Description
Stimulated Raman scattering (SRS) microscopy, with its label-free chemical contrast, has shown great potential in biomedical research. However, conventional point-scanning approaches based on ultrashort pulsed lasers suffer from low imaging throughput and long acquisition times, while high excitation power can induce phototoxicity, severely limiting their application in dynamic monitoring of living tissues. To address these limitations, this study proposes a low-damage, high-speed SRS imaging strategy. By introducing an ultrafast resonant galvo scanning system, tissue exposure time is significantly reduced. Combined with low-power laser excitation and deep learning–driven image enhancement algorithms, this approach effectively compensates for signal loss caused by noise reduction. The proposed strategy maintains high signal-to-noise ratio (SNR) imaging quality while substantially reducing tissue photodamage and thermal effects, providing a feasible technical pathway for label-free, large-field in vivo tissue examination and intraoperative pathological diagnosis.
