Optical neural networks leverage the unique properties of light propagation, interference, and diffraction to implement neural network functionalities, offering distinctive advantages including large bandwidth, low power consumption, parallel processing capability, and compact integration. These merits have led to their widespread applications in high-performance computing and artificial intelligence. Among various implementations, integrated optical neural networks (IONN) have emerged as a prominent research focus due to the superior integration density, cost-effectiveness, strong anti-noise capability, and excellent scalability offered by integrated photonic platforms. Here, we provide a comprehensive review of IONN development, fundamental computing principles, recent advances in two representative integrated platforms, and performance evaluation metrics across different architectures. Furthermore, we critically discuss future research directions and key challenges in this rapidly evolving field.

In the past decade, the strong coupling between traditional optical microcavities and excitons has been extensively studied, demonstrating broad application prospects in quantum computing and lasers fields. However, limited by the relatively low quality factor and small mode volume, the strong coupling is difficult to achieve in traditional optical microcavities. Compared to traditional optical microcavities, the bound states in the continuum (BIC) are special eigenstates that exist in the radiative continuum without any energy leakage. BICs possess ultra-high quality factors and strong light field confinement abilities, offering a new approach to realizing strong light?matter interactions. In this paper, we first review the development process of BICs and briefly introduce their generation mechanisms and classification. Then, we discuss the topic of BIC-assisted strong coupling phenomenon and present several related examples. Finally, we summarize the current challenges and future applications of BIC-assisted strong coupling in possible laser fields.

Imaging flow cytometry (IFC) is an emerging technology that combines the high-throughput analysis capabilities of traditional flow cytometry with high-resolution and high-specificity imaging of microscopes. Traditional flow cytometry is mainly used for rapid quantitative analysis of molecules at the single cell or subcellular level. The results are usually presented in the form of scatter plots, but lack information about cell morphology, structure, and subcellular signal distribution, which limits its application in cell structure research and intracellular molecular distribution analysis. IFC can image cells and has an extended depth of field function, allowing light from different focal planes to be focused on the detection plane at the same time. Therefore, through multiple detection channels, IFC can perform multi-parameter quantitative analysis of cell images, which not only retains the characteristics of traditional flow cytometry but also realizes high-resolution imaging and visualization of single cells. This article reviews the basic principles and latest progress of IFC from the perspective of two key technologies: optical imaging and image-guided cell sorting. It focuses on how this technology combines the high-throughput advantages of traditional flow cytometry with the high-resolution capabilities of microscopic imaging to overcome the limitations of traditional flow cytometry in cell morphology analysis and molecular distribution research, providing a reference for subsequent research.

The wavefront shaping technique based on classical optical disordered media shows significant application potential in the field of complex optical field distribution modulation and super-diffraction-limited imaging. However, its system performance depends on the construction of the transmission matrix and high-dimensional optical field measurements. At the same time, the bulky optical system architecture restricts its application potential in miniaturized and intelligent optical systems. To address this technical bottleneck, scholars have proposed engineered disordered metasurfaces, which break through the constraints between the optical memory effect and the angular scattering distribution of the traditional scattering media through the co-design of ordered?disordered structures at the sub-wavelength scale, and significantly enhance the system stability and polarization modulation performance. In this paper, we focus on the engineered disordered metasurface technology, sort out its progress and applications in the field of optical field modulation, and summarize the random phase model based on the disordered shape of meta-atoms and the positional disorder model based on the disorder of meta-atoms' position coordinates. The two design strategies effectively expand the control freedom of the optical system by introducing controllable engineering randomness, while maintaining deterministic control capabilities. This opens up new paths for applications such as the miniaturization of optical components, scattered imaging, and the integration of optical paths for light field control. Optical system design has entered a new stage of multi-dimensional collaborative control.

This article presents a systematic review of the physical mechanisms underlying symmetry-breaking-induced optical Bloch surface waves in one-dimensional photonic crystals and their applications in optical sensing, while exploring an artificial intelligence-driven collaborative design paradigm. The disruption of translational symmetry in one-dimensional photonic crystals through surface truncation enables the formation of localized surface states within the photonic bandgap, generating optical Bloch surface waves with subwavelength confinement characteristics. This mode enhances electromagnetic fields beyond conventional diffraction limits through interface evanescent field effects, thereby substantially increasing surface energy density and light?matter interaction efficiency. Through the application of transfer matrix methodology and photonic band theory, this paper examines the fundamental relationship between optical Bloch surface wave dispersion characteristics and localized field enhancement, with theoretical models validated through rigorous coupled-wave analysis and finite element simulation methods. The study establishes a comprehensive design framework for label-free optical Bloch surface wave biosensors, evaluating critical performance metrics including quality factor, sensitivity, and detection limits. Additionally, the paper discusses significant advances in optical Bloch surface waves across various applications, including nonlinear optical enhancement, micro-nano photonics manipulation, high-Q microcavity lasers, and on-chip interconnect devices. The conclusion offers insights into optimizing optical Bloch surface wave-based biosensor performance through artificial intelligence integration, providing theoretical foundations and technical pathways for advancing photonic sensor systems.