上海科技大学知识管理系统

In practical applications within the human body, it is often challenging to fully encircle the target tissue or organ, necessitating the use of limited-view arrays, which can lead to the loss of critical information. Addressing the reconstruction of photoacoustic sensor signals in limited-view detection spaces has become a focal point of current research. In this study, we introduce an unsupervised network designed to tackle the inverse problem in photoacoustic imaging, aiming to reconstruct high-quality photoacoustic images from sensor data acquired under limited-view conditions. We propose a method that leverages fully connected neural networks and Fourier feature positional encoding to learn an implicit continuous representation of photoacoustic images from limited observation data. The network is trained by minimizing the error between the predicted and actual sensor data. Experimental results show that our model out-performs three commonly used methods in terms of reconstruction quality.