Grayscale Changes Research Articles

Case Series of Changes in Greyscale and Power Doppler Signals on Salivary Gland Ultrasonography after Treatment in Patients with Sjögren's Syndrome.

The usefulness of greyscale (GS) in salivary gland ultrasonography for Sjögren's syndrome (SS) has been established; however, limited information is currently available on power Doppler signals (PDs), and changes after treatment remain unknown. PDs are considered to represent glandular inflammation, which indicates the worsening of GS in later years. We examined the changes in PDs in three immunosuppressant-treated SS patients. PDs decreased, along with GS and markers of disease activity, after treatment. PDs have the potential to provide insights into glandular inflammation in real time; however, large-scale studies on their clinical usefulness are needed.

Objective Comparison of White Light and Narrow-Band Imaging for Detecting Scars, Sulci and Nodules.

Narrow-band imaging (NBI) can improve detection of lesions in the aerodigestive tract. However, its role in benign lesions of the larynx is unclear. This study aims to determine whether NBI improves the detection of scars, sulci, and nodules compared to panchromatic lighting using objective image analysis. In total, 120 vocal folds (VFs) were analyzed with and without NBI (21 normal, 15 scars, 16 sulci, and 45 nodules). Each VF image had anterior, middle, and posterior thirds analyzed for brightness using an area morphometry software (Optimas 5.1a). The middle-third with the lesion was analyzed against surrounding VF segments for average and standard deviation (SD) in absolute grayscale. The use of panchromatic light resulted in greater illumination and grayscale values than NBI. All lesions tended to be in the mid-membranous fold. Under panchromatic light, change in brightness when comparing anterior versus middle (A-M) was +6.1% for normal, versus 6.5%, 8.1%, and 7.1% for sulci, nodules, and scars, respectively. Under NBI, they were 9.0% (normal), 12.3% (sulci), 13.7% (nodules), and 13.1% (scars). A greater SD of luminescence was observed at pathology sites (p < 0.05) when using NBI. The change in absolute grayscale at all lesion sites was greater when using NBI than when using panchromatic light (p < 0.05). NBI significantly enhanced the area of pathology in patients with nodules, sulci, and scars. Greater SD values in grayscale at pathologic sites were observed compared at normal sites. Thus, NBI may improve the detection of phonotraumatic lesions compared to panchromatic light. N/A Laryngoscope, 134:4066-4070, 2024.

Evolution of cavitation clouds under cavitation impinging jets based on three-view high-speed visualization

Cavitation cloud is an important index for evaluating the unsteady behaviour of cavitation water jets. However, when cavitation jets are applied in a restricted space, the optimal standoff distance is not applied. To study the evolution and collapse mechanisms of cavitation clouds at small standoff distances, this study conducted visualization research on the behaviour of submerged cavitation water jet impacting the target surface at three standoff distances from three-view using a high-speed visualization system. The effect of the optimal standoff distance and the small standoff distance on the morphology of cavitation clouds was explored by observing their shedding behaviour and evolution on the target surface, obtaining a pattern for the morphological evolution of cavitation clouds. The evolution of cavitation clouds under impact flow conditions can be divided into stages: incipient cavitation, ring vortex cavitation, spindle-shaped cavitation cloud, mushroom-type cavitation cloud, cavitation ring, ring vortex cavitation, and collapse. The periodicity of cavitation clouds on the target surface was confirmed by monitoring the grayscale change at the center of the target. Furthermore, the changing of cavitation jet structures is illustrated through Proper Order Decomposition (POD) analysis, showing that the frequency of cavitation clouds increased with the decrease of standoff distance within a certain range. The behaviour of cavitation flow impinging on the target surface at optimal standoff distance and small standoff distance was simulated using the Large Eddy Simulation (LES) turbulence model combined with the Zwart-Gerber-Belamri (ZGB) cavitation model. The results show that the numerical simulation can capture the high-pressure zone generated when the cavitation cloud collapses, and the area of the high-pressure zone increases with the standoff distance within a certain range, the concentration of cavitation bubbles on the target surface did not represent the intensity of erosion. The transient vortex structures of the flow field at different standoff distances were investigated based on the Q-criterion. The findings of this study not only provide critical technical support for the application of cavitation jets in confined spaces, but also contribute to a broader understanding of the erosion mechanism of cavitation jets.

