How Membrane Flexibility Impacts Permeation and Separation of Gas through Nanoporous Graphenes.

In recent years, extensive research has used molecular dynamics simulations to investigate gas separation through nanoporous graphene (NPG) membranes. However, most studies have considered graphene membranes as rigid, overlooking the impact of their inherent flexibility. This study systematically quantifies the effect of graphene flexibility on gas permeation by comparing the diffusion of various gases through flexible and rigid single-layer NPG models. The results demonstrate that flexibility notably increases permeance, particularly for gases with larger molecular diameters/pore size ratios, by allowing gas molecules greater mobility within the pore. Interestingly, the effect of flexibility boils down to the expansion of the average pore size, and the detail of the membrane's vibrational dynamics is of little importance in quantifying permeance. Our work shows that accounting for flexibility in molecular models improves the alignment of simulation results with experimental data, emphasizing the importance of considering membrane flexibility in predictive models of NPG membrane performance.

Long noncoding RNA SNHG3 interacts with microRNA-502-3p to mediate ITGA6 expression in liver hepatocellular carcinoma.

SNHG3, a long noncoding RNA (lncRNA), has been linked to poor outcomes in patients with liver hepatocellular carcinoma (LIHC). In this study, we found that SNHG3 was overexpressed in LIHC and associated with poor outcomes in patients with LIHC. Functional assays, including colony formation, spheroid formation, and in vivo assays showed that SNHG3 promoted stemness of cancer stem cells (CSC) and tumor growth in vivo by interacting with microRNA-502-3p (miR-502-3p). miR-502-3p inhibitor repressed the tumor-suppressing effects of SNHG3 depletion. Finally, by RNA pull-down, dual-luciferase reporter assay, m6A methylation level detection, and m6A-IP-qPCR assays, we found that miR-502-3p targeted YTHDF3 to regulate the translation of integrin alpha-6 (ITGA6) and targeted HBXIP to inhibit the m6A modification of ITGA6 through methyltransferase-like 3 (METTL3). Our study revealed that SNHG3 controls the YTHDF3/ITGA6 and HBXIP/METTL3/ITGA6 pathways by repressing miR-502-3p expression to sustain the self-renewal properties of CSC in LIHC.

Pluronic F127 hydrogel-loaded extracellular vesicles from adipose-derived mesenchymal stem cells promote tracheal cartilage regeneration via SCNN1B delivery.

Extracellular vesicles (EVs) derived from adipose-derived mesenchymal stem cells (AMSC-EVs) have been highlighted as a cell-free therapy due to their regenerative capability to enhance tissue and organ regeneration. Herein, we aimed to examine the mechanism of PF127-hydrogel@AMSC-EVs in promoting tracheal cartilage defect repair. Based on bioinformatics methods, SCNN1B was identified as a key gene for the osteogenic differentiation of AMSCs induced by AMSC-EVs. EVs were isolated from rat AMSCs and then loaded onto thermo-sensitive PF-127 hydrogel to develop PF127-hydrogel@AMSC-EVs. It was established that PF127-hydrogel@AMSC-EVs could effectively deliver SCNN1B into AMSCs, where SCNN1B promoted AMSC osteogenic differentiation. The promotive effect was evidenced by enhanced ALP activity, extracellular matrix mineralization, and expression of s-glycosaminoglycan, RUNX2, OCN, collagen II, PERK, and ATF4. Furthermore, the in vivo experiments revealed that PF127-hydrogel@AMSC-SCNN1B-EVs stimulated tracheal cartilage regeneration in rats through PERK/ATF4 signaling axis activation. Therefore, PF127-hydrogel@AMSC-SCNN1B-EVs may be a novel cell-free biomaterial to facilitate tracheal cartilage regeneration and cartilage injury repair.

From molecular sieving to gas effusion through nanoporous 2D graphenes: Comparison between analytical predictions and molecular simulations.

