Nonlinear optical microscopy (NLOM) is a technology that combines nonlinear optical effect with optical microscopy to generate contrast images by nonlinear light-matter interactions. Additionally, NLOM differs from conventional microscopy, which is typically based on linear interactions such as absorption, scattering, refraction, and fluorescence. In the past few decades, nonlinear optical imaging techniques have become important tools for detecting biomolecules, cells, and tissues at the micrometer and nanometer levels. The NLOM advancements promote and enhance the basic research on biology, pharmacy, and medicine. The nonlinear imaging techniques mainly include second harmonic generation (SHG), third harmonic generation (THG), two-photon excited fluorescence (TPEF), three-photon excited fluorescence (3PEF), coherent anti-Stokes Raman scattering (CARS) microscopy, and stimulated Raman scattering (SRS) microscopy. These techniques rely on tight focusing of ultrashort pulses with high photon density to excite nonlinear processes, which feature diffraction-limited spatial resolution and optical sectioning. Additionally, nonlinear optical microscopes employ near-infrared light sources that provide strong penetration power and cause minimal photodamage to tissues, allowing label-free imaging at the subcellular level. The nonlinear optical properties of different molecules in biological tissues enable molecular specificity and selectivity, making nonlinear optical imaging techniques widely applicable in biomedical imaging.

With the advances in biology, the applications of nonlinear optical imaging technology are expanding, and the complex structures and functions of living organisms pose new challenges to optical imaging. Biomedical research requires super-composite optical imaging technology to achieve multidimensional optical characterization of biological tissues and obtain comprehensive information about their microstructure and molecular metabolism. Multiple nonlinear contrastive imaging technologies eliminate the need for tedious tissue preparation and enable analysis of unlabeled tissue samples, which provides rich structural and functional information about complex organisms. Finally, the multimodal nonlinear optical imaging technology which integrates multiple optical characterization methods has emerged as a new direction in optical microscopy in recent years.

It is necessary to summarize and explore the existing research progress and future development trends to further promote the development of multimodal nonlinear optical imaging technology and contribute to relevant biomedical research. This will provide references for researchers in related fields.

The generation of nonlinear optical effects relies on focusing ultrashort pulse lasers to achieve extremely high peak intensity. When multiple photons simultaneously interact with excited fluorophores or specific structures, nonlinear optical signals are generated by light-matter interactions. A deep understanding about the generation process of various nonlinear effects is necessary to obtain optical images with high signal contrast and signal-to-noise ratio (SNR). Furthermore, selecting appropriate excitation conditions and detection methods is crucial for effective nonlinear optical imaging. We introduce the generation process of different nonlinear optical signals and their imaging mechanisms, mainly including multiphoton excitation fluorescence (MPEF), SHG/THG, coherent Raman scattering (CRS), and two-photon fluorescence lifetime microscope (TP-FLIM).

Multimodal nonlinear optical imaging technology allows for accurate and comprehensive multi-parameter optical physical information. It serves as an important tool in studying complex organisms and multi-threaded dynamic processes from a multi-dimensional perspective. This technology has extensive applications in biological research fields such as physiology, neurobiology, embryology, and tissue engineering. However, different nonlinear optical imaging systems have distinct requirements for optics and hardware in excitation conditions and detection methods. Therefore, the key to integrating multiple nonlinear optical imaging technologies lies in coordinating the synchronous excitation of multiple nonlinear effects and the simultaneous detection of multi-dimensional signals. Meanwhile, we elaborate on the technical challenges and solutions related to multimodal coupling in nonlinear optical imaging and introduce the research progress and biological applications of multimodal imaging with multiple coupling mechanisms.

Additionally, we review the optimization schemes for multimodal nonlinear optical imaging from three aspects of imaging speed, spatial resolution, and SNR to further improve the performance of multimodal optical imaging system. System miniaturization is discussed, and multimodal nonlinear optical endoscopy is extended to enable dynamic monitoring of the epidermis and internal organs of living organisms. Furthermore, nonlinear optical imaging microscopes can visualize the tissue structure and molecules in organism specificity. The imaging results require combined image processing methods for the quantitative detection of biological molecules and tissue structures. Therefore, we further introduce quantitative analysis methods for different nonlinear optical images.

Multimodal nonlinear optical microscopy, along with corresponding quantitative analysis methods, can conduct imaging and characterize the structure and physiological dynamic processes of biological tissues from multiple information dimensions. It represents an important branch of nonlinear optical microscopy development, with extensive applications in biomedical fields such as cell detection, cancer diagnosis, and brain imaging. Additionally, it holds significant potential, particularly in clinical pathological diagnosis. However, there are still several aspects of this technology to be further developed and improved. Firstly, in multimodal imaging, TP-FLIM imaging based on time-correlated single photon counting (TCSPC) requires a longer accumulation time for photons to obtain the lifetime decay curve. Simultaneously, spectral scanning in stimulated Raman scattering (SRS) imaging necessitates changing the position of time delay displacement tables. The two imaging methods still limit the imaging speed of the system and hinder the multi-parameter optical characterization for certain dynamic physiological processes. Therefore, there is still a room for further research on fast multimodal nonlinear optical imaging schemes.

Meanwhile, in practical applications, the images obtained from multi-parameter nonlinear optical imaging systems should be combined with corresponding analysis methods to extract relevant biochemical information. This requires extensive data processing and statistical analysis, particularly in the context of clinical pathological analysis. Exploring new analytical methods that enable rapid conversion from optical images to biological information will significantly enhance the clinical applicability of multimodal nonlinear optical imaging. In summary, despite the potential and utility in biomedical research presented by multimodal nonlinear optical microscopy, further advancements are needed to address challenges such as imaging speed and data analysis. By developing faster imaging schemes and exploring new analytical methods, the clinical applications of multimodal nonlinear optical imaging can be greatly enhanced.