Spectral analysis with highly collimated mini-LEDs as light sources for quantitative detection of direct bilirubin

Because the human eye cannot visually detect the results of direct bilirubin test papers accurately and quantitatively, this study proposes four different highly collimated mini light-emitting diodes (HC mini-LEDs) as light sources for detection. First, different concentrations of bilirubin were oxidized to biliverdin by FeCl3 on the test paper, and pictures were obtained with a smartphone. Next, the red, green, and blue (RGB) channels of the pictures were separated to average grayscale values, and their linear relationship with the direct bilirubin concentration was analyzed to detect bilirubin on the test paper noninvasively and quantitatively. The experimental results showed that when green HC mini-LEDs were used as the light sources and image analysis was performed using the G channel, for a direct bilirubin concentration range of 0.1–2 mg/dL, the G channel determination coefficient (R2) reached 0.9523 and limit of detection was 0.459 mg/dL. The detection method proposed herein has advantages such as rapid analysis, noninvasive detection, and digitization according to RGB grayscale changes in the images of the detection test paper.

The role of implants and implant prostheses on the accuracy and artifacts of cone-beam computed tomography: an in-vitro study

To assess the accuracy of CBCT in implant-supported prostheses and to evaluate metal artifacts with and without implants or implant prostheses. Accuracy and artifacts were assessed in the dried mandible at three points on the buccal and lingual cortical plates on the mandible's body near the crest and the base. On the buccal cortical plate, these points were labelled as A, B and C near the crest and D, E and F near the base of the body of the mandible. Similarly, points a to f were marked on the lingual cortical plate corresponding to points A to F. The study had two control groups, C0 for physical linear measurement (PLM) and C1 for radiographic linear measurement (RLM) and artifact assessment. There were seven test groups, TG 1 to 7, progressing from a single implant to implant full-arch prosthesis. For accuracy assessment, PLM was compared to RLM. CBCT artifacts were investigated in images integrated at 0.25 mm, 10 mm, and 20 mm at regions of interest on concentric circles at different intersecting angles by comparing grayscale values at C1 and TG1 to 7. The data were collected and statistically analyzed. A significant difference was observed between C0 and C1, and RLM in test groups at the superior axial plane. Similarly, PLM and test RLM in the sagittal plane at A-B, B-C, and D-E were statistically significant. A significant difference between PLM and RLM was also observed in the vertical plane at A-D, B-E, and C-F. Quantification of CBCT artifacts in the presence of implants or prostheses revealed that full-arch prostheses had the highest mean grayscale value, whereas single implants with a prosthesis had the lowest. The mean grayscale change was greatest around the implant and implant prosthesis. The mean grayscale value was maximum at 20 mm voxel integration scales (VIS) and lowest at 0.25 mm. CBCT is a clinically reliable device. Metal in implants or implant-supported prostheses prevents true assessment of the peri-implant area; therefore, lower VIS is suggested in the presence of implants or implant prostheses.

Error-diffusion-kernel parameters for binary pattern in 1-bit fringe projection profilometry.

In fringe projection profilometry, 1-bit processing of 8-bit raster patterns is a common method to suppress nonlinear errors in commercial projectors and realize high-speed projection in industrial projectors. In the process of generating 1-bit fringes from sinusoidal fringes, the generation of high-order harmonics is inevitable; choosing to introduce fewer high-order harmonics of the algorithm is conducive to defocus to obtain a better sinusoidal pattern. This paper proposes a method to expand the error-diffusion kernel of the conventional Floyd-Steinberg diffusion dithering algorithm from 2×3 to 3×5, which can reduce the grayscale change of surrounding pixels and generate 1-bit fringes with fewer high-order harmonics. Meanwhile, this paper optimizes the parameters of the 3×5 error-diffusion kernel and proposes the optimal parameters for this kind of diffusion kernel. The simulation results show that the fringes generated by the proposed 3×5 error-diffusion-kernel algorithms are closer to sinusoidal fringes after Gaussian low-pass filtering. The experimental results show that the accuracy of the 3×5 diffusion kernel algorithms is higher.

