RESUMEN

This article aims to combine deep learning with image analysis technology and propose an effective classification method for distal radius fracture types. Firstly, an extended U-Net three-layer cascaded segmentation network was used to accurately segment the most important joint surface and non joint surface areas for identifying fractures. Then, the images of the joint surface area and non joint surface area separately were classified and trained to distinguish fractures. Finally, based on the classification results of the two images, the normal or ABC fracture classification results could be comprehensively determined. The accuracy rates of normal, A-type, B-type, and C-type fracture on the test set were 0.99, 0.92, 0.91, and 0.82, respectively. For orthopedic medical experts, the average recognition accuracy rates were 0.98, 0.90, 0.87, and 0.81, respectively. The proposed automatic recognition method is generally better than experts, and can be used for preliminary auxiliary diagnosis of distal radius fractures in scenarios without expert participation.

Asunto(s)

RESUMEN

Background: Thyroid cancer (TC) prone to cervical lymph node (CLN) metastasis both before and after surgery. Ultrasonography (US) is the first-line imaging method for evaluating the thyroid gland and CLNs. However, this assessment relies mainly on the subjective judgment of the sonographer and is very much dependent on the sonographer's experience. This prospective study was designed to construct a machine learning model based on contrast-enhanced ultrasound (CEUS) videos of CLNs to predict the risk of CLN metastasis in patients with TC. Methods: Patients who were proposed for surgical treatment due to TC from August 2019 to May 2020 were prospectively included. All patients underwent US of CLNs suspected of metastasis, and a 2-minute imaging video was recorded. After target tracking, feature extraction, and feature selection through the lymph node imaging video, three machine learning models, namely, support vector machine, linear discriminant analysis (LDA), and decision tree (DT), were constructed, and the sensitivity, specificity, and accuracy of each model for diagnosing lymph nodes were calculated by leave-one-out cross-validation (LOOCV). Results: A total of 75 lymph nodes were included in the study, with 42 benign cases and 33 malignant cases. Among the machine learning models constructed, the support vector machine had the best diagnostic efficacy, with a sensitivity of 93.0%, a specificity of 93.8%, and an accuracy of 93.3%. Conclusions: The machine learning model based on US video is helpful for the diagnosis of whether metastasis occurs in the CLNs of TC patients.

RESUMEN

Despite laparoscopic-guided minimally invasive hepatectomy emerging as the primary approach for resecting hepatocellular carcinoma (HCC), there's still a significant gap in suitable biomaterials that seamlessly integrate with these techniques to achieve effective hemostasis and suppress residual tumors at the surgical margin. Electrospun films are increasingly used for wound closure, yet the employment of prefabricated electrospun films for hemostasis during minimally invasive HCC resection is hindered by prolonged operation times, complexity in implementation, limited visibility during surgery, and inadequate postoperative prevention of HCC recurrence. In this study, we integrated montmorillonite-iron oxide sheets into the PVP polymer framework, enhancing the resulting electrospun polyvinylpyrrolidone (PVP) /montmorillonite-iron oxide (MI) film (abbreviated as PMI) with robustness, hemostatic capability, and magnetocaloric properties. In contrast to the in vitro prefabricated electrospun films, the electrospun PMI film is designed to be formed in situ on liver wounds under laparoscopic guidance during hepatectomy. This design affords superior wound adaptability, facilitating meticulous wound closure and expeditious hemostasis, thereby simplifying the operative process and ultimately alleviating the workload of healthcare professionals. Moreover, when exposed to an alternating magnetic field, the film can efficiently ablate residual tumors, significantly augmenting the treatment efficacy of HCC. This article is protected by copyright. All rights reserved.

