Deep learning for image reconstruction in thermoacoustic tomography

Microwave-induced thermoacoustic tomography(TAT)is a rapidly-developing noninvasive imaging technique that integrates the advantages of microwave imaging and ultrasound imaging.While an image reconstruction algorithm is critical for the TAT,current reconstruction methods often creates significant artifacts and are computationally costly.In this work,we propose a deep learning-based end-to-end image reconstruction method to achieve the direct reconstruction from the sinogram data to the initial pressure density image.We design a new network architecture TAT-Net to transfer the sinogram domain to the image domain with high accuracy.For the scenarios where realistic training data are scarce or unavailable,we use the finite element method(FEM)to generate synthetic data where the domain gap between the synthetic and realistic data is resolved through the signal processing method.The TAT-Net trained with synthetic data is evaluated through both simulations and phantom experiments and achieves competitive performance in artifact removal and robustness.Compared with other state-of-the-art reconstruction methods,the TAT-Net method can reduce the root mean square error to 0.0143,and increase the structure similarity and peak signal-to-noise ratio to 0.988 and 38.64,respectively.The results obtained indicate that the TAT-Net has great potential applications in improving image reconstruction quality and fast quantitative reconstruction.

Developing microwave-induced thermoacoustic tomography:System,application,and reconstruction

A microwave-induced thermoacoustic imaging(MITAT)system is a non-destructive physical medical imaging method that combines the advantages of the high contrast of microwave imaging and the high resolution of ultrasound imaging.It uses the microwave as the excitation source and ultrasound as the information carrier.When different kinds of biological tissue absorb electromagnetic energy,it results in localized temperature rises.The thermal expansion will induce ultrasonic signals(i.e.,thermoacoustic signals),known as the thermoacoustic effect.The microwave absorption image of the sample can be reconstructed by algorithm processing.The MITAT contrast depends on different dielectric parameters of different kinds of tissue.We introduce the developed system and its application.In addition,the challenges and prospects of MITAT for further development are discussed.

Two-Dimensional Photoacoustic Imaging of Blood Vessel Networks within Biological Tissues

Photoacoustic tomography （PAT） is a powerful imaging merits and most compelling features of light and sound, technique for medical diagnosis because it combines the We describe a PAT experimental system constructed in our laboratory which consists of a Q-switched Nd：YAG pulse laser operating at 532nm with a 8-ns pulse width to generate the photoacoustic signals from a biological sample. Two-dimensional photoacoustic imaging of blood vessel networks lcm below the tissue surface is achieved, We also successfully demonstrate that the system is capable of imaging the blood vessels over the ex vivo rat brain with skull and skin intact.

MICROW AVE-INDUCED THERMOA COUSTIC IMAGING FOR EARLY BREAST CA NCER DETECTION

Microwave induced thermoacoustic tomography(TAT)is a noninvasive,nonionizing modality based on the inherent differences in microwave absorpt ion of malignant breast tissues and normal adipose dominated breast tissues.In this paper,a TAT system based on multielement acquisition system was built to receive signals.Slices from different layers in the sample were composed into a three-dimensional(3D)volume.Based on the 3D volume,inherent differences in microw ave absorption bet ween different biological tisues can be converted into structure information.Our experimental results of some minicked and human tumors indicate that TAT may potentially be used to detect early-stage breast cancers with high contrast.