Improving estimation of maize leaf area index by combining of UAV-based multispectral and thermal infrared data: The potential of new texture index

Crop leaf area index (LAI) is one of the important indicators to evaluate crop growth and guide field management, and can be used to predict crop yield. Spectral and thermal information extracted by multispectral (MS) and thermal infrared (TIR) sensors mounted on an unmanned aerial vehicle (UAV) can be used for LAI estimation. Image texture is sensitive to the changes in crop surface grayscale or color characteristics, and can be combined with spectral and thermal information to estimate the LAI. But single texture metric has limitations in LAI estimation. Therefore, the purpose of this study is to construct new texture indices based on texture metrics extracted from MS and TIR images, and combined spectral and thermal information to enhance the estimation accuracy of maize LAI. Three replicates of maize experiments under different irrigation treatments were conducted in 2020. The MS and TIR sensors were mounted on a UAV to acquire maize canopy images during critical growth stages and acquire field LAI value of samples synchronously. The LAI estimation models were established using MS data, TIR data, as well as their combination. These models were constructed by Back Propagation Neural Network (BPNN), Partial least squares regression (PLSR), and Random Forest Regression (RFR). Finally, the performance of LAI estimation models was evaluated by the coefficient of determination (R2), root mean square error (RMSE) and relative root mean square error (rRMSE). Results shown that: (i) Among the eight kinds of texture metrics extracted from MS and TIR images, the texture metric mean (MEA) has the best performance. Compared with single texture metrics, texture indices constructed by different metrics has stronger correlation with LAI. (ii) Adding texture indices to estimation models significantly improved model accuracy, especially multispectral three-texture index(MS-TTI) has higher LAI estimation potential than thermal infrared three-texture index(TIR-TTI). (iii) Compared with the use of MS or TIR data alone, the estimation model constructed by combining MS data and TIR data have better performance. The best estimation model obtained by the RFR method (R2 = 0.862, RMSE = 0.246 and rRMSE = 10.20 %) further improved the LAI estimation of maize, with R2 increasing by 6.55 % and 14.48 %, respectively. In conclusion, the combination of MS and TIR data can effectively improve the estimation accuracy of maize LAI, and also provide a feasible method for monitoring crop growth.

Research on heart rate extraction method based on mobile phone video.

As an important indicator of human health, heart rate is related to the diagnosis of cardiovascular diseases. In recent years, extracting the heart rate from the mobile phone image has become a research hotspot. However, the illumination intensity of the background, frame rate of the video, and resolution of the image influence heart rate detection accuracy. To overcome these problems, this study proposed a novel heart rate extraction method based on mobile video. Firstly, the mobile phone camera is engaged to record the finger video, the region of interest (ROI) is extracted through the iterative threshold, and the pulse signal is obtained according to the grayscale change of the resolution within the ROI. Then, a low-pass and a high-pass Butterworth filters are exploited to filter out the noise and interframes from the extracted pulse signal. Finally, an improved adaptive peak extraction algorithm is proposed to detect the pulse peaks and the heart rate derived from the difference in pulse peaks. The experimental results show that light intensity, frame rate and resolution all have an influence on the heart rate extraction accuracy, with the most obvious influence of light, the average accuracy of the experiment can reach 99.32% under good lighting conditions, while only 72.23% under poor lighting conditions. In terms of frame rate, increasing the frame rate from 30fps to 60fps, the accuracy is improved by 0.9%. For the resolution, increasing the resolution from 1080p to 2160p, the accuracy is improved by 1.12%. While comparing the proposed method with existing methods, the proposed method has a higher accuracy rate, which has important practical value and application prospects in telemedicine and daily monitoring.

