• Intelligent Imaging and Image Analysis: Investigating high-performance, interpretable, and self-supervised image analysis methods. Exploring new architectures that integrate imaging with AI to push the boundaries of scientific observation and promote new scientific discoveries.

• Foundation Models for Scientific Observation: Building multi-scale, cross-modality, and high-quality image datasets. Developing foundation models for frontier research to achieve accurate representation and mechanistic understanding of complex phenomena.

• Advanced Imaging Mechanisms and Systems: Combining imaging with robotics for automatic perception and active observation. Exploring new paradigms that integrate quantum physics with optical imaging, aiming to break the fundamental limits of classical systems.

• We have developed a series of intelligent methods to push the boundaries of scientific observation, providing new methods and tools for frontier research. These approaches have been widely adopted by leading institutions worldwide, including Yale University, the Max Planck Institute, and Boston University. Our work has significantly advanced the integration of AI and scientific observation, offering critical support for important discoveries in neuroscience, immunology, medicine, and related fields.

• Our research has been published in high-impact journals such as Nature Methods, Nature Biotechnology, Nature Computational Science, etc. Our team was invited to publish a perspective commentary in Nature Methods. Our work has also been featured in major media outlets, including People’s Daily (Overseas Edition) and Guangming Daily.

• Xinyang has been invited as a keynote or guest speaker at various academic conferences, including WAIC 2025, ACAIC 2025, and CICAI 2023. He was awarded the PhotoniX Prize in 2025 in recognition of his contributions in empowering scientific observation with AI to accelerate scientific discovery.

[1] Yixin Li, Qi Zhang, Yuanlong Zhang, Jiaqi Fan, Zhi Lu, Xinhong Xu, Xinyang Li#, et al. Unsupervised transfer learning enables multi-animal tracking without training annotation, Nature Methods, 2026.

[2] Xinyang Li*, Yixin Li*, Yiliang Zhou, Jiamin Wu, et al. "Real-time denoising enables high-sensitivity fluorescence time-lapse imaging beyond the shot-noise limit." Nature Biotechnology, 2023.

[3] Xinyang Li*, Yuanlong Zhang*, Jiamin Wu#, Qionghai Dai#. "Challenges and opportunities in bioimage analysis." Nature Methods, 2023.

[4] Xinyang Li*, Guoxun Zhang*, Jiamin Wu, Yuanlong Zhang, et al. "Reinforcing neuron extraction and spike inference in calcium imaging using deep self-supervised denoising." Nature Methods, 2021.

[5] Xinyang Li*, Xiaowan Hu*, Xingye Chen*, Jiaqi Fan, et al. "Spatial redundancy transformer for self-supervised fluorescence image denoising." Nature Computational Science, 2023.

[6] Xinyang Li*, Guoxun Zhang*, Hui Qiao*, Feng Bao, et al. (2021). Unsupervised content-preserving transformation for optical microscopy. Light: Science & Applications, 2021.

[7] Zhi Lu*, Wentao Chen*, Feihao Sun, Jiaqi Fan, Xinyang Li, et al. Leveraging spatial-angular redundancy for self-supervised denoising of 3D fluorescence imaging without temporal dependency. Nature Communications, 2025.

[8] Zhifeng Zhao*, Yiliang Zhou*, Bo Liu*, Jing He*, Jiayin Zhao, Yeyi Cai, Jingtao Fan, Xinyang Li, et al. "Two-photon synthetic aperture microscopy for minimally invasive fast 3D imaging of native subcellular behaviors in deep tissue." Cell, 2023.

[9] Guoxun Zhang*, Xiaopeng Li*, Yuanlong Zhang*, Xiaofei Han, Xinyang Li, et al. Bio-friendly long-term subcellular dynamic recording by self-supervised image enhancement microscopy. Nature Methods, 2023.

[10] Yuanlong Zhang*, Guoxun Zhang*, Xiaofei Han, Jiamin Wu, Ziwei Li, Xinyang Li, et al. Rapid detection of neurons in widefield calcium imaging datasets after training with synthetic data. Nature Methods, 2023.