FARN: Fetal Anatomy Reasoning Network for Detection with Global Context Semantic and Local Topology Relationship.

Accurate recognition of fetal anatomical structure is a pivotal task in ultrasound (US) image analysis. Sonographers naturally apply anatomical knowledge and clinical expertise to recognizing key anatomical structures in complex US images. However, mainstream object detection approaches usually treat each structure recognition separately, overlooking anatomical correlations between different structures in fetal US planes. In this work, we propose a Fetal Anatomy Reasoning Network (FARN) that incorporates two kinds of relationship forms: a global context semantic block summarized with visual similarity and a local topology relationship block depicting structural pair constraints. Specifically, by designing the Adaptive Relation Graph Reasoning (ARGR) module, anatomical structures are treated as nodes, with two kinds of relationships between nodes modeled as edges. The flexibility of the model is enhanced by constructing the adaptive relationship graph in a data-driven way, enabling adaptation to various data samples without the need for predefined additional constraints. The feature representation is further enhanced by aggregating the outputs of the ARGR module. Comprehensive experimental results demonstrate that FARN achieves promising performance in detecting 37 anatomical structures across key US planes in tertiary obstetric screening. FARN effectively utilizes key relationships to improve detection performance, demonstrates robustness to small-scale, similar, and indistinct structures, and avoids some detection errors that deviate from anatomical norms. Overall, our study serves as a resource for developing efficient and concise approaches to model inter-anatomy relationships.

HFSCCD: A Hybrid Neural Network for Fetal Standard Cardiac Cycle Detection in Ultrasound Videos.

In the fetal cardiac ultrasound examination, standard cardiac cycle (SCC) recognition is the essential foundation for diagnosing congenital heart disease. Previous studies have mostly focused on the detection of adult CCs, which may not be applicable to the fetus. In clinical practice, localization of SCCs needs to recognize end-systole (ES) and end-diastole (ED) frames accurately, ensuring that every frame in the cycle is a standard view. Most existing methods are not based on the detection of key anatomical structures, which may not recognize irrelevant views and background frames, results containing non-standard frames, or even it does not work in clinical practice. We propose an end-to-end hybrid neural network based on an object detector to detect SCCs from fetal ultrasound videos efficiently, which consists of 3 modules, namely Anatomical Structure Detection (ASD), Cardiac Cycle Localization (CCL), and Standard Plane Recognition (SPR). Specifically, ASD uses an object detector to identify 9 key anatomical structures, 3 cardiac motion phases, and the corresponding confidence scores from fetal ultrasound videos. On this basis, we propose a joint probability method in the CCL to learn the cardiac motion cycle based on the 3 cardiac motion phases. In SPR, to reduce the impact of structure detection errors on the accuracy of the standard plane recognition, we use XGBoost algorithm to learn the relation knowledge of the detected anatomical structures. We evaluate our method on the test fetal ultrasound video datasets and clinical examination cases and achieve remarkable results. This study may pave the way for clinical practices.

Bidirectional two-sample Mendelian randomization analysis identifies causal associations between cardiovascular diseases and frozen shoulder.

BACKGROUND: Although prior observational studies indicate an association between cardiovascular diseases (CVDs) and frozen shoulder (FS), the potential causal relationship between them remains uncertain. This study aims to explore the genetic causal relationship between CVDs and FS using Mendelian randomization (MR). METHODS: Genetic variations closely associated with FS were obtained from the FinnGen Consortium. Summary data for CVD, including atrial fibrillation (AF), coronary artery disease (CAD), heart failure (HF), myocardial infarction (MI), stroke, and ischemic stroke (IS), were sourced from several large-scale genome-wide association studies (GWAS). MR analysis was performed using inverse variance weighting (IVW), MR Egger, and weighted median methods. IVW, as the primary MR analysis method, complemented by other sensitivity analyses, was utilized to validate the robustness of the results. Further reverse MR analysis was conducted to explore the presence of reverse causal relationships. RESULTS: In the forward MR analysis, genetically determined risk of stroke and IS was positively associated with FS (OR [95% CI] = 1.58 (1.23-2.03), P < 0.01; OR [95% CI] = 1.46 (1.16-1.85), P < 0.01, respectively). There was no strong evidence of an effect of genetically predicted other CVDs on FS risk. Sensitivity analyses confirmed the robustness of the results. In the reverse MR analysis, no causal relationships were observed between FS and various CVDs. CONCLUSION: The study suggests that stroke increases the risk of developing FS. However, further basic and clinical research is needed to substantiate our findings.

