Extraction Method of Target Region for HIFU Therapy Based on Ultrasonic Image Semantic Segmentation

As an important link between human and nature, land use/land cover change (LUCC) has become a research hotspot since the first day of the issue initiated. Remote sensing observation supplies important data source for LUCC research. With the constant improvement of the remote sensed image spatial resolution and the continuous development of computer scientific algorithms and models, the study of semantic segmentation in the field of automatic interpretation of remote sensed images has promoted quickly. There have been a number of achievements focusing on the one-class extraction of LUCC classification system, but limited progress in multi-class classifying mission when facing high resolution remote sensing data by using semantic segmentation model, because of its sophisticated structure, the phenomenon of synonyms spectrum and foreign body with spectrum in images. Additionally, the limitation of the semantic segmentation model is a crucial issue which needs to be solv for precision of classification in class recognition and the perfection of semantic segmentation on pixel level as well. The High-Resolution network (HRnet) model was used and improved in the study for semantic segmentation because it has the advantage of low resolution loss. In the LUCC classification work, the distribution of classes (such as cultivated land, woodland, grassland, other land, water, etc.) are usually extremely imbalanced, that leads to the limited classification recognition results of land types when using HRnet. So a combined Loss function of Lovasz Softmax Loss+Cross Entropy Loss was used in HRnet to classify the land cover of remote sensing images, to meet the requirements of high-resolution image classification not only in terms of low resolution loss but also in terms of class recognition. The combined loss function provided the direction of the model, and the appropriate loss function improved the performance of the model when faced with the imbalanced distribution of land classes. The results are as follows: 1) Compared with Deeplab and Unet, HRnet model performed better in segmentation results; 2) HRnet model with combined Loss function Lovasz Softmax Loss+Cross Entropy Loss (HRnet+) had better performance, and its test accuracy PA(Pixel Accuracy) value was 0.9577, MIOU (Mean Intersection over Union) value was 0.8801, and Kappa value was 0.9455. This indicated that the HRnet model with low loss of resolution could achieve better results in the automatic segmentation of high-resolution remote sensing images, and the combined loss function constructed in this study could effectively solve the problem of imbalanced distribution of classes in remote sensing images, and the edge optimization ability was also more significant.

This article addresses the semantic segmentation of laparoscopic surgery images, placing special emphasis on the segmentation of structures with a smaller number of observations. As a result of this study, adjustment parameters are proposed for deep neural network architectures, enabling a robust segmentation of all structures in the surgical scene. The U-Net architecture with five encoder-decoders (U-Net5ed), SegNet-VGG19, and DeepLabv3+ employing different backbones are implemented. Three main experiments are conducted, working with Rectified Linear Unit (ReLU), Gaussian Error Linear Unit (GELU), and Swish activation functions. The applied loss functions include Cross Entropy (CE), Focal Loss (FL), Tversky Loss (TL), Dice Loss (DiL), Cross Entropy Dice Loss (CEDL), and Cross Entropy Tversky Loss (CETL). The performance of Stochastic Gradient Descent with momentum (SGDM) and Adaptive Moment Estimation (Adam) optimizers is compared. It is qualitatively and quantitatively confirmed that DeepLabv3+ and U-Net5ed architectures yield the best results. The DeepLabv3+ architecture with the ResNet-50 backbone, Swish activation function, and CETL loss function reports a Mean Accuracy (MAcc) of 0.976 and Mean Intersection over Union (MIoU) of 0.977. The semantic segmentation of structures with a smaller number of observations, such as the hepatic vein, cystic duct, Liver Ligament, and blood, verifies that the obtained results are very competitive and promising compared to the consulted literature. The proposed selected parameters were validated in the YOLOv9 architecture, which showed an improvement in semantic segmentation compared to the results obtained with the original architecture.

The existing semantic segmentation network deelabv3+ has the problem of weak segmentation ability to small-scale target objects and rough edge segmentation. The method of parallel connection of multiple resolution subnets in HRNet network is introduced. After deeplabv3+ down sampling, the network layers of different sizes were fused with features, and the decode side was fused with up sampling step by step to improve the edge segmentation accuracy. Attention mechanism is added before feature fusion to improve the recognition ability of small object. At the end, the edge is refined again by using CRF random vector field. The test is carried out on Pascal VOC 2012, the experimental results show that: the segmentation is more detailed from the image edge details, the recognition of small objects is more accurate, the Pixel Accuracy (PA) and Mean Intersection over Union (MIOU) are improved compared with deeplabv3+.

