组

研究理念

SEEK THE TRUTH！

求真、务实！

研究目标

TRUTH课题组长期围绕“智慧医疗的可信性”这一核心问题和行业重大需求，以医学影像、临床数据、组学数据、穿刺手术机器人、导丝手术机器人、微纳手术机器人等为主要研究对象，开展生物医学信息的可信性智能分析、手术机器人的可信性智能导航研究。具体目标包括：

（1）生物医学信息的可信性智能分析：致力于开发生物医学信息的可信性智能分析技术，通过深度学习、知识图谱方法，系统评估多源异构信息（如文献、临床数据、组学数据）的可靠性、时效性、一致性及证据等级，构建动态可信度量化模型，为精准医疗决策、疾病风险预测提供可解释、可溯源的智能验证支持。

（2）手术机器人的可信性智能导航：聚焦手术机器人可信性智能导航，通过多模态感知融合与强化学习框架，构建实时动态空间映射模型，实现亚毫米级位姿纠偏与组织形变补偿，建立涵盖操作安全性、路径合规性、决策可解释性的多维度可信评估体系，为介入手术提供高精度抗扰导航，保障人机协同操作的安全与效率。

研究方向

（1）医学影像质量提升：医学影像去噪、超分辨率、质量增强

（2）医学影像病灶分割：医学影像单目标分割、多语义分割、多实例分割，医学影像半监督、弱监督分割

（3）计算机辅助诊断疗效预测：医学辅助诊断大模型、多模态信息融合诊断、多任务联合诊断、治疗手段疗效预测、预后生存期预测

（4）手术机器人环境感知：血管三维重建、腔体三维重建、器官三维重建，内窥镜图像分割与三维重建

（5）手术机器人运动规划：手术机器人人体内任务规划、路径规划、轨迹规划

课题组近年来研究成果：

1. 医学影像多信息集成学习

[1] Pengfei Lyu, Xiaosheng Yu, Jianning Chi, Hao Wu, Chengdong Wu, Jagath C. Rajapakse. TwinsTNet: Broad-View Twins Transformer Network for Bi-Modal Salient Object Detection. IEEE Transactions on Image Processing, 2025, 34: 2796-2810. (中科院1区top期刊，IF: 10.8)

[2] Jianning Chi*, Jia-hui Chen, Bo Wu, Jin Zhao, Kai Wang, Xiaosheng Yu, Wenjun Zhang, and Ying Huang. A Dual-Branch Cross-Modality-Attention Network for Thyroid Nodule Diagnosis Based on Ultrasound Images and Contrast-Enhanced Ultrasound Videos. IEEE Journal of Biomedical and Health Informatics, 2025, 29(2): 1269-1282. (中科院1区top期刊，IF: 6.7)

[3] Jianning Chi*, Zhiyi Sun, Xiaoying Han, et al. PILN: A Posterior Information Learning Network for Blind Reconstruction of Lung CT Images, Computer Methods and Programs in Biomedicine, 2023, 232: 107449. (中科院2区top期刊，IF: 6.1)

[4] Jianning Chi*, Huixuan Wu, Yujin Shi, Zelan Li, Xiaohu Sun, Lu Wang, Zitian Zhang, Yuehua Gong. Clustering Multimodal Ensemble Learning for Predicting Gastric Cancer Neoadjuvant Chemotherapy Efficacy, 2024 IEEE International Conference on Bioinformatics and Biomedicine, 2024: 5257-5263. (生物信息学领域国际顶级会议CCF-B)

[5] Junjie Shi, Puhong Duan, Xiaoguang Ma, Jianning Chi, Yong Dai. Frefusion: Frequency Domain Transformer for Infrared and Visible Image Fusion, IEEE Transactions on Multimedia, 2025, Early Access. (中科院1区top期刊，IF: 8.4)

[6] Jianning Chi*, et al. Thyroid Nodule Classification in Ultrasound Images by Fine-Tuning Deep Convolutional Neural Network, Journal of Digital Imaging, 2017, 30(4): 477-486. (中科院2区期刊，IF: 4.4)

[7] Jianning Chi*, Chengdong Wu, Xiaosheng Yu, Hao Chu, Peng Ji. Saliency Detection via Integrating Deep Learning Architecture and Low-level Features, Neurocomputing, 2019, 352: 75-92. (中科院2区top期刊，IF: 6.0)

[8] Jianning Chi*, Jin Zhao, Siqi Wang, Xiaosheng Yu, Chengdong Wu. LGDNet: Local Feature Coupling Global Representations Network for Pulmonary Nodules Detection, Medical and Biological Engineering and Computing, 2024, online. (中科院2区期刊，IF: 3.2)

[9] Yang Jiang, Shuang Zhang, Jianning Chi*. Multi-Modal Brain Tumor Data Completion based on Reconstruction Consistency Loss, Journal of Digital Imaging, 2023, 36: 1794-1807. (中科院2区期刊，IF: 4.4)

