A New Deep Learning Network for Mitigating Limited-view and Under-sampling Artifacts in Ring-shaped Photoacoustic Tomography. (September 2020)

• Record Type: Journal Article
• Authors: Zhang, Huijuan; LI, Hongyu; Nyayapathi, Nikhila; Wang, Depeng; Le, Alisa; Ying, Leslie; Xia, Jun
• Is Part Of: Computerized medical imaging and graphics. Volume 84(2020)
• Publication Date: 2020-09
• Subjects: Deep learning -- Photoacoustic Tomography -- Limited-view -- Undersampling artifact -- Ring-shaped transducer array; Diagnostic imaging -- Periodicals; Imaging systems in medicine -- Periodicals; Diagnosis, Radioscopic -- Data processing -- Periodicals; Diagnostic Imaging -- Periodicals; Imagerie pour le diagnostic -- Périodiques; Diagnostic imaging; Periodicals; Electronic journals; Electronic journals; 616.0754
• DOI: 10.1016/j.compmedimag.2020.101720
• Languages: English
• ISSNs: 0895-6111

Abstract

Highlights: The designed deep-learning neural network can successfully remove under-sampling and limited-view artifacts in ring-array-based photoacoustic tomography. The designed deep-learning neural network can be applied in both numerical and experimental data. The designed deep-learning network can be implemented in circular arrays with different number of elements. The designed deep-learning network is superior to the conventional compressed sensing algorithm. Abstract: Photoacoustic tomography (PAT) is a hybrid technique for high-resolution imaging of optical absorption in tissue. Among various transducer arrays proposed for PAT, the ring-shaped transducer array is widely used in cross-sectional imaging applications. However, due to the high fabrication cost, most ring-shaped transducer arrays have a sparse transducer arrangement, which leads to limited-view problems and under-sampling artifacts. To address these issues, we paired conventional PAT reconstruction with deep learning, which recently achieved a breakthrough in image processing and tomographic reconstruction. In this study, we designed a convolutional neural network (CNN) called a ring-array deep learning network (RADL-net), which can eliminate limited-view and under-sampling artifacts in PAT images. The method was validated on a three-quarter ring transducer array using numerical simulation, phantom imaging, and in vivo imaging. Our results indicate that the proposed RADL-net significantly improves the quality of reconstructed images on a three-quarter ring transducer array. The method is also superior to the conventional compressed sensing (CS) algorithm.