In interpreting a scene for numerous applications, including autonomous driving and robotic navigation, semantic segmentation is crucial. Compared to single-modal data, multi-modal data allow us to extract a richer set of features, which is the benefit of improving segmentation accuracy and effect. We propose a point cloud semantic segmentation method, and a fusion graph convolutional network (FGCN) which extracts the semantic information of each point involved in the two-modal data of images and point clouds. The two-channel k-nearest neighbors (KNN) module of the FGCN was created to address the issue of the feature extraction's poor efficiency by utilizing picture data. Notably, the FGCN utilizes the spatial attention mechanism to better distinguish more important features and fuses multi-scale features to enhance the generalization capability of the network and increase the accuracy of the semantic segmentation. In the experiment, a self-made semantic segmentation KITTI (SSKIT) dataset was made for the fusion effect. The mean intersection over union (MIoU) of the SSKIT can reach 88.06%. As well as the public datasets, the S3DIS showed that our method can enhance data features and outperform other methods: the MIoU of the S3DIS can reach up to 78.55%. The segmentation accuracy is significantly improved compared with the existing methods, which verifies the effectiveness of the improved algorithms.

Ultrasound B-Mode imaging systems are often used to image tissues to which High Intensity Focused Ultrasound (HIFU) is being applied. Since HIFU operates at frequencies in the same range as a B-mode imager, the two technologies interfere when they are simultaneously applied to the patient. This causes the imaging window to 'whiteout' and obscure the target region, thus defeating the purpose of using the ultrasonic imager in the first place. The system in this paper controls the interference window by pulsing the HIFU at a frame rate synchronized to the ultrasonic imager. The frame rate is either approximated by using an on-board oscillator, or it is captured from stray electromagnetic waves received from the ultrasound transducer. The system is capable of delivering 100 W of peak electrical output power to a HIFU transducer. It is controlled by a custom software application, and synchronizes with B-Mode imaging systems with frame rates from 10-100 Hz. An on-board Direct Digital Synthesizer generates HIFU RF from 1-5 MHz. The complete system provides precision positioning of the interference window in a small and portable desktop package.

Despite the success of crack segmentation, the traditional segmentation strategy and networks are not necessarily suitable for fatigue cracks with unique visual features. Towards the indistinct crack features, a network named Fatigue Crack Unet (FC-Unet) was proposed, where the novel split attention mechanism was incorporated to boost diverse feature extraction. Scaled by various ratios, raw images were cropped and constituted three datasets to represent different segmentation strategies. FC-Unet was trained and tested on each dataset, where the class imbalance problem was alleviated by the adjustment of loss functions. Results showed that the segmentation performance is mainly limited to noise interference. Preserving abundant global contexts to distinguish noises, the global inference is the superior strategy to the previous local inference. Leveraging the wide receptive field in global inference, the down-sampling rate can be decreased to save resolution losses. Combining the focus-based and region-based loss functions, the accuracy of imbalanced data was further improved. Compared with nine classical networks, the proposed FC-Unet enjoyed a powerful backbone and achieved a superior metric of 83.1% mean intersection over union (MIoU).

To accurately evaluate residual plastic film pollution in pre-sowing cotton fields, a method based on modified U-Net model was proposed in this research. Images of pre-sowing cotton fields were collected using UAV imaging from different heights under different weather conditions. Residual films were manually labelled, and the degree of residual film pollution was defined based on the residual film coverage rate. The modified U-Net model for evaluating residual film pollution was built by simplifying the U-Net model framework and introducing the inception module, and the evaluation results were compared to those of the U-Net, SegNet, and FCN models. The segmentation results showed that the modified U-Net model had the best performance, with a mean intersection over union (MIOU) of 87.53%. The segmentation results on images of cloudy days were better than those on images of sunny days, with accuracy gradually decreasing with increasing image-acquiring height. The evaluation results of residual film pollution showed that the modified U-Net model outperformed the other models. The coefficient of determination(R2), root mean square error (RMSE), mean relative error (MRE) and average evaluation time per image of the modified U-Net model on the CPU were 0.9849, 0.0563, 5.33% and 4.85 s, respectively. The results indicate that UAV imaging combined with the modified U-Net model can accurately evaluate residual film pollution. This study provides technical support for the rapid and accurate evaluation of residual plastic film pollution in pre-sowing cotton fields.

Single-pixel imaging (SPI) receives widespread attention due to its superior anti-interference capabilities, and image segmentation technology can effectively facilitate its recognition and information extraction. However, the complexity of the target scene and plenty of imaging time in SPI make it challenging to achieve high-quality and concise segmentation. In this paper, we investigate the image-free intricate scene semantic segmentation in SPI. Using “learned”illumination patterns allows for the full extraction of the object’s spatial information, thereby enabling pixel-level segmentation results through the decoding of the received measurements. Simulation and experimentation show that, in the absence of image reconstruction, the mean intersection over union (MIoU) of segmented image can reach higher than 85%, and the Dice coefficient (DICE) close to 90% even at the sampling ratio of 5%. Our approach may be favorable to applications in medical image segmentation and autonomous driving field.