In this paper, we study the permeation of polyatomic gas molecules through 2D graphene membranes. Using equilibrium molecular dynamics simulations, we investigate the permeation of pure gas compounds (CH4, CO2, O2, N2, and H2) through nanoporous graphene membranes with varying pore sizes and geometries. Our simulations consider the recrossing mechanism, often neglected in previous studies, which has a significant effect on permeation for intermediate pore size to molecular diameter ratios. We find that the permeation process can be decoupled into two steps: the crossing process of gas molecules through the pore plane and the escaping process from the pore region to a neighboring adsorption site, which prevents recrossing. To account for these mechanisms, we use a permeance model expressed as the product of the permeance for the crossing process and the probability of molecule escape. This phenomenological model is extended to account for small polyatomic gas molecules and to describe permeation regimes ranging from molecular sieving to effusion. The proposed model captures the temperature dependence and provides insights into the key parameters of the gas/membrane interaction controlling the permeance of the system. This work lays the foundation for predicting gas permeance and exploring membrane separation factors in 2D materials such as graphene.

"Phoenix in Flight": an unique fruit morphology ensures wind dispersal of seeds of the phoenix tree (Firmiana simplex (L.) W. Wight).

BACKGROUND: Many seed plants produce winged diaspores that use wind to disperse their seeds. The morphology of these diaspores is directly related to the seed dispersal potential. The majority of winged diaspores have flat wings and only seeds; however, some angiosperms, such as Firmiana produce winged fruit with a different morphology, whose seed dispersal mechanisms are not yet fully understood. In this study, we observed the fruit development of F. simplex and determined the morphological characteristics of mature fruit and their effects on the flight performance of the fruit. RESULTS: We found that the pericarp of F. simplex dehisced early and continued to unfold and expand during fruit development until ripening, finally formed a spoon-shaped wing with multiple alternate seeds on each edge. The wing caused mature fruit to spin stably during descent to provide a low terminal velocity, which was correlated with the wing loading and the distribution of seeds on the pericarp. When the curvature distribution of the pericarp surface substantially changed, the aerodynamic characteristics of fruit during descent altered, resulting in the inability of the fruit to spin. CONCLUSIONS: Our results suggest that the curved shape and alternate seed distribution are necessary for the winged diaspore of F. simplex to stabilize spinning during wind dispersal. These unique morphological characteristics are related to the early cracking of fruits during development, which may be an adaptation for the wind dispersal of seeds.

A perspective on the diagnosis of cracked tooth: imaging modalities evolve to AI-based analysis.

Despite numerous clinical trials and pre-clinical developments, the diagnosis of cracked tooth, especially in the early stages, remains a challenge. Cracked tooth syndrome is often accompanied by dramatic painful responses from occlusion and temperature stimulation, which has become one of the leading causes for tooth loss in adults. Current clinical diagnostical approaches for cracked tooth have been widely investigated based on X-rays, optical light, ultrasound wave, etc. Advances in artificial intelligence (AI) development have unlocked the possibility of detecting the crack in a more intellectual and automotive way. This may lead to the possibility of further enhancement of the diagnostic accuracy for cracked tooth disease. In this review, various medical imaging technologies for diagnosing cracked tooth are overviewed. In particular, the imaging modality, effect and the advantages of each diagnostic technique are discussed. What's more, AI-based crack detection and classification methods, especially the convolutional neural network (CNN)-based algorithms, including image classification (AlexNet), object detection (YOLO, Faster-RCNN), semantic segmentation (U-Net, Segnet) are comprehensively reviewed. Finally, the future perspectives and challenges in the diagnosis of the cracked tooth are lighted.

Modeling gas permeation mechanisms through 2D membranes: Comparison between a phenomenological model and extensive molecular simulations.