[10] 迟剑宁*, 于晓升, 张艺菲. 融合深度网络和浅层纹理特征的甲状腺结节癌变超声图像诊断，中国图象图形学报, 2018, 23(10): 1582-1593. (中文核心期刊)

    基于双分支交叉模态注意力的甲状腺结节诊断方法          基于噪声分类预估指导的医学影像超分辨率重建            基于影像与文本特征聚类集成学习的医学影像分类

2. 医学影像噪声与组织信号解耦分析

[1] Jianning Chi*, Zhiyi Sun, Liuyi Meng, Siqi Wang, Xiaosheng Yu, Xiaolin Wei, and Bin Yang. Low-dose CT image super-resolution with noise suppression based on prior degradation estimator and self-guidance mechanism. IEEE Transactions on Medical Imaging, 2025, 44(2): 601-617.(中科院1区top期刊，IF: 8.9)

[2] Jianning Chi*, Zhiyi Sun, Huan Wang, Pengfei Lyu, Xiaosheng Yu, Chengdong Wu. CT Image Super-Resolution Reconstruction based on Global Hybrid Attention, Computers in Biology and Medicine, 2022, 150: 106112. (中科院1区top期刊，IF: 7.7)

[3] Jianning Chi*, Mark Eramian. Enhancement of Textural Differences Based on Morphological Component Analysis, IEEE Transactions on Image Processing, 2015, 24(9): 2671-2684. (中科院1区top期刊，IF: 10.6)

[4] Jianning Chi*, Zhiyi Sun, Tianli Zhao, Huan Wang, et al. Low-Dose CT Image Super-Resolution Network with Dual-Guidance Feature Distillation and Dual-Path Content Communication, Medical Image Computing and Computer Assisted Intervention – MICCAI 2023, Lecture Notes in Computer Science, 2023, 14229: 99-108. (国际顶级医学影像计算会议,CCF-B)

[5] Jianning Chi*, Zhiyi Sun, et al. A Hybrid Framework of Dual-Domain Signal Restoration and Multi-depth Feature Reinforcement for Low-Dose Lung CT Denoising, Journal of Digital Imaging, 2024, online. (中科院2区期刊，IF: 4.4)

[6] Jianning Chi*, Jian Miao, Jia-hui Chen, Huan Wang, Xiaosheng Yu, Ying Huang. DSTAN: A Deformable Spatial-Temporal Attention Network with Bidirectional Sequence Feature Refinement for Speckle Noise Removal in Thyroid Ultrasound Video, Journal of Digital Imaging, 2024. (中科院2区期刊，IF: 4.4)

[7] Jianning Chi, Xiaolin Wei, Zhiyi Sun*, Yongming Yang, Bin Yang. Low-Dose CT Image Super-Resolution Network with Noise Inhibition Based on Feedback Feature Distillation Mechanism, Journal of Digital Imaging, 2024, online (中科院2区期刊，IF: 4.4)

[8] Jianning Chi*, Mark Eramian. Enhancing Textural Differences using Wavelet-based Texture Characteristics Morphological Component Analysis: A Preprocessing Method for Improving Image Segmentation, Computer Vision and Image Understanding, 2017, 158: 49-61. (中科院3区期刊，IF: 4.5)

[9] Jianning Chi*, Mark Eramian. Wavelet-based Texture-characteristic Morphological Component Analysis for Colour Image Enhancement, 2016 IEEE International Conference on Image Processing, 2016.9.25-2016.9.28, Phoenix, AZ, USA. (国际顶级图像处理与计算机视觉会议，CCF-C)

[10] Huan Wang, Jianning Chi, Chengdong Wu, et al. Degradation Adaption Local-to-Global Transformer for Low-Dose CT Image Denoising, Journal of Digital Imaging, 2023, 36: 1894–1909. (中科院2区期刊，IF: 4.4)

    基于空间-时序注意力的乘性噪声与组织信号解耦      基于先验退化估计和自引导机制的噪声抑制与组织器官超分辨率            基于形态学成分分析的图像增强

3. 医学影像病灶与背景区域分割

[1] Jianning Chi*, Geng Lin, Zelan Li, Wenjun Zhang, Jia-hui Chen, Ying Huang. Coarse for Fine: Bounding Box Supervised Thyroid Ultrasound Image Segmentation Using Spatial Arrangement and Hierarchical Prediction Consistency, IEEE Journal of Biomedical and Health Informatics, 2025, early access.(中科院1区top期刊，IF: 6.7)

[2] Jianning Chi*, Zelan Li, Huixuan Wu, Wenjun Zhang, and Ying Huang. Beyond Point Annotation: A Weakly Supervised Network Guided by Multi-Level Labels Generated from Four-Point Annotation for Thyroid Nodule Segmentation in Ultrasound Image, 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, Hyderabad, India, 2025: 1-5.(国际信号处理顶级会议，CCF-B)