MLMSeg: A multi-view learning model for ultrasound thyroid nodule segmentation.

Accurate segmentation of the thyroid gland in ultrasound images is an essential initial step in distinguishing between benign and malignant nodules, thus facilitating early diagnosis. Most existing deep learning-based methods to segment thyroid nodules are learned from only a single view or two views, which limits the performance of segmenting nodules at different scales in complex ultrasound scanning environments. To address this limitation, this study proposes a multi-view learning model, abbreviated as MLMSeg. First, a deep convolutional neural network is introduced to encode the features of the local view. Second, a multi-channel transformer module is designed to capture long-range dependency correlations of global view between different nodules. Third, there are semantic relationships of structural view between features of different layers. For example, low-level features and high-level features are endowed with hidden relationships in the feature space. To this end, a cross-layer graph convolutional module is proposed to adaptively learn the correlations of high-level and low-level features by constructing graphs across different layers. In addition, in the view fusion, a channel-aware graph attention block is devised to fuse the features from the aforementioned views for accurate segmentation of thyroid nodules. To demonstrate the effectiveness of the proposed method, extensive comparative experiments were conducted with 14 baseline methods. MLMSeg achieved higher Dice coefficients (92.10% and 83.84%) and Intersection over Union scores (86.60% and 73.52%) on two different thyroid datasets. The exceptional segmentation capability of MLMSeg for thyroid nodules can greatly assist in localizing thyroid nodules and facilitating more precise measurements of their transverse and longitudinal diameters, which is of significant clinical relevance for the diagnosis of thyroid nodules.

A knowledge-interpretable multi-task learning framework for automated thyroid nodule diagnosis in ultrasound videos.

Ultrasound has become the most widely used modality for thyroid nodule diagnosis, due to its portability, real-time feedback, lack of toxicity, and low cost. Recently, the computer-aided diagnosis (CAD) of thyroid nodules has attracted significant attention. However, most existing techniques can only be applied to either static images with prominent features (manually selected from scanning videos) or rely on 'black boxes' that cannot provide interpretable results. In this study, we develop a user-friendly framework for the automated diagnosis of thyroid nodules in ultrasound videos, by simulating the typical diagnostic workflow used by radiologists. This process consists of two orderly part-to-whole tasks. The first interprets the characteristics of each image using prior knowledge, to obtain corresponding frame-wise TI-RADS scores. Associated embedded representations not only provide diagnostic information for radiologists but also reduce computational costs. The second task models temporal contextual information in an embedding vector sequence and selectively enhances important information to distinguish benign and malignant thyroid nodules, thereby improving the efficiency and generalizability of the proposed framework. Experimental results demonstrated this approach outperformed other state-of-the-art video classification methods. In addition to assisting radiologists in understanding model predictions, these CAD results could further ease diagnostic workloads and improve patient care.

Appraising the causal relationship between thyroid function and rheumatoid arthritis: a two-sample bidirectional Mendelian randomization study.