RESUMEN

The precise delineation of urban aquatic features is of paramount importance in scrutinizing water resources, monitoring floods, and devising water management strategies. Addressing the challenge of indistinct boundaries and the erroneous classification of shadowed regions as water in high-resolution remote sensing imagery, we introduce WaterDeep, which is a novel deep learning framework inspired by the DeepLabV3 + architecture and an innovative fusion mechanism for high- and low-level features. This methodology first creates a comprehensive dataset of high-resolution remote sensing images, then progresses through the Xception baseline network for low-level feature extraction, and harnesses densely connected Atrous Spatial Pyramid Pooling (ASPP) modules to assimilate multi-scale data into sophisticated high-level features. Subsequently, the network decoder amalgamates the elemental and intricate features and applies dual-line interpolation to the amalgamated dataset to extract aqueous formations from the remote images. Experimental evidence substantiates that WaterDeep outperforms its existing deep learning counterparts, achieving a stellar overall accuracy of 99.284%, FWIoU of 95.58%, precision of 97.562%, recall of 95.486%, and F1 score of 96.513%. It also excels in the precise demarcation of edges and the discernment of shadows cast by urban infrastructure. The superior efficacy of the proposed method in differentiating water bodies in complex urban environments has significant practical applications in real-world contexts.

RESUMEN

BACKGROUND Hepatocellular carcinoma (HCC) poses a significant threat to human life and is the most prevalent form of liver cancer. The intricate interplay between apoptosis, a common form of programmed cell death, and its role in immune regulation stands as a crucial mechanism influencing tumor metastasis. MATERIAL AND METHODS Utilizing HCC samples from the TCGA database and 61 anoikis-related genes (ARGs) sourced from GeneCards, we analyzed the relationship between ARGs and immune cell infiltration in HCC. Subsequently, we identified long non-coding RNAs (lncRNAs) associated with ARGs, using the least absolute shrinkage and selection operator (LASSO) regression analysis to construct a robust prognostic model. The predictive capabilities of the model were then validated through examination in a single-cell dataset. RESULTS Our constructed prognostic model, derived from lncRNAs linked to ARGs, comprised 11 significant lncRNAs: NRAV, MCM3AP-AS1, OTUD6B-AS1, AC026356.1, AC009133.1, DDX11-AS1, AC108463.2, MIR4435-2HG, WARS2-AS1, LINC01094, and HCG18. The risk score assigned to HCC samples demonstrated associations with immune indicators and the infiltration of immune cells. Further, we identified Annexin A5 (ANXA5) as the pivotal gene among ARGs, with it exerting a prominent role in regulating the lncRNA gene signature. Our validation in a single-cell database elucidated the involvement of ANXA5 in immune cell infiltration, specifically in the regulation of mononuclear cells. CONCLUSIONS This study delves into the intricate correlation between ARGs and immune cell infiltration in HCC, culminating in the development of a novel prognostic model reliant on 11 ARGs-associated lncRNAs. Furthermore, our findings highlight ANXA5 as a promising target for immune regulation in HCC, offering new perspectives for immune therapy in the context of HCC.

Asunto(s)

RESUMEN

BACKGROUND: Bladder prolapse is a common clinical disorder of pelvic floor dysfunction in women, and early diagnosis and treatment can help them recover. Pelvic magnetic resonance imaging (MRI) is one of the most important methods used by physicians to diagnose bladder prolapse; however, it is highly subjective and largely dependent on the clinical experience of physicians. The application of computer-aided diagnostic techniques to achieve a graded diagnosis of bladder prolapse can help improve its accuracy and shorten the learning curve. PURPOSE: The purpose of this study is to combine convolutional neural network (CNN) and vision transformer (ViT) for grading bladder prolapse in place of traditional neural networks, and to incorporate attention mechanisms into mobile vision transformer (MobileViT) for assisting in the grading of bladder prolapse. METHODS: This study focuses on the grading of bladder prolapse in pelvic organs using a combination of a CNN and a ViT. First, this study used MobileNetV2 to extract the local features of the images. Next, a ViT was used to extract the global features by modeling the non-local dependencies at a distance. Finally, a channel attention module (i.e., squeeze-and-excitation network) was used to improve the feature extraction network and enhance its feature representation capability. The final grading of the degree of bladder prolapse was thus achieved. RESULTS: Using pelvic MRI images provided by a Huzhou Maternal and Child Health Care Hospital, this study used the proposed method to grade patients with bladder prolapse. The accuracy, Kappa value, sensitivity, specificity, precision, and area under the curve of our method were 86.34%, 78.27%, 83.75%, 95.43%, 85.70%, and 95.05%, respectively. In comparison with other CNN models, the proposed method performed better. CONCLUSIONS: Thus, the model based on attention mechanisms exhibits better classification performance than existing methods for grading bladder prolapse in pelvic organs, and it can effectively assist physicians in achieving a more accurate bladder prolapse diagnosis.

