Deep Learning Enhanced Volumetric Photoacoustic Imaging of Vasculature in Human

• Published in: Advanced science Vol. 10; no. 29; pp. e2301277 - n/a
• Main Authors: Zheng, Wenhan, Zhang, Huijuan, Huang, Chuqin, Shijo, Varun, Xu, Chenhan, Xu, Wenyao, Xia, Jun
• Format: Journal Article
• Language: English
• Published: Germany John Wiley & Sons, Inc 01-10-2023; John Wiley and Sons Inc; Wiley
• Subjects: 3D vascular Imaging; Acoustics; Algorithms; Arrays; Deep Learning; Humans; Image Processing, Computer-Assisted; Imaging, Three-Dimensional - methods; Lasers; linear transducer arrays; Medical imaging; Medical research; Neural networks; Noise; Photoacoustic Techniques - methods; photoacoustic tomography; Tomography; Tomography, X-Ray Computed; deep learning; 3D vascular Imaging; linear transducer arrays; photoacoustic tomography
• ISSN: 2198-3844; 2198-3844
• DOI: 10.1002/advs.202301277

The development of high‐performance imaging processing algorithms is a core area of photoacoustic tomography. While various deep learning based image processing techniques have been developed in the area, their applications in 3D imaging are still limited due to challenges in computational cost and memory allocation. To address those limitations, this work implements a 3D fully‐dense (3DFD) U‐net to linear array based photoacoustic tomography and utilizes volumetric simulation and mixed precision training to increase efficiency and training size. Through numerical simulation, phantom imaging, and in vivo experiments, this work demonstrates that the trained network restores the true object size, reduces the noise level and artifacts, improves the contrast at deep regions, and reveals vessels subject to limited view distortion. With these enhancements, 3DFD U‐net successfully produces clear 3D vascular images of the palm, arms, breasts, and feet of human subjects. These enhanced vascular images offer improved capabilities for biometric identification, foot ulcer evaluation, and breast cancer imaging. These results indicate that the new algorithm will have a significant impact on preclinical and clinical photoacoustic tomography.
A 3D fully‐dense U‐net is implemented in linear array based photoacoustic tomography. The trained network restores the true object size, reduces the noise level, improves the contrast at deep regions, and addresses the limited view problem. The performance is further validated in various human imaging applications, such as biometric identification, foot ulcer evaluation, and breast cancer imaging.