Search

• A deep learning-based extended aperture image reconstruction method is proposed to overcome aperture limit. • Signals from limited and full-apertures were used as input and target, respectively, to train the proposed method. • The proposed image reconstruction method is evaluated by simulation, experimental data, and in vivo studies. • Our reconstruction method significantly improves the image quality of the boundary region in plane-wave ultrasound imaging. B-mode images undergo degradation in the boundary region because of the limited number of elements in the ultrasound probe. Herein, a deep learning-based extended aperture image reconstruction method is proposed to reconstruct a B-mode image with an enhanced boundary region. The proposed network can reconstruct an image using pre-beamformed raw data received from the half-aperture of the probe. To generate a high-quality training target without degradation in the boundary region, the target data were acquired using the full-aperture. Training data were acquired from an experimental study using a tissue-mimicking phantom, vascular phantom, and simulation of random point scatterers. Compared with plane-wave images from delay and sum beamforming, the proposed extended aperture image reconstruction method achieves improvement at the boundary region in terms of the multi-scale structure of similarity and peak signal-to-noise ratio by 8% and 4.10 dB in resolution evaluation phantom, 7% and 3.15 dB in contrast speckle phantom, and 5% and 3 dB in in vivo study of carotid artery imaging. The findings in this study prove the feasibility of a deep learning-based extended aperture image reconstruction method for boundary region improvement. [ABSTRACT FROM AUTHOR]

Simple Summary: Surgical therapy is critical to pancreatic cancer survival. The segmentation of pancreatic cancer in endoscopic ultrasonography (EUS) images can provide critical characteristics of the pancreatic cancer for surgical therapy. However, EUS has high operator dependency, and it requires a considerable level of experience and competency to stage pancreatic cancer. Deep learning approaches have been used on EUS images, but there have been no studies on the segmentation of pancreatic cancer. The purpose of this study is to segment pancreatic cancer from EUS images using a neural network model with deep attention features, called DAF-Net. The significance of this study lies in the successful segmentation performance of the pancreatic cancer using the DAF-Net, regardless of the size and location of the cancer, and its usefulness in preoperative planning. Endoscopic ultrasonography (EUS) plays an important role in diagnosing pancreatic cancer. Surgical therapy is critical to pancreatic cancer survival and can be planned properly, with the characteristics of the target cancer determined. The physical characteristics of the pancreatic cancer, such as size, location, and shape, can be determined by semantic segmentation of EUS images. This study proposes a deep learning approach for the segmentation of pancreatic cancer in EUS images. EUS images were acquired from 150 patients diagnosed with pancreatic cancer. A network with deep attention features (DAF-Net) is proposed for pancreatic cancer segmentation using EUS images. The performance of the deep learning models (U-Net, Attention U-Net, and DAF-Net) was evaluated by 5-fold cross-validation. For the evaluation metrics, the Dice similarity coefficient (DSC), intersection over union (IoU), receiver operating characteristic (ROC) curve, and area under the curve (AUC) were chosen. Statistical analysis was performed for different stages and locations of the cancer. DAF-Net demonstrated superior segmentation performance for the DSC, IoU, AUC, sensitivity, specificity, and precision with scores of 82.8%, 72.3%, 92.7%, 89.0%, 98.1%, and 85.1%, respectively. The proposed deep learning approach can provide accurate segmentation of pancreatic cancer in EUS images and can effectively assist in the planning of surgical therapies. [ABSTRACT FROM AUTHOR]

An image reconstruction method that can simultaneously provide high image quality and frame rate is necessary for diagnosis on cardiovascular imaging but is challenging for plane-wave ultrasound imaging. To overcome this challenge, an end-to-end ultrasound image reconstruction method is proposed for reconstructing a high-resolution B-mode image from radio frequency (RF) data. A modified U-Net architecture that adopts EfficientNet-B5 and U-Net as the encoder and decoder parts, respectively, is proposed as a deep learning beamformer. The training data comprise pairs of pre-beamformed RF data generated from random scatterers with random amplitudes and corresponding high-resolution target data generated from coherent plane-wave compounding (CPWC). To evaluate the performance of the proposed beamforming model, simulation and experimental data are used for various beamformers, such as delay-and-sum (DAS), CPWC, and other deep learning beamformers, including U-Net and EfficientNet-B0. Compared with single plane-wave imaging with DAS, the proposed beamforming model reduces the lateral full width at half maximum by 35% for simulation and 29.6% for experimental data and improves the contrast-to-noise ratio and peak signal-to-noise ratio, respectively, by 6.3 and 9.97 dB for simulation, 2.38 and 3.01 dB for experimental data, and 3.18 and 1.03 dB for in vivo data. Furthermore, the computational complexity of the proposed beamforming model is four times less than that of the U-Net beamformer. The study results demonstrate that the proposed ultrasound image reconstruction method employing a deep learning beamformer, trained by the RF data from scatterers, can reconstruct a high-resolution image with a high frame rate for single plane-wave ultrasound imaging. • An end-to-end image reconstruction method is proposed for high-resolution plane-wave ultrasound imaging. • A modified U-Net architecture is developed as a deep learning beamformer trained by pre-beamformed radio frequency (RF) data. • Our beamforming model is trained by simulation data from only point targets and evaluated by in vivo experimental data. • Our image reconstruction method improves the image quality with reduced computational time for higher frame rate imaging. [ABSTRACT FROM AUTHOR]

Infrared thermal imaging has been widely used to show the correlation between thermal characteristics of the body and muscle activation. This study aims to investigate a method using thermal imaging to visualize and differentiate target muscles during resistance training. Thermal images were acquired to monitor three target muscles (i.e., biceps brachii, triceps brachii, and deltoid muscle) in the brachium while varying the training weight, duration, and order of training. The acquired thermal images were segmented and converted to heat maps. By generating difference heat maps from pairs of heat maps during training, the target muscles were clearly visualized, with an average temperature difference of 0.86 °C. It was observed that training order had no significant effect on skin surface temperature. The difference heat maps were also used to train a convolutional neural network (CNN) to show the feasibility of target muscle classification, with an accuracy of 92.3%. This study demonstrated that infrared thermal imaging could be effectively utilized to locate and differentiate target muscle activation during resistance training. [ABSTRACT FROM AUTHOR]

Mucinous cystic neoplasms (MCN) and serous cystic neoplasms (SCN) account for a large portion of solitary pancreatic cystic neoplasms (PCN). In this study we implemented a convolutional neural network (CNN) model using ResNet50 to differentiate between MCN and SCN. The training data were collected retrospectively from 59 MCN and 49 SCN patients from two different hospitals. Data augmentation was used to enhance the size and quality of training datasets. Fine-tuning training approaches were utilized by adopting the pre-trained model from transfer learning while training selected layers. Testing of the network was conducted by varying the endoscopic ultrasonography (EUS) image sizes and positions to evaluate the network performance for differentiation. The proposed network model achieved up to 82.75% accuracy and a 0.88 (95% CI: 0.817–0.930) area under curve (AUC) score. The performance of the implemented deep learning networks in decision-making using only EUS images is comparable to that of traditional manual decision-making using EUS images along with supporting clinical information. Gradient-weighted class activation mapping (Grad-CAM) confirmed that the network model learned the features from the cyst region accurately. This study proves the feasibility of diagnosing MCN and SCN using a deep learning network model. Further improvement using more datasets is needed. [ABSTRACT FROM AUTHOR]