Segmenting medical images via explicit–implicit attention aggregation

Medical image segmentation methods using the attention mechanism have achieved great results. However, these attention mechanisms primarily focus on the evaluation of explicit response values, ignoring the important information provided by the implicit energy, i.e., the grayscale change frequency contained in the data. The deficiency in the attention mechanism mentioned above limits its ability to aggregate valid features in medical image segmentation tasks. We propose an efficient network for medical image segmentation, named TransMIS. This network innovatively designs an implicit attention mechanism to interoperate with the explicit attention mechanism and fuses multi-level features with long-range and local dependencies to construct a robust feature representation, effectively improving its segmentation performance. Specifically, TransMIS introduces two new components: (1) A Multi-Scale Compression Self-Attention Mechanism (MSCS) that uses the sparse representation to replace complex feature tensors for matrix multiplication, which can effectively enhance feature representation while reducing the amount of computation. (2) An Explicit–Implicit Channel Mixer Module (CMM) that combines implicit energy quantification and explicit response value evaluation methods, which can bridge the discrepancies between different semantic information and fuse multi-level features effectively. With extensive experiments on multiple benchmark datasets, it shows that TransMIS outperforms current state-of-the-art methods for medical image segmentation. We also conduct rigorous ablation experiments to verify the effectiveness of the proposed components.

Kinetics and modeling of Pb (II) adsorption in pellet biochar based on micro-computed tomography characterization

Biochar, a cost-effective adsorbent for the removal of heavy metals from aqueous solutions, has gained increasing attention. In this study, an advanced micro-computed tomography (micro-CT) system was used to investigate the adsorption kinetics by direct localization and visualization of Pb (II) on wheat straw pellet biochar. The normalized digital images indicating the dynamic changes of Pb (II) adsorption on biochar samples at different initial Pb (II) concentrations of 100, 200, 300, and 400 mg/L and adsorption times were obtained. It was found that image grayscale (GS) changes over adsorption time (t) followed the power function, GSe/GSt=2.45∗t-0.27. Based on this finding, modified pseudo-first-order (PFO) and pseudo-second-order (PSO) models incorporated with time-dependent kinetic constants kPFOt=KPFO∗GSe/GSt and kPSOt=KPSO∗GSe/GSt were proposed, resulting in a better interpretation of the adsorption mechanism. The micro-CT-guided novel approach demonstrated visual evidence-based superiority and should prove valuable to the existing body of research in related fields.

Computed vs. film-based radiographs' contour artifacts influence diagnosis of secondary caries

To test local grey-scale changes on dental bitewing radiographs near filling margins for image acquisition. Forty approximal preparations in caries-free amalgam filled teeth and bitewing radiographs were acquired under standardized conditions applying four techniques. Film-based analog radiographs were digitized using flat-bed scanner (FDR). Phosphor-plate computed radiographs (PCR) were directly acquired by scanning VistaScan imaging plates. Image quality was tested using Preset Filter (PF) or manually applied IntraOral Fine Filter (IF) to enhance digital images. Local changes from digital imaging processing were assessed by comparing the margin-near (MN) and margin-far (MF) zone by a multivariate repeated measurements analysis. All images were acquired with 8-bit depth (256 shades). Dentine was displayed in 79 shades for FDR and 54 shades for PCR. PF or IF locally modify bitewing radiographs by darkening marginal dentine by 8 or 29 shades, respectively. The sharpest display of the margin (shades per pixel) from dentine to filling was found for IF (26.2), followed by FDR (23.2), PF (15.3) and PCR (8.3). Computed radiography with phosphor plates generate more homogeneous images compared to flatbed-digitized film-based radiographs. The filling margin was sharpest represented with the IF filter at the detriment of an artificial darkening of the dentine near the margin of the filling. Contour artifacts by filters have the potential to confound diagnosis of secondary caries. Algorithms and filters for sensor data processing, causing local changes above 2% of the dynamic range by non-continuous mathematical functions, should only be applied with caution, manually when diagnosing and reversibly.