Background: Hypothyroidism and hyperthyroidism are observationally associated with rheumatoid arthritis (RA), but causality is unclear. To evaluate the causal relationship between thyroid function and RA, we conducted a two-Sample bidirectional Mendelian Randomization (MR) study. Methods: Single nucleotide polymorphisms associated with six phenotypes were selected from the FinnGen biobank database, The ThyroidOmics Consortium database, and the IEU Open GWAS database. For the forward MR analysis, we selected hypothyroidism (N=213,390), Graves' disease (GD) (N=199,034), other types of hyperthyroidism (N=190,799), free thyroxine (FT4, N=49,269), and thyroid-stimulating hormone (TSH, N=54,288) as the five related thyroid function phenotypes for exposure, with RA (N=58,284) as the outcome. Reverse MR analysis selected RA as the exposure and five phenotypes of thyroid function as the outcome. The Inverse variance weighting (IVW) method was used as the primary analysis method, supplemented by weighted median (WM) and MR-Egger methods. Cochran's Q test, MR-PRESSO, MR-Egger regression methods, and leave-one-out analysis were employed to assess sensitivity and pleiotropy. Results: Forward MR evidence indicates that genetic susceptibility to hypothyroidism is associated with an increased risk of RA (ORIvw=1.758, P=7.61×10-5). Reverse MR evidence suggests that genetic susceptibility to RA is associated with an increased risk of hypothyroidism (ORIvw=1.274, P=3.88×10-20), GD (ORIvw=1.269, P=8.15×10-05), and other types of hyperthyroidism (ORIvw=1.141, P=1.80×10-03). There is no evidence to support a forward or reverse causal relationship between genetic susceptibility to RA and FT4, TSH. Conclusion: Our results provide genetic evidence supporting bidirectional causal relationships between thyroid function and RA. These findings inform preventive strategies and interventions targeting RA and thyroid dysfunction.

TransFSM: Fetal Anatomy Segmentation and Biometric Measurement in Ultrasound Images Using a Hybrid Transformer.

Biometric parameter measurements are powerful tools for evaluating a fetus's gestational age, growth pattern, and abnormalities in a 2D ultrasound. However, it is still challenging to measure fetal biometric parameters automatically due to the indiscriminate confusing factors, limited foreground-background contrast, variety of fetal anatomy shapes at different gestational ages, and blurry anatomical boundaries in ultrasound images. The performance of a standard CNN architecture is limited for these tasks due to the restricted receptive field. We propose a novel hybrid Transformer framework, TransFSM, to address fetal multi-anatomy segmentation and biometric measurement tasks. Unlike the vanilla Transformer based on a single-scale input, TransFSM has a deformable self-attention mechanism so it can effectively process multi-scale information to segment fetal anatomy with irregular shapes and different sizes. We devised a BAD to capture more intrinsic local details using boundary-wise prior knowledge, which compensates for the defects of the Transformer in extracting local features. In addition, a Transformer auxiliary segment head is designed to improve mask prediction by learning the semantic correspondence of the same pixel categories and feature discriminability among different pixel categories. Extensive experiments were conducted on clinical cases and benchmark datasets for anatomy segmentation and biometric measurement tasks. The experiment results indicate that our method achieves state-of-the-art performance in seven evaluation metrics compared with CNN-based, Transformer-based, and hybrid approaches. By Knowledge distillation, the proposed TransFSM can create a more compact and efficient model with high deploying potential in resource-constrained scenarios. Our study serves as a unified framework for biometric estimation across multiple anatomical regions to monitor fetal growth in clinical practice.

Appraising causal risk and protective factors for rheumatoid arthritis.

Aims: Mendelian randomization (MR) is considered to overcome the bias of observational studies, but there is no current meta-analysis of MR studies on rheumatoid arthritis (RA). The purpose of this study was to summarize the relationship between potential pathogenic factors and RA risk based on existing MR studies. Methods: PubMed, Web of Science, and Embase were searched for MR studies on influencing factors in relation to RA up to October 2022. Meta-analyses of MR studies assessing correlations between various potential pathogenic factors and RA were conducted. Random-effect and fixed-effect models were used to synthesize the odds ratios of various pathogenic factors and RA. The quality of the study was assessed using the Strengthening the Reporting of Observational Studies in Epidemiology using Mendelian Randomization (STROBE-MR) guidelines. Results: A total of 517 potentially relevant articles were screened, 35 studies were included in the systematic review, and 19 studies were eligible to be included in the meta-analysis. Pooled estimates of 19 included studies (causality between 15 different risk factors and RA) revealed that obesity, smoking, coffee intake, lower education attainment, and Graves' disease (GD) were related to the increased risk of RA. In contrast, the causality contribution from serum mineral levels (calcium, iron, copper, zinc, magnesium, selenium), alcohol intake, and chronic periodontitis to RA is not significant. Conclusion: Obesity, smoking, education attainment, and GD have real causal effects on the occurrence and development of RA. These results may provide insights into the genetic susceptibility and potential biological pathways of RA.