Asunto(s)

RESUMEN

Aim: To explore the overall methylation changes in liver tissues during the formation of gallstones, as well as the key pathways and genes involved in the process. Methods: Reduced-representation bisulfite sequencing and RNA sequencing were conducted on the liver tissues of mice with gallstones and control normal mice. Results: A total of 8705 differentially methylated regions in CpG and 1410 differentially expressed genes were identified. The joint analysis indicated that aberrant DNA methylation may be associated with dysregulated gene expression in key pathways such as cholesterol metabolism and bile secretion. Conclusion: We propose for the first time that methylation changes in some key pathway genes in liver tissue may be involved in the formation of gallstones.

RESUMEN

Relapse and unresectability have become the main obstacle for further improving hepatocellular carcinoma (HCC) treatment effect. Currently, single therapy for HCC in clinical practice is limited by postoperative recurrence, intraoperative blood loss and poor patient outcomes. Multidisciplinary therapy has been recognized as the key to improving the long-term survival rate for HCC. However, the clinical application of HCC synthetic therapy is restricted by single functional biomaterials. In this study, a magnetic nanocomposite hydrogel (CG-IM) with iron oxide nanoparticle-loaded mica nanosheets (Iron oxide nanoparticles@Mica, IM) is reported. This biocompatible magnetic hydrogel integrated high injectability, magnetocaloric property, mechanical robustness, wet adhesion, and hemostasis, leading to efficient HCC multidisciplinary therapies including postoperative tumor margin treatment and percutaneous locoregional ablation. After minimally invasive hepatectomy of HCC, the CG-IM hydrogel can facilely seal the bleeding hepatic margin, followed by magnetic hyperthermia ablation to effectively prevent recurrence. In addition, CG-IM hydrogel can inhibit unresectable HCC by magnetic hyperthermia through the percutaneous intervention under ultrasound guidance.

Asunto(s)

RESUMEN

Background: Inhomogeneity within tumors can reflect tumor angiogenesis. Existing research into the quantization of angiogenesis mainly focuses on time-intensity curve parameters but has produced inconsistent results. In clinical work, it is difficult to achieve standardization and consistency for manual judgement of the inhomogeneity of contrast-enhanced images, while the artificial intelligence technology may be helpful. The aim of this study was to assess whether computers can assist in the artificial classification of tumor inhomogeneity in contrast-enhanced ultrasound (CEUS) images of rectal cancer. Methods: A total of 500 contrast-enhanced ultrasonograms were retrospectively collected, which was verified of rectal cancer pathologically from 2016 to 2018 as training set. All images are from 18-80 years old patients with rectal cancer in our hospital. These tumors are usually located in the middle and lower segment of the rectum, which can be completely observed on ultrasound. The images were divided into 3 categories according to the inhomogeneous distribution of contrast agents inside the tumors. Computing methods were used to simulate manual classification. Computer processing steps included segmentation, gray level quantization, dimension reduction, and classification. The results of 6 different gray level quantization, 2 dimensionality reduction methods, and 3 classifiers were compared, from which the optimal parameters were selected in each step. The performance of computer classification was evaluated using manual classification results as the reference. Ninety-seven ultrasonograms of contrast-enhanced rectal tumors were collected as validation set from 2018.1 to 2018.6. Results: The optimal gray level was set at 32. Principal component analysis (PCA) was the first choice for dimensionality reduction. The best classifier was support vector machines (SVM). The accuracy of computer classification was 87.80% (439/500). The accuracy of computer classification in the validation cohort was 60.82%. The area under the curve (AUC) of class 1, 2, and 3 were 0.76, 0.41, and 0.48, respectively. Conclusions: Results showed that the computer methods are competent for classifying inhomogeneity of contrast-enhanced rectal cancers inside ultrasonograms.

