Multiphoton super-resolution imaging via virtual structured illumination

Imaging in thick biological tissues is often degraded by sample-induced aberrations, which reduce image quality and resolution, particularly in super-resolution techniques. While hardware-based adaptive optics, which correct aberrations using wavefront shaping devices, provide an effective solution, their complexity and cost limit accessibility. Computational methods offer simpler alternatives but struggle with complex aberrations due to the incoherent nature of fluorescence. Here, we present a deep-tissue super-resolution imaging framework that addresses these challenges with minimal hardware modification. By replacing the photodetector in a standard laser-scanning microscope with a camera, we measure an incoherent response matrix (IRM). A dual deconvolution algorithm is developed to decompose the IRM into excitation and emission optical transfer functions and the object spectrum. The proposed method simultaneously corrects excitation and emission point-spread functions (PSFs), achieving a resolution of {\lambda}/4, comparable to structured illumination microscopy. Unlike existing computational methods that rely on vector decomposition of a single convoluted PSF, our matrix-based approach enhances image reconstruction, particularly for high spatial frequency components, enabling super-resolution even in the presence of complex aberrations. We validated this framework with two-photon super-resolution imaging, achieving a lateral resolution of 130 nanometers at a depth of 180 micrometers in thick mouse brain tissue.