A Coarse-to-Fine Feature Match Network Using Transformers for Remote Sensing Image Registration

Feature matching is a core step in multi-source remote sensing image registration approaches based on feature. However, for existing methods, whether traditional classical SIFT algorithm or deep learning-based methods, they essentially rely on generating descriptors from local regions of feature points, which can lead to low matching success rates due to various challenges, including gray-scale changes, content changes, local similarity, and occlusions between images. Inspired by the human approach of finding rough corresponding regions globally and then carefully comparing local regions, and the excellent global attention property of transformers, the proposed feature matching network adopts a coarse-to-fine matching strategy that utilizes both global and local information between images to predict corresponding feature points. Importantly, the network has great flexibility of matching corresponding points for any feature points and can be effectively trained without strong supervised signals of corresponding feature points and only require the true geometric transformation between images. The qualitative experiment illustrate the effectiveness of the proposed network by matching feature points extracted by SIFT or sampled uniformly. In the quantitative experiments, we used feature points extracted by SIFT, SuperPoint, and LoFTR as the keypoints to be matched. We then calculated the mean match success ratio (MSR) and mean reprojection error (MRE) of each method at different thresholds in the test dataset. Additionally, boxplot graphs were plotted to visualize the distributions. By comparing the MSR and MRE values as well as their distributions with other methods, we can conclude that the proposed method consistently outperforms the comparison methods in terms of MSR at different thresholds. Moreover, the MSR of the proposed method remains within a reasonable range compared to the MRE of other methods.

15‐2: A Novel Compensation Algorithm for Multi‐Frame‐Drop on AMOLED Display

In this paper, a novel compensation algorithm is proposed for improving multi‐frame drop (MFD) on AMOLED display. The MFD phenomenon occurs as the image switching or sliding when the display with extremely low brightness, resulting in the color shift and distortion at the edge of moving objects. Because the current is too small and the emission time is too short at extremely low brightness, more than one frame is needed to alleviate the impact of the parasitic capacitance between the gate and drain node of driving TFT (DTFT), and the gradual changing of drain voltage (VD) causes the DTFT to move from the linear to saturation region, which also aggravates this phenomenon. This paper analyzes the MFD phenomenon based on the existing 7T1C pixel circuits, records the grayscale changes of multiple frames, and gives the different offsets at dropped frames. By experiment results, this method is not changing the pixel circuit, such as the 8T1C, which could compensate for the dropped frames for reducing MFD issues and is easily implemented in the display driver IC for many different OLED panels.

Detail Feature Inpainting of Art Images in Online Educational Videos based on Double Discrimination Network

In order to improve the image detail restoration effect of online education videos and improve the quality of online education images, a method for image detail restoration of online education videos based on dual discriminant networks is proposed. Utilize the grayscale changes in the selected spatial domain to enhance the detailed features of the original video art image, and construct a dual discriminant generative adversarial network model to repair the enhanced image details. Using a U-Net structured generation network to generate images similar to detail regions, and using image detail features as input, the initial image restoration is completed by training the generation network. Improve the authenticity and consistency of output image details through global and local discriminant networks. The experimental results show that this method has a good image enhancement effect in restoring image detail features in online education video art, improving the dynamic range and grayscale contrast of the image, and has a significant effect on image detail repair. The average peak signal-to-noise ratio of this method is 30.108, and the average structural similarity is 0.961, proving that the repair effect of this method is good. The average PSNR of this method under Gaussian noise interference is 29.68, which proves the robustness of this method.