Two-dimensional (2D) membranes based on perforated graphene have great potential in the field of separation of chemical species for a variety of applications, including gas treatment. In addition to recent experimental studies, several works simulate the mechanisms of gas permeation through this type of membrane using molecular dynamics, but few combine different techniques to ensure that their method of choice captures all relevant mechanisms. In particular, the re-crossing mechanism leading a gas molecule that has crossed the plane of the membrane to rapidly re-cross it in the opposite direction has never been documented. In this work, we study gas permeation through a simplified 2D membrane model. We combine equilibrium and non-equilibrium molecular dynamics simulations to quantify the impact of these re-crossing mechanisms on the values of the computed transport coefficients. Using non-equilibrium simulations as reference, we show that the equilibrium simulation techniques commonly used can lead to a significant overestimation of the transport properties of the membrane. We propose a simple method to probe the re-crossing dynamics during equilibrium simulations, making it possible to compute correct values of the transport coefficient without the need for non-equilibrium simulations. Furthermore, by analyzing the phenomenology observed in the simulations, we derive an analytical formula for the permeance that takes the form of an Arrhenius law with a non-trivial temperature dependent prefactor. In excellent agreement with our simulation results, this model provides a simple theoretical framework that captures the main mechanisms involved in gas permeation through 2D membranes, including the effect of re-crossing.

KDM5B promotes self-renewal of hepatocellular carcinoma cells through the microRNA-448-mediated YTHDF3/ITGA6 axis.

Histone methylation plays important roles in mediating the onset and progression of various cancers, and lysine-specific demethylase 5B (KDM5B), as a histone demethylase, is reported to be an oncogene in hepatocellular carcinoma (HCC). However, the mechanism underlying its tumorigenesis remains undefined. Hence, we explored the regulatory role of KDM5B in HCC cells, aiming to identify novel therapeutic targets for HCC. Gene Expression Omnibus database and StarBase were used to predict important regulatory pathways related to HCC. Then, the expression of KDM5B and microRNA-448 (miR-448) in HCC tissues was detected by RT-qPCR and Western blot analysis. The correlation between KDM5B and miR-448 expression was analysed by Pearson's correlation coefficient and ChIP experiments, and the targeting of YTH N6-methyladenosine RNA binding protein 3 (YTHDF3) by miR-448 was examined by luciferase assay. Additionally, the effect of KDM5B on the proliferation, migration, invasion and apoptosis as well as tumorigenicity of transfected cells was assessed using ectopic expression and depletion experiments. KDM5B was highly expressed in HCC cells and was inversely related to miR-448 expression. KDM5B demethylated H3K4me3 on the miR-448 promoter and thereby inhibited the expression of miR-448, which in turn targeted YTHDF3 and integrin subunit alpha 6 (ITGA6) to promote the malignant phenotype of HCC. Moreover, KDM5B accelerated HCC progression in nude mice via the miR-448/YTHDF3/ITGA6 axis. Our study uncovered that KDM5B regulates the YTHDF3/ITGA6 axis by inhibiting the expression of miR-448 to promote the occurrence of HCC.

KPNA2 interaction with CBX8 contributes to the development and progression of bladder cancer by mediating the PRDM1/c-FOS pathway.

BACKGROUND: Bladder cancer (BCa) is a common malignancy characterized by high heterogeneity, yet the current treatment modalities are limited. The aim of the present investigation was to unravel the functional role of Karyopherin alpha 2 (KPNA2), a tumor facilitator identified in multiple malignancies, in the progression of BCa. METHODS: BCa tissues and adjacent normal tissues were surgically resected and analyzed from patients with BCa to determine the expression profile of KPNA2 and Chromobox 8 (CBX8) by RT-qPCR, Western blot analysis and immunohistochemistry. The relationship among KPNA2, CBX8 and PR domain zinc finger protein 1 (PRDM1) was explored by co-immunoprecipitation and chromatin-immunoprecipitation. The functions of KPNA2, CBX8 and PRDM1 on BCa cell proliferation, migration and invasion were evaluated. Next, a nude mouse model of BCa was established for validating the roles of KPNA2, CBX8 and PRDM1 in vivo. RESULTS: KPNA2 and CBX8 were highly expressed in BCa and are in association with dismal oncologic outcomes of patients with BCa. KPNA2 promoted nuclear import of CBX8. CBX8 downregulated PRDM1 by recruiting BCOR in the promoter region of PRDM1. Overexpression of KPNA2 promoted the malignant behaviors of BCa cells, which was counteracted by silencing of CBX8. Overexpressing PRDM1 attenuated the progression of BCa by inhibiting c-FOS expression. The tumor-promoting effects of KPNA2 via the PRDM1/c-FOS pathway were also validated in vivo. CONCLUSION: Collectively, our findings attached great importance to the interplay between KPNA2 and CBX8 in BCa in mediating the development and progression of BCa, thus offering a promising candidate target for better BCa patient management.

