On May 25, the doctoral dissertation defense was held in Room 407 of the Main Building at Tsinghua University. PhD students Yifan Chen, Yiliang Zhou, Yeyi Cai, and Haoxiang Wang successfully completed their dissertation defenses. Faculty members and students from the laboratory gathered to witness this important moment. The four students carried out in-depth research in computational imaging, neuroscience, and intelligent simulation, demonstrating strong research ability and innovative achievements.

The defense committee was chaired by Professor Yongdong Zhang from the University of Science and Technology of China. Committee members included Professor Qionghai Dai, Professor Jianhua Tao, Professor Jia Liu, Associate Professor Jiamin Wu, Associate Professor Hui Qiao from Tsinghua University, and Professor Yue Deng from Beihang University. The defense attracted many faculty members and students to attend and exchange ideas.

PhD candidates with Professer Qionghai Dai and Associate Professor Jiamin Wu

• Research Highlights

Yifan Chen | High-Precision Light Field Sensing and Display

Yifan Chen’s doctoral dissertation is titled “Research on High-Precision Light Field Sensing and Display Methods Addressing the Spatial Bandwidth Product Bottleneck.” His research focused on key challenges in high-precision 3D imaging and glasses-free 3D display technologies.

Main contributions include:

1. Proposing a multi-angle corrected scanning light-field microscopy method for high-resolution 3D imaging of large transparent samples;

2. Developing a super-Snell scanning light-field display system with high resolution and a viewing angle expanded to 150°;

3. Establishing a high-precision light-field simulation and visual optimization framework to improve display quality and visual experience.

Yifan Chen with the dissertation committee

Yiliang Zhou | High-Throughput Cross-Modal Two-Photon Computational Microscopy

Yiliang Zhou’s doctoral dissertation is titled “Research on High-Throughput Cross-modal Two-photon Computational Microscopy.” His work focused on overcoming throughput and modality limitations in deep intravital two-photon microscopy.

Main contributions include:

1. Proposing a high-throughput two-photon synthetic aperture microscopy method with long-term autofocusing capability;

2. Developing a self-supervised fluorescence lifetime estimation method based on photon-event sensing, reducing the required photon dose by three-orders of magnitude;

3. Establishing a cross-modal imaging framework combining fluorescence intensity and fluorescence lifetime information for biological applications such as tumor microenvironment analysis.

Yiliang Zhou with the dissertation committee

Yeyi Cai | Dynamic Encoding of Functional Modules in Large-scale Cortical Neurons

Yeyi Cai’s doctoral dissertation is titled “Dynamic Encoding of Functional Modules in Large-scale Cortical Neurons.” Her research focused on dynamic neural network analysis, functional module identification, and information interaction mechanisms in cortical systems.

Main contributions include:

1．Building an experimental platform for deep brain stimulation and cortical recording to study neural dynamic encoding;

2.Proposing the FRID method for efficient and accurate identification of functional modules from mesoscale calcium imaging data;

3.Revealing long-term interactions across neural modules during cognitive tasks through synchronized imaging experiments and network modeling.

Yeyi Cai with the dissertation committee

Haoxiang Wang | Differentiable Dynamic Field Simulation and Application with Neural Field Representation

Haoxiang Wang’s doctoral dissertation is titled “Differentiable Dynamic Field Simulation and Application with Neural Field Representation.” His work focused on neural-field-based physical simulation and high-fidelity data generation.

Main contributions include:

1.Developing a neural-field physical simulation method based on multi-scale hash grids for efficient and accurate simulation;

2.Proposing a physics-driven implicit neural field framework for stable fluid simulation on complex surfaces;

3.Building a high-fidelity data generation framework based on the lattice Boltzmann method for realistic fluid simulation and real-time smoke rendering.

Haoxiang Wang with the dissertation committee

• Closing Remarks

During the defense, the four PhD students presented their research work and answered questions from the committee members. The committee highly recognized the originality, theoretical significance, and application value of the dissertations.

After anonymous voting by the committee, all four students successfully passed their doctoral dissertation defenses. The defense session concluded successfully.

Congratulations to all four PhD graduates! We wish them continued success in their future research and careers.