This paper introduces a novel platform designed to enhance the imaging quality of handheld photoacoustic imaging (PAI) systems, addressing the limitations of current portable PAI devices. The platform integrates the MultiResU-Net imaging enhancement algorithm with a Ferroelectric random-access memory (FeRAM) crossbar array, enabling efficient in-memory computing that is highly suitable for deep neural networks involving extensive matrix multiplications. The hardware implementation is optimized for low-power operation on edge devices, and a specifically designed algorithmic strategy is introduced to accurately simulate hardware variations with a time complexity of O(mn). The feasibility and effectiveness of this approach are demonstrated through simulations using synthesized and in vivo data, showing a more than tenfold improvement in imaging resolution. The neural network inference is significantly accelerated, completing within microseconds, thereby fully supporting real-time imaging. The entire platform is compact, with dimensions of 25×25×20 cm3, making it a portable, high-resolution, real-time imaging solution for personalized healthcare.

Recent advancements have marked significant progress in photoacoustic imaging as an effective method for acquiring deep bio-tissue visuals in modern medical clinical therapy and the efficacy of U-Net and its variants has been established for imaging quality enhancement in this field. Unlike common computer vision datasets such as ImageNet [1] and PASCAL VOC [2], biomedical images exhibit highly structured patterns, low spatial resolution, and single-channel modality, as shown in Fig. 1. Additionally, the U-Net parameters trained for medical super-resolution tasks demonstrate a high sparsity ratio, making them suitable for implementation on edge-computing platforms. Therefore, developing an energy-efficient photoacoustic imaging setup in this area is a natural progression. However, this development is constrained by the current neural network architectures, which are built around a U-Net backbone. The multi-stage feature extractor, skip connection integration across different blocks, and the encoder-decoder backbone design pose significant challenges to cutting-edge computational hardware platforms. In this study, a scalable, sparsity-supported neural network accelerator architecture for bio-tissue imaging quality enhancement is proposed to meet the stringent requirements of latency and energy efficiency, as depicted in Fig. 2. This architecture achieves desired performance improvements by exploring the sparsity possibilities in neural network during the training process and implementing an end-to-end pixel-first hardware design to minimize data movement and support sparsity computation. Compared with the state-of-the-art related works, this optimized architecture has achieved minimum on-chip storage overhead and the fastest frame for the application of photoacoustic imaging quality enhancement. The scalable architecture has also been implemented on a Xilinx XCZU9EG FPGA and attains a performance of PSNR@ 24 dB and a frame rate of 164 fps at a working frequency of 250 MHz.

Acoustic resolution photoacoustic microscopy (AR-PAM) is a novel medical imaging modality, which can be used for both structural and functional imaging in deep bio-tissue. However, the imaging resolution is degraded and structural details are lost since its dependency on acoustic focusing, which significantly constrains its scope of applications in medical and clinical scenarios. To address the above issue, model-based approaches incorporating traditional analytical prior terms have been employed, making it challenging to capture finer details of anatomical bio-structures. In this paper, we proposed an innovative prior named group sparsity prior for simultaneous reconstruction, which utilizes the non-local structural similarity between patches extracted from internal AR-PAM images. The local image details and resolution are improved while artifacts are also introduced. To mitigate the artifacts introduced by patch-based reconstruction methods, we further integrate an external image dataset as an extra information provider and consolidate the group sparsity prior with a deep denoiser prior. In this way, complementary information can be exploited to improve reconstruction results. Extensive experiments are conducted to enhance the simulated and in vivo AR-PAM imaging results. Specifically, in the simulated images, the mean peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) values have increased from 16.36 dB and 0.46 to 27.62 dB and 0.92, respectively. The in vivo reconstructed results also demonstrate the proposed method achieves superior local and global perceptual qualities, the metrics of signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) have significantly increased from 10.59 and 8.61 to 30.83 and 27.54, respectively. Additionally, reconstruction fidelity is validated with the optical resolution photoacoustic microscopy (OR-PAM) data as reference image.

