Cascade neural approximating for few-shot super-resolution photoacoustic angiography

Abstract

High-resolution photoacoustic angiography images are reconstructed from undersampled images with the help of a super-resolution deep neural network, enhancing the ability of the photoacoustic angiography systems to image dynamic processes in living tissues. However, image degradations are difficult to estimate due to a lack of knowledge of the point spread function and noise sources, resulting in poor generalization capability of the trained super-resolution model. In this work, a high-order residual cascade neural network was developed to reconstruct high-resolution vascular images, which is a neural approximating approach used to remove image degradations of photoacoustic angiography. To handle overfitting in training super-resolution model with a limited dataset, we proposed a BicycleGAN based image synthesis method in data preparation, achieving a strong regularization by forging realistic photoacoustic vascular images that act to essentially increase the training dataset. The quantitative analysis of the reconstructed results shows that the high-order residual cascade neural network surpassed the other residual super-resolution neural networks. Most importantly, we demonstrated that the generalized model could be achieved despite the limited training dataset, promising to be a methodology for few-shot super-resolution photoacoustic angiography.

Photoacoustic imaging plus X: a review

“…Evaluations showed that it achieved substantially higher vessel contrast at depths over 2 cm in vivo. Other super-resolution methods include works by Ma et al 107 and He et al 108 High fidelity deconvolution methods, such as using RRDBNet, 109 also leverage deep learning for resolution improvement. RRDBNet is a deep residual network tailored for image deconvolution.…”

“…Evaluations showed that it achieved substantially higher vessel contrast at depths over 2 cm in vivo . Other super-resolution methods include works by Ma et al 107 . and He et al 108 …”

Iterative-in-iterative super-resolution biomedical imaging using one real image

“…8,9 Some researchers have proposed the restoration of resolution through deep neural networks, which would allow the focus to shift towards improving imaging speed during data collection. 10,11 However, training such models necessitates an extensive number of high-resolution images, which poses a challenge in the context of limited biomedical datasets for deep learning. The shortage of biomedical data raises the question of whether it is possible to train a neural network using a minimal number of high-resolution images or even just one.…”

Characterization of cerebrovascular changes in mice treated with alcohol by photoacoustic imaging

“…Due to the high optical absorption contrast and good ultrasound resolution, photoacoustic imaging (PAI) has proven to be a promising tool for vascular imaging based on endogenous contrast with blood hemoglobin [13,14]. So far, PAI has been applied in studies of functional connectivity [15], neurovascular coupling [16], ischemic stroke [17] and obesity [18].…”