Three-dimensional Dirac material anode enables concentrated solar thermionic converters.

Concentrated solar thermionic converters (CSTCs) are proposed by using three-dimensional (3D) Dirac material (DM) as the novel anode, significantly improving device performance. A theoretical model is developed to investigate the optimal performance of CSTCs. Under a solar concentration of 500, the CSTC reaches a maximum conversion efficiency of 11.8%. Furthermore, the optimal working condition and parametric selection criteria of the CSTC parameters under different solar concentrations are determined. Importantly, we demonstrate that 3D DM as an anode has more advantages over metal and graphene in CSTCs. This work thus offers a theoretical foundation for the exploration of solar thermionic application in using 3D DMs as novel electrodes.

A method of crack detection based on digital image correlation for simulated cracked tooth.

BACKGROUND: Early clinical cracked tooth can be a perplexing disorder to diagnose and manage. One of the key problems for the diagnosis of the cracked tooth is the detection of the location of the surface crack. METHODS: This paper proposes an image-based method for the detection of the micro-crack in the simulated cracked tooth. A homemade three-axis motion platform mounted with a telecentric lens was built as an image acquisition system to observe the surface of the simulated cracked tooth, which was under compression with a magnitude of the masticatory force. By using digital image correlation (DIC), the deformation map for the crown surface of the cracked tooth was calculated. Through image analysis, the micro-crack was quantitatively visualized and characterized. RESULTS: The skeleton of the crack path was successfully extracted from the image of the principal strain field, which was further verified by the image from micro-CT. Based on crack kinematics, the crack opening displacement was quantitatively calculated to be 2-10 µm under the normal mastication stress, which was in good agreement with the value reported in the literature. CONCLUSIONS: The crack on the surface of the simulated cracked tooth could be detected based on the proposed DIC-based method. The proposed method may provide a new solution for the rapid clinical diagnosis of cracked teeth and the calculated crack information would be helpful for the subsequent clinical treatment of cracked teeth.

Comparative Assessment of Various Low-Dissipation Combined Models for Three-Terminal Heat Pump Systems.

Thermally driven heat pump systems play important roles in the utilization of low-grade thermal energy. In order to evaluate and compare the performances of three different constructions of thermally driven heat pump and heat transformer, the low-dissipation assumption has been adopted to establish the irreversible thermodynamic models of them in the present paper. By means of the proposed models, the heating loads, the coefficients of performance (COPs) and the optimal relations between them for various constructions are derived and discussed. The performances of different constructions are numerically assessed. More importantly, according to the results obtained, the upper and lower bounds of the COP at maximum heating load for different constructions are generated and compared by the introduction of a parameter measuring the deviation from the reversible limit of the system. Accordingly, the optimal constructions for the low-dissipation three-terminal heat pump and heat transformer are determined within the frame of low-dissipation assumption, respectively. The optimal constructions in accord with previous research and engineering practices for various three-terminal devices are obtained, which confirms the compatibility between the low-dissipation model and endoreversible model and highlights the validity of the application of low-dissipation model for multi-terminal thermodynamic devices. The proposed models and the significant results obtained enrich the theoretical thermodynamic model of thermally driven heat pump systems and may provide some useful guidelines for the design and operation of realistic thermally driven heat pump systems.