RESUMEN

This study aimed to explore the role of a stimulator of interferon (IFN) gene (STING) agonist in breast cancer (BCa) immunotherapy. Clinical samples were collected from 37 patients with BCa. A tumor-bearing mouse model was established by injecting 4T1 cells into the mammary fat pad of mice. STING agonist and atezolizumab were injected in the mice twice a week for 2 weeks. Peripheral blood, tumor mass, lung, liver, brain cortex and kidney samples of the tumor-bearing mice were collected. Anti-IFN alpha receptor subunit 1 (IFNAR1) was used to treat 4T1 cells. Tumor tissues of patients with BCa exhibited lower STING and high programmed cell death protein 1 and programmed death-ligand 1 protein expressions. The STING agonist inhibited 4T1 cell growth in mice (P < 0.001) and increased the IFN-ß level and phosphorylation of STING, TBK1, IRF3 and STAT1 in tumor mass of tumor-bearing mice (P < 0.001). It synergized with atezolizumab to inhibit 4T1 cell growth in mice and increased tumor necrosis factor-α, IFN-ß, interleukin-10 and IFN-γ levels in the peripheral blood and tumor mass (P < 0.01). It synergized with atezolizumab to increase CD8+ cytotoxic T cells and decrease FOXP3+ Treg cells in the tumor-bearing mouse model. The STING agonist was nontoxic to the lung, liver, brain cortex and kidney. Anti-IFNAR1 reversed the STING agonist promotion on TBK1, IRF3 and STAT1 phosphorylation in 4T1 cells (P < 0.01). STING agonists enhance the efficacy of atezolizumab in BCa immunotherapy by activating the IFN-ß signaling pathway.

Asunto(s)

RESUMEN

Transrectal contrast-enhanced ultrasound (CEUS) is an important examination for rectal tumors. The inhomogeneity of the CEUS images has important clinical significance. However, there is no objective method to evaluate this index. In this study, a method based on gray-level co-occurrence matrix (GLCM) is proposed to extract texture features of images and grade these images according the inhomogeneity. Specific processes include compressing the gray level of the image, calculating the texture statistics of gray level co-occurrence matrix, combining feature selection and principal component analysis (PCA) for dimensionality reduction, and training and validating quadratic discriminant analysis (QDA). After ten cross-validation, the overall accuracy rate of machine classification was 87.01%, and the accuracy of each level was as follows: Grade Ⅰ 52.94%, Grade Ⅱ 96.48% and Grade Ⅲ 92.35% respectively. The proposed method has high accuracy in judging grade Ⅱ and Ⅲ images, which can help to identify the grade of inhomogeneity of contrast-enhanced ultrasound images of rectal tumors, and may be used to assist clinical doctors in judging the grade of inhomogeneity of contrast-enhanced ultrasound of rectal tumors.