[3]Jianning Chi, Mingyang Sun, Zelan Li, Geng Lin, Ying Huang. Adaptive box-level supervision with superpixel shape guidance for ultrasound image segmentation, Visual Computer, 2025, early access.(中科院3区，IF： 3.0)

[4] Jianning Chi*, Zelan Li, Zhiyi Sun, Xiaosheng Yu, Huan Wang. Hybrid Transformer UNet for Thyroid Segmentation from Ultrasound Scans, Computers in Biology and Medicine, 2023, 153: 106453.(中科院1区top期刊，IF: 7.7)

[5] Jianning Chi*, Xiaoying Han, Chengdong Wu, Huan Wang, Peng Ji. X-Net: Multi-branch UNet-like Network for Liver and Tumor Segmentation from 3D Abdominal CT Scans, Neurocomputing, 2021, 459: 81-96.(中科院2区top期刊，IF: 6.0)

[6] Jianning Chi*, Shuang Zhang, Xiaoying Han, Huan Wang, Chengdong Wu, Xiaosheng Yu. MID-UNet: Multi-input Directional UNet for COVID-19 Lung Infection Segmentation from CT images, Signal Processing: Image Communication, 2022, 108: 116835. (中科院3区期刊，IF: 3.5)

[7] Xiaoliang Lei, Xiaosheng Yu, Jianning Chi, Ying Wang, Jingsi Zhang, Chengdong Wu. Brain Tumor Segmentation in MR Images Using a Sparse Constrained Level Set Algorithm, Expert Systems with Applications, 2021, 168: 114262.(中科院1区top期刊，IF: 8.5)

[8] Siqi Wang, Xiaosheng Yu, Wenzhuo Jia, Jianning Chi, et al. Optic Disc Detection based on Fully Convolutional Network and Weighted Matrix Recovery Model, Medical and Biological Engineering and Computing, 2023, 61: 3319-3333.(中科院2区期刊，IF: 3.2)

[9] 雷晓亮,于晓升,迟剑宁,王莹,吴成东.稀疏形状先验的脑肿瘤图像分割,中国图象图形学报, 2019, 24(12): 2222-2232.(中文核心期刊)

[10] 王莹,于晓升,迟剑宁,雷晓亮,吴成东.双层水平集描述眼底图像视杯视盘分割,中国图象图形学报, 2020, 25(6): 1260-1270.(中文核心期刊)

          基于空间分布与语义感知合理的医学影像弱监督分割         基于多层次信息提取与多尺度约束的医学影像弱监督分割          基于空间特征时序关联的甲状腺区域分割

4. 基于视觉的机器人规划

[1] Suyi Liu, Fang Xu, Chengdong Wu, Jianning Chi, Xiaosheng Yu, Longxing Wei, Chuanjiang Leng. CMT-6D: a lightweight iterative 6DoF pose estimation network based on cross-modal Transformer, Visual Computer, 2025, 41(3): 2011-2027.(中科院3区， IF： 3)

[2] Suyi Liu, Jianning Chi, Chengdong Wu, Fang Xu, Xiaosheng Yu. SGT-Net: A Transformer-Based Stratified Graph Convolutional Network for 3D Point Cloud Semantic Segmentation, CMC-Computers Materials and Continua, 2024, 79(3): 4471-4489.(中科院3区，IF：2.1) 

课题组所获专利及软件著作权：

1. 专利

[1] 吴成东, 张艺菲, 迟剑宁. 一种基于深度学习网络和浅层纹理特征融合的甲状腺超声图像结节分析方法. 中国专利, CN110211116A, 2019.9.6.

[2] 吴成东, 张子昂, 迟剑宁. 一种基于改进的卷积神经网络的医学图像分割方法. 中国专利, CN110570431A, 2020.1.7.

2.软件著作权

[1] 迟剑宁, 吴博, 吴慧璇, 林庚, 凌修伟. 甲状腺多模态影像自监督软件v1.0. 计算机软件著作, 2025SR0182374, 2025.1.26.

[2] 迟剑宁, 林庚, 吴慧璇, 凌修伟, 吴博. 基于边界框监督的甲状腺超声图像弱监督分割软件v1.0. 计算机软件著作， 2025SR0182790, 2025.1.26.

[3] 迟剑宁, 吴慧璇, 林庚, 凌修伟, 吴博. 基于双分支跨模态注意网络的甲状腺超声图像结节诊断软件v1.0. 计算机软件著作, 2025SR0188545, 2025.1.27.

[4] 迟剑宁, 凌修伟, 吴慧璇, 林庚, 吴博. 基于扩散模型与CNN特征融合的子宫腺肌症诊断软件v1.0. 计算机软件著作, 2025SR0281147, 2025.2.18.