Deep-E Enhanced Photoacoustic Tomography Using Three-Dimensional Reconstruction for High-Quality Vascular Imaging

Abstract

Linear-array-based photoacoustic computed tomography (PACT) has been widely used in vascular imaging due to its low cost and high compatibility with current ultrasound systems. However, linear-array transducers have inherent limitations for three-dimensional imaging due to the poor elevation resolution. In this study, we introduced a deep learning-assisted data process algorithm to enhance the image quality in linear-array-based PACT. Compared to our earlier study where training was performed on 2D reconstructed data, here, we utilized 2D and 3D reconstructed data to train the two networks separately. We then fused the image data from both 2D and 3D training to get features from both algorithms. The numerical and in vivo validations indicate that our approach can improve elevation resolution, recover the true size of the object, and enhance deep vessels. Our deep learning-assisted approach can be applied to translational imaging applications that require detailed visualization of vascular features.

Deep Learning Enhanced Volumetric Photoacoustic Imaging of Vasculature in Human

“…The training network was developed based on the fully dense U‐net, [ 30 ] which has been widely used in biomedical imaging applications with great performance. [ 27 , 29 , 57 ] Compared to traditional 2D neural network approaches, 3D‐trained networks can better reveal the vascular structure in all 3D planes by exploring the data's volumetric information rather than cross‐sectional images. Compared with the conventional 3D U‐net network, the 3D FD U‐net leveraged the benefit of dense connectivity in each layer, allowing it to learn additional features and improve the model performance.…”

“…In a follow‐up study, we combined Deep‐E with 3D reconstructed data and added experimental noise to the training. [ 27 ] However, since the training is still performed in 2D, the misalignment between frames still exists.…”

Simplifying vein detection for intravenous procedures: A comparative assessment through near‐infrared imaging system

“…Vital designs, including the Accu Vein, 22 Vein Viewer, 21 and Vascu Luminator, 23 have been proposed for vein detection. Besides, many researchers have also explored alternative approaches for vessel visualization, such as bio-photonic applications, 24 ultrasound imaging processes, 25 transillumination solutions, 26 use of multisource heterogeneous fundus 27 datasets for image quality assessment, and photoacoustic imaging 28 to improve the vein detection. Despite the effectiveness of these imaging methods in assisting with cannulation, their prevailing adoption is hindered by the elevated cost of these devices and the potential ionizing effects associated with some of them.…”

A Systematic Literature Review of 3D Deep Learning Techniques in Computed Tomography Reconstruction

“…In this method, a patient is swiftly rotated around a narrow beam, creating signals processed by the machine's computer to produce cross-sectional pictures or slices. Once the scanner has gathered a number of these slices, which are known as tomography pictures, they are stacked together to create three-dimensional representations of the patient [5][6][7][8][9]. These three-dimensional representations are created by applying an algorithm to the raw data, resulting in image slices that are then reconstructed into a 3D volume [10,11].…”