Asunto(s)

RESUMEN

STUDY OBJECTIVES: There is a lack of data regarding adherence trajectories when switching from continuous positive airway pressure (CPAP) to adaptive servoventilation (ASV) in the context of persistent or treatment-emergent central sleep apnea (CSA). This study investigated 90-day adherence rates in patients with sleep apnea based on the type of positive airway pressure (PAP) device used and any switching of PAP modality over time. METHODS: Telemonitoring data were obtained from a United States PAP database. Eligible patients were a 30% random sample who started PAP, plus all who started ASV, from January 1, 2015 to October 2, 2015. All received PAP and had at least one session with usage of 1 hour or more. Adherence and device usage were determined in three groups: started on CPAP and stayed on CPAP (CPAP only); started on ASV and stayed on ASV (ASV only); started on CPAP, switched to ASV (Switch). The United States Medicare definition of adherence was used. RESULTS: The study included 198,890 patients; 189,724 (CPAP only), 8,957 (ASV only) and 209 (Switch). In the Switch group, average apnea-hypopnea index decreased significantly on ASV versus CPAP. At 90 days, adherence rates were 73.8% and 73.2% in the CPAP only and ASV only groups. In the Switch group, CPAP adherence was 62.7%, improving to 76.6% after the switch to ASV. Mean device usage at 90 days was 5.27, 5.31, and 5.73 h/d in the CPAP only, ASV only, and Switch groups, respectively. CONCLUSIONS: Treatment-emergent or persistent CSA during CPAP reduced therapy adherence, but adherence improved early after switching from CPAP to ASV.

Asunto(s)

RESUMEN

BACKGROUND: The emergence of central sleep apnea (CSA) during positive airway pressure (PAP) therapy has been observed clinically in approximately 10% of obstructive sleep apnea titration studies. This study assessed a PAP database to investigate trajectories of treatment-emergent CSA during continuous PAP (CPAP) therapy. METHODS: U.S. telemonitoring device data were analyzed for the presence/absence of emergent CSA at baseline (week 1) and week 13. Defined groups were as follows: obstructive sleep apnea (average central apnea index [CAI] < 5/h in week 1, < 5/h in week 13); transient CSA (CAI ≥ 5/h in week 1, < 5/h in week 13); persistent CSA (CAI ≥ 5/h in week 1, ≥ 5/h in week 13); emergent CSA (CAI < 5/h in week 1, ≥ 5/h in week 13). RESULTS: Patients (133,006) used CPAP for ≥ 90 days and had ≥ 1 day with use of ≥ 1 h in week 1 and week 13. The proportion of patients with CSA in week 1 or week 13 was 3.5%; of these, CSA was transient, persistent, or emergent in 55.1%, 25.2%, and 19.7%, respectively. Patients with vs without treatment-emergent CSA were older, had higher residual apnea-hypopnea index and CAI at week 13, and more leaks (all P < .001). Patients with any treatment-emergent CSA were at higher risk of therapy termination vs those who did not develop CSA (all P < .001). CONCLUSIONS: Our study identified a variety of CSA trajectories during CPAP therapy, identifying several different clinical phenotypes. Identification of treatment-emergent CSA by telemonitoring could facilitate early intervention to reduce the risk of therapy discontinuation and shift to more efficient ventilator modalities.

Asunto(s)

RESUMEN

How do we retrieve cartoon characters accurately? Or how to synthesize new cartoon clips smoothly and efficiently from the cartoon library? Both questions are important for animators and cartoon enthusiasts to design and create new cartoons by utilizing existing cartoon materials. The first key issue to answer those questions is to find a proper representation that describes the cartoon character effectively. In this paper, we consider multiple features from different views, i.e., color histogram, Hausdorff edge feature, and skeleton feature, to represent cartoon characters with different colors, shapes, and gestures. Each visual feature reflects a unique characteristic of a cartoon character, and they are complementary to each other for retrieval and synthesis. However, how to combine the three visual features is the second key issue of our application. By simply concatenating them into a long vector, it will end up with the so-called "curse of dimensionality," let alone their heterogeneity embedded in different visual feature spaces. Here, we introduce a semisupervised multiview subspace learning (semi-MSL) algorithm, to encode different features in a unified space. Specifically, under the patch alignment framework, semi-MSL uses the discriminative information from labeled cartoon characters in the construction of local patches where the manifold structure revealed by unlabeled cartoon characters is utilized to capture the geometric distribution. The experimental evaluations based on both cartoon character retrieval and clip synthesis demonstrate the effectiveness of the proposed method for cartoon application. Moreover, additional results of content-based image retrieval on benchmark data suggest the generality of semi-MSL for other applications.

