IRR-Net: A Joint Learning Framework for Image Reconstruction and Recognition of Photoacoustic Tomography

Figure 6

Reconstructed images using our joint learning framework, where the image reconstruction modules consists of 1, 2, 3, and 4 DGD units, respectively, and the metrics for evaluating the image quality. (a) Images reconstructed from the simulation data; (b) images reconstructed from the scanning data of the experimental phantoms; (c) images reconstructed from the whole-body scanning data of living mice; (d) PSNR and SSIM of the reconstructed images from the simulation data; (e) CR and CNR of the phantom images; (f) CR and CNR of the in vivo images; and (g) training time and running time.

(a)

(b)

(c)

(d)

(e)

(f)

(g)