In acoustic resolution PA microscopy (AR-PAM), high signal-to-noise ratio (SNR) images require averaging pulse-illuminated A-scans, slowing data acquisition. Multi-wavelength PA imaging further reduces scan speed, leading to issues like photobleaching of exogenous contrast agents due to prolonged exposure. Traditional noise removal algorithms fall short, while deep learning models like U-Net, though effective, also compromise image contrast. We propose a platform-flexible denoising conditional GAN (modified Pix2Pix) to generate high SNR images with single pulse illumination, reducing AR-PAM scan time by 30-fold. Tested across various systems, our model minimizes photobleaching, enhances efficiency, and remains adaptable to diverse hardware setups without retraining, broadening its preclinical and clinical applications.

Photoacoustic (PA) imaging is hybrid imaging modality with good optical contrast and spatial resolution. Portable, cost-effective, smaller footprint LEDs are rapidly becoming important PA optical sources. However, the key challenge faced by the LED-based systems is the low fluence that is generally compensated by high frame averaging; consequently reducing acquisition frame-rate. In this study, we present a simple deep learning U-Net framework that enhances the signal-to-noise ratio (SNR) and contrast of the low number of frame-averaged PA images. The SNR increased by approximately 4-fold for both in-class in vitro phantoms (4.39 ± 2.55) and out-of-class in vivo models (4.27 ± 0.87). We also demonstrate the noise invariancy of the network and discuss the downsides (blurry outcome and fails to reduce the salt & pepper noise). Overall, the developed U-Net framework can provide a real-time image enhancement platform for clinically translatable low-cost and low-energy light source-based PA imaging systems.