Photoacoustic imaging (PAI) is a biomedical imaging technique that provides optical contrasts at depth through the generation of acoustic waves with light. Handheld systems can be made for 3D navigation to image the vasculature and its oxygenation. However, these systems are impacted by limited view artifacts that hide some structures and by low contrast-to-noise when using a sparse array. Photoacoustic Fluctuation imaging (PAFI) enables to solve these two issues and we will develop further this technique towards quantitative view-full SO2 imaging. One of the challenges is to correct the spectral coloring effects due to the tissue surrounding the vessels. To this end, we will employ a highly novel multi-modal combination. Beyond these advances concerning PAFI, this technique still suffers from its low temporal resolution. Relying on deep neural networks (DNN) image enhancement abilities, we will improve the frame rate towards the ultimate limit of SO2 images obtained from single shot multispectral images. A main challenge for DNN is to provide quantitative predictions. To address this issue, FULBOX proposes novel experimental approaches to enable real-time full-view imaging of blood oxygenation, which, coupled to state-of-the-art Ultrasound Doppler will enrich the diagnosis for several pathologies.