Real-Time Histogram-Based Automated Signal Exposure Correction on FPGA for Optoacoustic Imaging

Optoacoustic (OA) imaging combines optical excitation and ultrasound receive beamforming to render images of deep tissues with strong functional contrast, coupled with high spatial and temporal resolution. The widespread adoption of OA imaging critically depends on user-friendly, compact systems not requiring experts for manual parameter tuning. Manual adjustments for optimal Signal-to-Noise Ratio (SNR), often tailored to each target, impede high-frame-rate acquisition and are prone to image degradation if done incorrectly. To address these challenges, we propose a fully automated histogram-based signal exposure correction method based on Automated Gain Control (AGC). Our approach analyzes the histogram distribution of raw Analog Front-end (AFE) data to classify exposure conditions and dynamically adjust gain settings without user intervention. The algorithm is implemented on a mid-range FPGA (Xilinx Kria K26), operating at 400 MHz, across 16 input channels, and with minimal hardware overhead. We validate our approach on an in-vivo human angiography dataset and demonstrate that it reliably avoids overexposed and underexposed configurations and achieves image quality metrics comparable to expert-tuned acquisitions.