An embedded gene selection method using knockoffs optimizing neural network.

BACKGROUND: Gene selection refers to find a small subset of discriminant genes from the gene expression profiles. How to select genes that affect specific phenotypic traits effectively is an important research work in the field of biology. The neural network has better fitting ability when dealing with nonlinear data, and it can capture features automatically and flexibly. In this work, we propose an embedded gene selection method using neural network. The important genes can be obtained by calculating the weight coefficient after the training is completed. In order to solve the problem of black box of neural network and further make the training results interpretable in neural network, we use the idea of knockoffs to construct the knockoff feature genes of the original feature genes. This method not only make each feature gene to compete with each other, but also make each feature gene compete with its knockoff feature gene. This approach can help to select the key genes that affect the decision-making of neural networks. RESULTS: We use maize carotenoids, tocopherol methyltransferase, raffinose family oligosaccharides and human breast cancer dataset to do verification and analysis. CONCLUSIONS: The experiment results demonstrate that the knockoffs optimizing neural network method has better detection effect than the other existing algorithms, and specially for processing the nonlinear gene expression and phenotype data.

microRNA-448 inhibits stemness maintenance and self-renewal of hepatocellular carcinoma stem cells through the MAGEA6-mediated AMPK signaling pathway.

Hepatocellular carcinoma (HCC) occurs mainly in patients with chronic liver disease and cirrhosis. Increasing evidence has identified the involvement of microRNAs (miRNAs) acting as essential regulators in the progression of HCC. As predicted by microarray analysis, miR-448 might potentially affect HCC progression by regulating the melanoma-associated antigen (MAGEA). Therefore, the present investigation focused on exploring whether or not miR-448 and MAGEA6 were involved in the self-renewal and stemness maintenance of HCC stem cells. The interaction among miR-448, MAGEA6, and the AMPK signaling pathway was evaluated. It was noted that miR-448 targeted and downregulated MAGEA6, thus activating the AMP-activated protein kinase (AMPK) signaling pathway in HCC. Furthermore, for the purpose of exploring the functional relevance of MAGEA6 and miR-448 on the sphere formation, colony formation, and invasion and migration of HCC stem cells, the CD133+ CD44 + HCC stem cells were sorted and treated with the mimic or inhibitor of miR-448, small interfering RNA (siRNA) against MAGEA6 or an AMPK activator AICAR. MAGEA6 silencing or miR-448 overexpression was demonstrated to inhibit the abilities of sphere formation, colony formation, cell migration, and invasion of HCC cells. Afterwards, a rescue experiment was conducted and revealed that MAGEA6 silencing reversed the effects of miR-448 inhibitor on stemness maintenance and self-renewal of HCC stem cells. Finally, after the in vivo experiment was carried out, miR-448 was observed to restrain the tumor formation and stemness in vivo. Altogether, miR-448 activates the AMPK signaling pathway by downregulating MAGEA6, thus inhibiting the stemness maintenance and self-renewal of HCC stem cells, which identifies miR-448 as a new therapeutic strategy for HCC.

Functional rs6265 polymorphism in the brain-derived neurotrophic factor gene confers protection against neurocognitive dysfunction in posttraumatic stress disorder among Chinese patients with hepatocellular carcinoma.