Asunto(s)

RESUMEN

This paper proposes a technique to denoise the worm artifacts of elastogram using 2-D wavelet shrinkage denoising method. Firstly, strain estimate matrix including worm artifacts was decomposed to 3 levels by 2-D discrete wavelet transform with Sym8 wavelet function, and the thresholds were obtained using Birg6-Massart algorithm. Secondly, all the high frequency coefficients on different levels were quantized by using hard threshold and soft threshold function. Finally, the strain estimate matrix was reconstructed by using the 3rd layer low frequency coefficients and other layer quantized high frequency coefficients. The simulation results illustrated that the present technique could efficiently denoise the worm artifacts, enhance the elastogram performance indices, such as elastographic signal-to-noise ratio (SNRe) and elastographic contrast-to-noise ratio (CNRe), and could increase the correlation coefficient between the denoised elastogram and the ideal elastogram. In comparison with 2-D low-pass filtering, it could also obtain the higher elastographic SNRe and CNRe, and have clearer hard lesion edge. In addition, the results demonstrated that the proposed technique could suppress worm artifacts of elastograms for various applied strains. This work showed that the 2-D wavelet shrinkage denoising could efficiently denoise the worm artifacts of elastogram and enhance the performance of elastogram.

Asunto(s)

RESUMEN

Correspondence construction of objects in key frames is the precondition for inbetweening and coloring in 2-D computer-assisted animation production. Since each frame of an animation consists of multiple layers, objects are complex in terms of shape and structure. Therefore, existing shape-matching algorithms specifically designed for simple structures such as a single closed contour cannot perform well on objects constructed by multiple contours with an open shape. This paper introduces a semisupervised patch alignment framework for complex object correspondence construction. In particular, the new framework constructs local patches for each point on an object and aligns these patches in a new feature space, in which correspondences between objects can be detected by the subsequent clustering. For local patch construction, pairwise constraints, which indicate the corresponding points (must link) or unfitting points (cannot link), are introduced by users to improve the performance of correspondence construction. This kind of input is convenient for animation software users via user-friendly interfaces. A dozen of experimental results on our cartoon data set that is built on industrial production suggest the effectiveness of the proposed framework for constructing correspondences of complex objects. As an extension of our framework, additional shape retrieval experiments on MPEG-7 data set show that its performance is comparable with that of a prominent algorithm published in T-PAMI 2009.

RESUMEN

In order to get real time ultrasound images with clear structure and improved contrast, an adaptive ultrasound sound speed optimization method based on image contrast analysis was investigated. It firstly introduced the dynamic beamforming of ultrasound system, as well as the definition of assumed system's sound speed and the true sound speed propagated in tissues the degrade image quality due to their mismatch was also discussed. After given the pixel gray level value based ultrasound image contrast ratio, the basic idea to precisely estimate the true sound speed for real time system sound speed was proposed. Algorithms have been verified both in tissue-mimicking phantoms with known sound speeds and in vivo ultrasound images, compared with other existing method. The testing results showed that this new method not only produced accurate sound speed for ultrasound image optimization, but also finely met the critical computation requirement for real time applications.

Asunto(s)

RESUMEN

In this paper the theory of autocorrelation algorithm for color flow mapping is analyzed and a new way for ultrasound color flow velocity estimation and real-time display is proposed. The method sets up a mapping table which directly relates to the dynamic display range and has only 256 entries using an inverse mapping method instead of calculating the arctangent value directly. This method is ideal for software implementation and offers an interactive way to the user for changing the dynamic range of flow velocity and thus to increasing the display resolution.

Asunto(s)