Effects of Astronomical Cycles on Laminated Shales of the Paleogene Shahejie Formation in the Dongying Sag, Bohai Bay Basin, China

Laminated shales are widely developed in the Dongying Sag and have attracted much attention as an oil reservoir. Macroscopically, these shales generally have multi-scale cyclicity, which is closely related to the development of laminae. Therefore, analyzing the origin of their cyclicity is helpful to understanding the formation mechanism of laminated shales and the vertical heterogeneity of shale reservoirs, which are of great significance for continental shale oil exploration and development. In this study, a gamma ray (GR) logging series, high-resolution elemental geochemical data, high-resolution core scanning photos and grayscale data, and mineralogical data were used to characterize the cyclicity of shale at different scales, and their relationship with different astronomical cycles was discussed. The results show that the Es3L and Es4U shale in the Dongying Sag has cyclicity from the meter-scale to the ten-meter scale and then to the hundred-meter scale, which is mainly manifested by periodic changes in organic matter abundance, mineral composition, element abundance, and grayscale. These cycles of different scales coincide with different astronomical periods. Specifically, the hundred-meter scale cyclicity is mainly controlled by the very long orbital period; the ten-meter scale cyclicity is mainly related to the eccentricity cycle; while the precession period is the main driver of the meter-scale cyclicity. Finally, we propose a simplified model for illustrating the formation of rhythmic organic-rich shale. This study is helpful to understanding the origin of continental organic-rich shale and predicting shale reservoir properties.

Challenges in DIC measurement of segregation frost heave and its solutions by adaptive updating algorithm

Two challenges exist in the non-contact measurements of local frost heave of soil, such as the grayscale change caused by the generation of pore ice and the discontinuity of the deformation field caused by segregation cracks. To overcome the upper two challenges, a new imaging analysis algorithm based on the digital image correlation (DIC) method is developed in this study, leading to a modified DIC method for the deformation analysis of segregation frost heave. The proposed DIC method is applied to the segregation frost heave test in a soil column. The results demonstrate that the new algorithm can effectively reduce the differences in soil grayscale distribution and reduce the disappearance of correlation peaks due to segregation cracks. With the proposed DIC method, the non-contact measurement of frost heave deformation can achieve a resolution of less than 0.5 mm.

Locating robust patterns based on invariant of LTP-based features

• A novel concept of robust patterns (called RLTP) is introduced by considering the invariance of LTP-based features. • A crucial derivative of CLTP (called RCLTP) is proposed to be in accordance with RLTP. • An efficient application is presented to take into account RCLTP for local DT representation. • Local optical-flow-based features are taken into account DT motions for boosting the discrimination. • Just addressing the robust patterns, our RCLTP descriptor obtains very good performance compared to state of the art. Efficiently representing Dynamic Textures (DTs) based on salient features is one of the considerable challenges in computer vision. Locating these features can be obstructed due to the impact of encoding factors. In this article, a novel concept of Robust Local Ternary Patterns (RLTP) is introduced in consideration of the invariance of Local Ternary Patterns (LTP) subject to the deviation of thresholds. Our locating process is able to simultaneously encapsulate the discrimination of local features, and deal with the noise sensibility caused by a small gray-scale change of local neighbors. RLTP is then adapted to the completed LTP model to form an efficient operator for capturing completely properties of RLTP. Finally, RLTP is taken into account for the DT description, where the robust patterns of spatial-temporal features and optical-flow-based motions are exploited to improve the performance. Experiments have clearly corroborated the efficacy of our proposed approach.

Long-term outcome of ultrasound-guided focused ultrasound ablation for gestational trophoblastic neoplasia in the cesarean scar: a case report