A novel mixed-signal adaptable silicon-based coherent photoacoustic (PA) sensing system-on-chip (SoC) is proposed to detect various kinds of target signals robustly under high noise and strong interferences in compact chip-level, attaining precise in vivo sensing for physiological signs monitoring and enhanced bio-imaging. Based on the configurable coherent PA sensing SoC architecture supported by on-chip Gilbert cell-based multiplier, a digital processing module, and DACs, in-phase (I) and quadrature (Q) templates generated by digital module on-chip are configurable to be with a high correlation coefficient to the target PA signal, attaining detection and reconstruction of target signals in a coherent detection mode. The correlation between the received PA signal and the templates is implemented efficiently, assuring accurate tracking and precise reconstruction on the target PA signals at the chip level. Based on the integrated PA SoC fabricated by the TSMC 65-nm CMOS process, precise in vivo sensing and imaging can be assured at the edge. Further, as PA detection leverages optical and ultrasound sensing, in vivo imaging on in-depth vessels or other tissues can be attained. The mixed-signal PA SoC paves the way for sustainable health monitoring and owns immense potential for early disease diagnostics based on in vivo blood temperature sensing and vessel imaging.

Optical resolution photoacoustic microscopy (OR-PAM) imaging method can achieve high lateral resolution $(\lt 5 \mu \mathrm{m})$, while the penetration depth for OR is shallow (up to $1 \sim 2$ mm). In contrast, acoustic resolution photoacoustic microscopy (AR-PAM) imaging only has limited lateral resolution $(\gt 50 \mu \mathrm{m})$ but with deeper penetration depth up to several millimeters (3-10 mm). Enlighted by the recent progress in the field of machine learning, we proposed to enhance AR-PAM to OR-PAM while maintaining its high penetration depth merit with deep neural network, where a novel network structure named MultiResU-Net is employed. By training the network with OR images obtained with real setup and AR images simulated with physical model, the network is able to enhance the image quality of simulated AR image a huge extent that is similar to OR image. More importantly, the trained model is applied to real AR imaging system for both phantom and in vivo image enhancement. When compared with corresponding ground truth OR images, it can be fully substantiated that our proposed method realized the AR to OR target in real photoacoustic microscopy imaging system.

Acoustic resolution photoacoustic micros- copy (AR-PAM) can achieve deeper imaging depth in biological tissue, with the sacrifice of imaging resolution compared with optical resolution photoacoustic microscopy (OR-PAM). Here we aim to enhance the AR-PAM image quality towards OR-PAM image, which specifically includes the enhancement of imaging resolution, restoration of micro-vasculatures, and reduction of artifacts. To address this issue, a network (MultiResU-Net) is first trained as generative model with simulated AR-OR image pairs, which are synthesized with physical transducer model. Moderate enhancement results can already be obtained when applying this model to in vivo AR imaging data. Nevertheless, the perceptual quality is unsatisfactory due to domain shift. Further, domain transfer learning technique under generative adversarial network (GAN) framework is proposed to drive the enhanced image's manifold towards that of real OR image. In this way, perceptually convincing AR to OR enhancement result is obtained, which can also be supported by quantitative analysis. Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index (SSIM) values are significantly increased from 14.74 dB to 19.01 dB and from 0.1974 to 0.2937, respectively, validating the improvement of reconstruction correctness and overall perceptual quality. The proposed algorithm has also been validated across different imaging depths with experiments conducted in both shallow and deep tissue. The above AR to OR domain transfer learning with GAN (AODTL-GAN) framework has enabled the enhancement target with limited amount of matched in vivo AR-OR imaging data.

Photoacoustic Microscopy (PAM) optical resolution (OR) imaging method is suited to get high resolution bio-tissue image but suffers from shallow penetration depth. By contrast, photoacoustic acoustic resolution (AR) imaging has deeper penetration depth but with degraded imaging resolution. Inspired by the current advances in the field of deep neural network (DNN), we proposed a new DNN framework named Prior Residual U-Net (PRU-Net), which combines U-Net with global residual block and image prior for AR image to OR image resolution enhancement. It helps to aggregate the advantages of both imaging methods without the cost of building extra physical setup. By training the model with experimentally obtained OR image and simulated AR image pairs, the model is able to enhance the image quality from AR image towards OR image to a huge extent. The enhancement results of sub-images and complete image have both validated this method's effectiveness qualitatively and quantitatively.