Photonic neuromorphic computing promises revolutionary advances in parallel and high-speed processing, yet a key challenge persists: co-integrating nonlinearity, dense connectivity, and intrinsi

Photonic neuromorphic computing promises revolutionary advances in parallel and high-speed processing, yet a key challenge persists: co-integrating nonlinearity, dense connectivity, and intrinsic memory monolithically to enable brain-inspired, spatiotemporal information processing. Here, we overcome this challenge by introducing a monolithic photonic neural network based on a metasurface operating at bound state in the continuum (BIC). The BIC mode mediates strong, long-range coupling across the lattice, supporting a reconfigurable recurrent network topology in hardware. Concurrently, the gain medium provides both optical nonlinearity for neuronal activation and a finite carrier lifetime that serves as a built-in, analog temporal memory. This synergy enables computation to emerge directly from the collective spatiotemporal dynamics of the driven-dissipative photonic system, effectively realizing a physical reservoir computer on a chip. We experimentally validate a minimal yet physically complete system on benchmark tasks—brain MRI image classification and human action recognition—achieving 92.16% and 85.25% accuracies, respectively. This work establishes a scalable pathway toward ultrafast, energy-efficient neuromorphic intelligence where processing is an inherent property of tailored light–matter interaction.show less

Fiber-optic imaging enables flexible, minimally invasive, deep-tissue imaging of otherwise inaccessible biological targets using optical fiber devices. Methods based on direct signal measurement

Fiber-optic imaging enables flexible, minimally invasive, deep-tissue imaging of otherwise inaccessible biological targets using optical fiber devices. Methods based on direct signal measurements have been widely adopted in biomedical research and clinical applications. However, imaging performance is bottlenecked by limited modal capacity, fiber-induced distortion, and reduced robustness. Driven by advances in fiber optics and modern computational imaging, computational fiber-optic imaging methods can effectively address these challenges, enabling high-fidelity, robust fiber-optic imaging that benefits fundamental research and clinical practice. This review first introduces the significance and challenges of direct fiber-optic imaging methods and contrasts their framework with that of computational fiber-optic imaging. Major optical-fiber devices and representative computational fiber-optic imaging strategies are then reviewed. Finally, the imaging performance and applications of established computational approaches, as well as future directions, are discussed and synthesized.show less

Single-photon imaging is an advanced optical technology that uses single-photon detectors to capture extremely weak light signals or individual photons carrying quantum information. It achieves

Single-photon imaging is an advanced optical technology that uses single-photon detectors to capture extremely weak light signals or individual photons carrying quantum information. It achieves high sensitivity, high temporal resolution, and multi-dimensional imaging by relying on single-photon detectors. Among many single-photon detection technologies, superconducting nanowire single-photon detectors (SNSPDs) have demonstrated superior performance and are widely used in both quantum information science and classical light-detection scenarios. Expanding single-pixel SNSPD detectors into large-scale arrays and integrating them with cryogenic systems, imaging optics, reconstruction algorithms, and other auxiliary units to build a practical high-performance single-photon camera is an inspiring goal in the community and will surely promote single-photon imaging applications. Although this task is difficult and challenging for superconducting electronics, significant progress has already been made. This review summarizes the developmental history of SNSPD arrays, various readout architectures, and remaining challenges, aiming to encourage cross-disciplinary research to accelerate the deployment and implementation of SNSPD cameras.show less

With the advent of the Quantum 2.0 era, technologies increasingly exploit nonclassical properties of quantum mechanics. Quantum imaging (QI) uses nonclassical phenomena such as quantum entanglem

With the advent of the Quantum 2.0 era, technologies increasingly exploit nonclassical properties of quantum mechanics. Quantum imaging (QI) uses nonclassical phenomena such as quantum entanglement, antibunching, and squeezing to develop new imaging modalities and enhance existing biomedical imaging techniques. Although QI has the potential to surpass certain classical limitations in biomedical imaging, it has seen limited adoption in biomedicine. This limited adoption may reflect a knowledge gap between the QI and biomedical imaging communities. In this review, we introduce biomedical QI through an overview and comparison of the key QI properties and methods relevant to biomedical imaging. We discuss fundamentally new methods enabled by QI, such as quantum ghost imaging, quantum imaging with undetected photons, and biphoton imaging, while also reviewing biomedical imaging methods that QI has already enhanced. Additionally, we compare the performance characteristics of quantum imaging cameras and representative biomedical QI methods. Finally, we discuss the limitations of QI and the factors that may limit future biomedical QI research.show less