Deep-sea mining vehicles rely on forward-looking sonar systems to collect critical environmental data for mapping, navigation, and operational safety. Real-time semantic segmentation of seabed terrain from sonar images is essential for accurately interpreting the underwater environment. However, these images present several challenges, including low resolution, substantial noise, and indistinct textures and shapes. Additionally, seabed terrain features vary across multiple spatial scales, which complicates the extraction of meaningful information using conventional image processing techniques. Such limitations hinder the precision and reliability of mapping efforts necessary for effective deep-sea operations. To address these challenges, a multi-scale semantic segmentation model has been developed specifically for sonar images. The model incorporates a downsampling module based on Discrete Wavelet Transform (WT-DM), which suppresses noise while preserving critical terrain features, ensuring clearer input for subsequent analysis. Moreover, a TriScale Attention Module (TAM) is introduced to capture features at varying spatial resolutions. This module enhances segmentation performance by allowing the model to focus on relevant patterns across different scales. In the absence of publicly available datasets, experiments were conducted using a custom-built seabed terrain dataset. The proposed model achieved a precision of 82.4% and a mean intersection over union (MIoU) of 79.6%, demonstrating its effectiveness. Comparative evaluations with other mainstream models further confirmed the superior performance and competitiveness of the proposed approach.

Semantic segmentation of weed and crop images is a key component and prerequisite for automated weed management. For weeds in unmanned aerial vehicle (UAV) images, which are usually characterized by small size and easily confused with crops at early growth stages, existing semantic segmentation models have difficulties to extract sufficiently fine features. This leads to their limited performance in weed and crop segmentation of UAV images. We proposed a fine-grained feature-guided UNet, named FG-UNet, for weed and crop segmentation in UAV images. Specifically, there are two branches in FG-UNet, namely the fine-grained feature branch and the UNet branch. In the fine-grained feature branch, a fine feature-aware (FFA) module was designed to mine fine features in order to enhance the model's ability to segment small objects. In the UNet branch, we used an encoder-decoder structure to realize high-level semantic feature extraction in images. In addition, a contextual feature fusion (CFF) module was designed for the fusion of the fine features and high-level semantic features, thus enhancing the feature discrimination capability of the model. The experimental results showed that our proposed FG-UNet, achieved state-of-the-art performance compared to other semantic segmentation models, with mean intersection over union (MIOU) and mean pixel accuracy (MPA) of 88.06% and 92.37%, respectively. The proposed method in this study lays a solid foundation for accurate detection and intelligent management of weeds. It will have a positive impact on the development of smart agriculture. © 2024 Society of Chemical Industry.

Study of bubble behavior in a gas–solid dense-phase fluidized bed based on deep learning

Aiming at the problems of low detection accuracy, small proportion of crack pixels in images and sensitivity to noise in highway pavement crack detection, an improved UNet model based on multi-scale and attention mechanism (UNet-MSAM) was proposed. In the encoding stage, in order to sense the crack information of different scales in pavement images, the UNet-MSAM added multi-scale feature fusion mechanism. In addition, UNet-MSAM introduced an attention mechanism to model the relationship between the channel and position on the low-level features of the crack, instead of simply integrating the low-level and high-level features of the crack like UNet. The F1 score of UNet-MSAM on DeepCrack dataset was 84.13%, and the Mean Intersection over Union (MIOU) was 85.61%, which was better than the other three methods such as UNet. The above experimental results show that the proposed network model can accurately detect cracks.

High Intensity Focused Ultrasound (HIFU) is one of promising non-invasive thermal ablation techniques of tumor. In this paper, we present a segmentation method based on Support Vector Machine (SVM) for HIFU image-guided system where SVM is used to construct the prior model about the intensity and the shape of the structure from the training set of images and the boundaries. When segmenting a novel image, we improved level set method by incorporating this prior model. Segmentation results are demonstrated on ultrasonic images. It shows that the prior model makes segmentation process more robust and faster.

Image processing technology assists physicians in the analysis of athletes’ human motion injuries, not only to improve the accuracy of athletes’ injury detection but also to improve the localization and recognition of injury locations. It is important to accurately segment human motion injury ultrasound medical images. To address many problems such as poor effect of traditional ultrasonic medical image segmentation algorithm for a sports injury. Therefore, we propose a segmentation algorithm for human motion injury ultrasound medical images using deep feature fusion. First, the accurate estimated value of human posture is extracted and combined with image texture features and image gray value as the target feature value of the ultrasonic medical image of human motion injury. Second, the image features are deeply fused by an adaptive fusion algorithm to enhance the image resolution. Finally, the best segmentation value of the image is obtained by the trained support vector machine to realize the accurate segmentation of human motion injury ultrasonic medical image. The results show that the average accuracy of the posture accurate estimation of the proposed algorithm is 95.97%; the segmentation time of the human motion injury ultrasound medical image of the proposed algorithm is below 150 ms; and the convergence of the algorithm is completed when the number of iterations is 3. The maximum segmentation error rate is 2.68%. The image segmentation effect is consistent with the ideal target segmentation effect. The proposed algorithm has important application value in the field of ultrasonic medical diagnosis of sports injury.

Robust Prediction of Isodose Distribution with a Fully Convolutional Networks (FCN)-Based Deep Learning Model