Posttraumatic stress disorder (PTSD) is a psychiatric disorder that plagues trauma survivors. Evidence shows that brain-derived neurotrophic factor (BDNF) may be involved in the occurrence and development of PTSD. Here we tried to demonstrate whether BDNF gene polymorphisms are correlated with neurocognitive function following PTSD in patients with hepatocellular carcinoma (HCC). This study included 102 patients with HCC complicated with PTSD, 146 HCC patients, and 152 healthy volunteers. Initially, we evaluated the neurocognitive function of the study subjects. Next, we measured BDNF G11757C and rs6265 polymorphisms by polymerase chain reaction-restriction fragment length polymorphism. The correlation of BDNF polymorphisms and BDNF level with HCC complicated with PTSD was evaluated. The results revealed that HCC complicated with PTSD showed decreased serum BDNF level and Mini-mental state examination (MMSE) score. Serum BDNF level of HCC and HCC complicated with PTSD was positively correlated with MMSE score. GA + AA allele and A allele of rs6265 increased the risk of PTSD among patients with HCC. GA and AA genotypes of rs6265 were correlated with the decreased MMSE score of HCC complicated with PTSD. Haplotype GA of rs6265 and G11757C increased the risk of PTSD for HCC, while haplotype CG decreased this risk. Lastly, the logistic regression analysis suggested that low BDNF level was a contributor to HCC complicated with PTSD, while GG genotype of rs6265 served as a protective factor. Collectively, this study defines the GG genotype of BDNF rs6265 polymorphism as a protector to HCC complicated with PTSD. In addition, these results provided a promising target for PTSD prevention in patients with HCC.

Mycolactone-producing Mycobacterium marinum infection in captive Hong Kong warty newts and pathological evidence of impaired host immune function.

A mass mortality event of captive Hong Kong warty newts Paramesotriton hongkongensis with non-granulomatous necrotic lesions occurred in Taipei Zoo, Taiwan, in 2014. Clinically, the sick newts were lethargic and often covered with water mold Saprolegnia sp. on the skin of the body trunk or extremities. Predominant pathological findings were multifocal non-granulomatous necrotic lesions in the liver, spleen, and kidneys and severe skin infection with Saprolegnia sp., with deep invasion and involvement of underlying muscles. The possibility of ranavirus infection was ruled out by negative PCR results. Unexpectedly, abundant intralesional acid-fast positive bacilli were found in the necrotic lesions of the liver, spleen, and kidney in all 14 sick newts. PCR targeting the hsp65, ITS region, and partial 16S rRNA genes was performed, and the sequence identity from amplified amplicons of hsp65 and partial 16S rRNA genes was 100% identical to that of the corresponding gene fragment of Mycobacterium marinum. Further molecular investigations demonstrated that the current M. marinum was a mycolactone-producing mycobacterium with the presence of esxA/esxB genes. Mycolactone is a plasmid-encoded, immunosuppressive, and cytotoxic toxin. The possible immunosuppression phenomenon characterized by systemic non-granulomatous necrotic lesions caused by M. marinum and the unusual deep invasive infection caused by water mold might be associated with the immunosuppressive effect of mycolactone. Therefore, it should be noted that non-granulomatous necrotic lesions in amphibians can be caused not only by ranavirus infection but also by mycobacteriosis.

Executive function in cancer patients with posttraumatic stress disorder.

Background Cancer patients with posttraumatic stress disorder can lead to their noncompliant behaviors. However, less is known about the neurocognitive functioning of posttraumatic stress disorder in general cancer types or patient populations. The current study attempted to examine the prevalence of posttraumatic stress disorder and their relationships with executive function in individuals with cancer. Methods A total of 285 cancer patients with posttraumatic stress disorder and 150 healthy individuals were recruited for the present study. The Clinician Administered PTSD Scale, Tower of Hanoi, Wisconsin Card Sorting Test, and Wechsler Adult Intelligence Scale-Revised Chinese revision were administered to all participants. Results Significant differences in the score of Tower of Hanoi, Wisconsin Card Sorting Test, and Wechsler Adult Intelligence Scale-Revised Chinese revision were observed between the posttraumatic stress disorder group and the healthy control group ( p < 0.001). Significant correlations were found between all posttraumatic stress disorder symptoms and executive function. Conclusions These findings suggest that individuals with cancer-related posttraumatic stress disorder exhibit more severe impairment in executive function than healthy controls do.

Application of smart hydrogel materials in cartilage injury repair: A systematic review and meta-analysis.

