Time-domain waveform-informed photoacoustic microscopy

Abstract

Conventional photoacoustic microscopy (PAM) relies on scalar amplitude projection and struggles with weakly absorbing samples due to weak signal generation. We present waveform-informed photoacoustic microscopy (WiPAM), a waveform-domain contrast strategy demonstrated here in a photoacoustic shadow-casting microscopy configuration and centered on time-domain waveform morphology. Unlike standard amplitude-based projection, WiPAM utilizes the Karhunen–Loève expansion to identify spatiotemporally coherent latent features from the full photoacoustic waveforms. By treating the pixel-wise time-domain signal as a multidimensional information carrier, this approach captures hybrid opto-acoustic interactions that remain decoupled from scalar amplitude. Demonstrated on weakly absorbing biological samples, WiPAM achieved over 40-fold contrast enhancement and a signal-to-noise ratio of 55.6 dB. Furthermore, WiPAM enables label-free differentiation of compositionally heterogeneous tissues, yielding a significantly higher statistical separability (Bhattacharyya distance of 1.28 vs 0.15 for conventional methods). The method offers a simple, stable, and cost-effective framework adaptable to existing PAM systems, extending photoacoustic imaging toward multidimensional, label-free characterization of previously invisible biological structures.