Enhancement of Acoustic Microscopy Lateral Resolution: A Comparison Between Deep Learning and Two Deconvolution Methods.
Researchers
Journal
Modalities
Models
Abstract
Scanning Acoustic Microscopy (SAM) provides high resolution images of biological tissues. Since higher transducer frequencies limit penetration depth, image resolution enhancement techniques could help in maintaining sufficient lateral resolution without sacrificing penetration depth. Compared to existing SAM research, the current work introduces two novelties. First, Deep Learning (DL) is used to improve lateral resolution of 180 MHz SAM images, comparing it with two deconvolution-based approaches. Secondly, 316 MHz images are used as ground truth in order to quantitatively evaluate image resolution enhancement. The samples used were mouse and rat brain sections. The results demonstrate that DL can closely approximate ground truth (NRMSE = 0.056, PSNR = 28.4 dB) even with a relatively limited training set (4 images, each smaller than 1 mm W 1 mm). The current study suggests the high potential of using DL as a single image superresolution method in SAM.