BackgroundThe treatment of gestational trophoblastic neoplasia (GTN) is one of the success stories in medical oncology. GTN in the cesarean scar is a rare entity, but most cases need to be treated with hysterectomy or localized uterine lesion resection because of chemoresistant lesions and/or massive bleeding. We present a patient with post-molar GTN in the cesarean scar who was non-invasively treated with ultrasound-guided high intensity focused ultrasound (HIFU) to preserve the uterus and fertility.Case presentationA 32-year-old woman was diagnosed with low-risk GTN (FIGO Stage I: 2 prognostic score) after partial hydatidiform mole. The 5th cycle of chemotherapy was interrupted because of persistent hepatic toxicity and impaired ovarian reserve function. However, the uterine lesion persisted (diameter of residual uterine lesion in the cesarean scar: 2.0 cm). Therefore, ultrasound-guided HIFU treatment was performed. A significant gray-scale change was observed during the HIFU treatment. Color Doppler ultrasonography and contrast-enhanced ultrasound (CEUS) was performed to evaluate the ablation effectiveness. Color Doppler ultrasonography showed disappearance of the signal of vascularity and CEUS showed no perfusion in the lesion located in the cesarean scar. The uterine lesion was obviously shrunken one month after HIFU treatment. Menstrual cycle resumed 48 days after HIFU. HIFU treatment decreased the number of chemotherapy cycles and there was complete disappearance of the GTN lesion at 4-month follow-up. The patient has shown no signs of recurrence as of 58-month follow-up.ConclusionUltrasound-guided HIFU may be a useful alternative to lesion resection for GTN in the cesarean scar in patients who show chemoresistance or are not suitable for chemotherapy. It has the potential to ablate the residual uterine lesion noninvasively to preserve the uterus and fertility, avoiding perioperative risks of lesion resection, especially acute bleeding.

High-Intensity Focused Ultrasound in Interventricular Septal Myocardial Ablation.

High-intensity focused ultrasound (HIFU) can cause necrotic damage in deep tissues through thermal ablation and cavitation, without significant damage to the surrounding tissues. High blood perfusion of heart affects the energy deposition. This study aimed to evaluate the effect of cooling of coronary blood flow for HIFU ablation.Continuous and pulsed HIFU (2000 J) at duty cycles of 100% and 25% were examined for their capacity to ablate the perfused porcine heart tissue in vitro. After ablation, grayscale changes and pathological features were observed or measured, and the area and volume of tissue necrosis were calculated.The cardiomyocytes in the lesions underwent necrosis with a clear boundary. The endocardial surface was intact without necrosis. The three-dimensional morphology of the lesions appeared approximately as ellipsoids. With the increase in perfusion speed, the necrotic volume in the target area was gradually reduced.HIFU has the potential to become a new minimally invasive surgery for ventricular septal myocardial ablation. Reduction of coronary blood flow can improve the ablation effect.

Liver tumor segmentation and classification using FLAS-UNet++ and an improved DenseNet.

The incidence of liver tumors is among the top three in China. The treatments of benign and malignant tumors are different. Accurate diagnosis plays an important role in guiding the treatment of tumors. The aim of this study is to solve the following: (1) blurred boundary between the liver tumor and other organs causes incorrect segmentation of liver tumor boundaries; (2) large difference in tumor size and the diversity in texture and grayscale are major challenges in liver tumor classification tasks. Firstly, the liver tumor is segmented from the original CT images by a tumor segmentation network, UNet++ with fusion loss and atrous spatial pyramid pooling (FLAS-UNet++). The proposed segmentation method can solve the problem of tumor edge segmentation error by learning the tumor edge information. Secondly they are adaptively cropped according to the tumor volume to reduce the over-fitting and over-sensitivity of the deep network. Thirdly an improved Dense Block is designed to pay more attention to the changes in grayscale and texture between benign and malignant tumors. Finally, the features extracted from the network combined with tumor volume, patient's sex and age, are sent to a classifier for diagnosis. Liver tumor segmentation results show that the dice, HD95 reached 71.9%, 12.1 mm, respectively. The classification results show that the accuracy, specificity, sensitivity and area under curve reached 82.4%, 79.8%, 84.4%, 87.5%, respectively. The segmentation and classification results are both better than other's methods and mainstream networks. In order to solve existing problems of liver tumor CT image classification methods, our method realizes the accurate segmentation and classification of liver tumors in CT images and has important clinical application value.