OBJECTIVE: Cartilage injury is a common clinical condition, and treatment approaches have evolved over time from traditional conservative and surgical methods to regenerative repair. In this context, hydrogels, as widely used biomaterials in the field of cartilage repair, have garnered significant attention. Particularly, responsive hydrogels (also known as "smart hydrogels") have shown immense potential due to their ability to respond to various physicochemical properties and environmental changes. This paper aims to review the latest research developments of hydrogels in cartilage repair, utilizing a more systematic and comprehensive meta-analysis approach to evaluate the research status and application value of responsive hydrogels. The goal is to determine whether these materials demonstrate favorable therapeutic effects for subsequent clinical applications, thereby offering improved treatment methods for patients with cartilage injuries. METHOD: This study employed a systematic literature search method to summarize the research progress of responsive hydrogels by retrieving literature on the subject and review studies. The search terms included "hydrogel" and "cartilage," covering data from database inception up to October 2023. The quality of the literature was independently evaluated using Review Manager v5.4 software. Quantifiable data was statistically analyzed using the R language. RESULTS: A total of 7 articles were retrieved for further meta-analysis. In the quality assessment, the studies demonstrated reliability and accuracy. The results of the meta-analysis indicated that responsive hydrogels exhibit unique advantages and effective therapeutic outcomes in the field of cartilage repair. Subgroup analysis revealed potential influences of factors such as different types of hydrogels and animal models on treatment effects. CONCLUSION: Responsive hydrogels show significant therapeutic effects and substantial application potential in the field of cartilage repair. This study provides strong scientific evidence for their further clinical applications and research, with the hope of promoting advancements in the treatment of cartilage injuries.

Semantic segmentation for tooth cracks using improved DeepLabv3+ model.

Objective: Accurate and prompt detection of cracked teeth plays a critical role for human oral health. The aim of this paper is to evaluate the performance of a tooth crack segmentation model (namely, FDB-DeepLabv3+) on optical microscopic images. Method: The FDB-DeepLabv3+ model proposed here improves feature learning by replacing the backbone with ResNet50. Feature pyramid network (FPN) is introduced to fuse muti-level features. Densely linked atrous spatial pyramid pooling (Dense ASPP) is applied to achieve denser pixel sampling and wider receptive field. Bottleneck attention module (BAM) is embedded to enhance local feature extraction. Results: Through testing on a self-made hidden cracked tooth dataset, the proposed method outperforms four classical networks (FCN, U-Net, SegNet, DeepLabv3+) on segmentation results in terms of mean pixel accuracy (MPA) and mean intersection over union (MIoU). The network achieves an increase of 11.41% in MPA and 12.14% in MIoU compared to DeepLabv3+. Ablation experiments shows that all the modifications are beneficial. Conclusion: An improved network is designed for segmenting tooth surface cracks with good overall performance and robustness, which may hold significant potential in computer-aided diagnosis of cracked teeth.

Fetal Ultrasound Standard Plane Detection With Coarse-to-Fine Multi-Task Learning.

The ultrasound standard plane plays an important role in prenatal fetal growth parameter measurement and disease diagnosis in prenatal screening. However, obtaining standard planes in a fetal ultrasound video is not only laborious and time-consuming but also depends on the clinical experience of sonographers to a certain extent. To improve the acquisition efficiency and accuracy of the ultrasound standard plane, we propose a novel detection framework that utilizes both the coarse-to-fine detection strategy and multi-task learning mechanism for feature-fused images. First, traditional manually-designed features and deep learning-based features are fused to obtain low-level shared features, which can enhance the model's feature expression ability. Inspired by the process of human recognition, ultrasound standard plane detection is divided into a coarse process of plane type classification and a fine process of standard-or-not detection, which is implemented via an end-to-end multi-task learning network. The region-of-interest area is also recognised in our detection framework to suppress the influence of a variable maternal background. Extensive experiments are conducted on three ultrasound planes of the first-class fetal examination, i.e., the femur, thalamus, and abdomen ultrasound images. The experiment results show that our method outperforms competing methods in terms of accuracy, which demonstrates the efficacy of the proposed method and can reduce the workload of sonographers in prenatal screening.