The METS-IR is independently related to bone mineral density, FRAX score, and bone fracture among U.S. non-diabetic adults: a cross-sectional study based on NHANES.

AIM: The purpose of this study was to investigate the association between the metabolic score for insulin resistance (METS-IR) and bone mineral density (BMD) in American non-diabetic adults. METHODS: We conducted a cross-sectional study with 1114 non-diabetic adults from the National Health and Nutrition Examination Survey cycle (2013-2014). The associations between METS-IR and BMD of total femur and spine were assessed by the multiple linear regression and verified the non-linear relationship with a smooth curve fit and threshold effect model. Furthermore, we evaluated the relationship between METS-IR, FRAX score, and history of bone fractures. RESULTS: We found that BMD of the total femur and spine increased by 0.005 g/cm3 and 0.005 g/cm3, respectively, for a one-unit increase of METS-IR in all participants. This positive association was more pronounced among higher METS-IR participants, and there was a non-linear relationship, which was more significant when the MTTS-IRfemur was < 41.62 or the METS-IRspine was < 41.39 (ßfemur = 0.008, ßspine = 0.011, all P < 0.05). We also found that METS-IR was positively correlated with both FRAX scores in all female participants. However, METS-IR was positively correlated only with the 10-year hip fracture risk score in male participants with fractures. No significant association between METS-IR and a history of bone fractures. CONCLUSIONS: In American non-diabetic adults, there is a correlation between elevated levels of METS-IR within the lower range and increased BMD as well as decreased risk of fractures, suggesting that METS-IR holds promise as a novel biomarker for guiding osteoporosis (OP) prevention. However, it is important to carefully balance the potential benefits and risks of METS-IR in OP.

Deep endpoints focusing network under geometric constraints for end-to-end biometric measurement in fetal ultrasound images.

Biometric measurements in fetal ultrasound images are one of the most highly demanding medical image analysis tasks that can directly contribute to diagnosing fetal diseases. However, the natural high-speckle noise and shadows in ultrasound data present big challenges for automatic biometric measurement. Almost all the existing dominant automatic methods are two-stage models, where the key anatomical structures are segmented first and then measured, thus bringing segmentation and fitting errors. What is worse, the results of the second-stage fitting are completely dependent on the good performance of first-stage segmentation, i.e., the segmentation error will lead to a larger fitting error. To this end, we propose a novel end-to-end biometric measurement network, abbreviated as E2EBM-Net, that directly fits the measurement parameters. E2EBM-Net includes a cross-level feature fusion module to extract multi-scale texture information, a hard-soft attention module to improve position sensitivity, and center-focused detectors jointly to achieve accurate localizing and regressing of the measurement endpoints, as well as a loss function with geometric cues to enhance the correlations. To our knowledge, this is the first AI-based application to address the biometric measurement of irregular anatomical structures in fetal ultrasound images with an end-to-end approach. Experiment results showed that E2EBM-Net outperformed the existing methods and achieved the state-of-the-art performance.

The role of NSAID in mediating the effect of genetically predicted major depressive disorder on osteomyelitis: A Mendelian randomization study.

BACKGROUND: Osteomyelitis and major depressive disorder (MDD) are significant health concerns with potential interconnections. However, the underlying mechanisms linking these conditions remain unknown. This study aimed to investigate the potential mediating role of non-steroidal anti-inflammatory drug (NSAID) medication in the association between MDD and the risk of osteomyelitis. METHODS: We utilized summary data from large-scale genome-wide association studies (GWAS) to perform Mendelian randomization (MR) mediation analysis. Instrumental variables were selected based on genome-wide significance, and instrumental strength was assessed using F-statistics. Univariable and multivariable MR analyses were conducted to estimate causal effects and proportions mediated by NSAID medication. RESULTS: The univariable MR analysis revealed significant associations between MDD and osteomyelitis (odds ratio [OR] = 1.44, 95 % confidence interval [CI]: 1.18-1.874) and between MDD and NSAID medication (OR = 1.36, 95 % CI 1.24-1.49). In the multivariable MR analysis, the direct effect of MDD on osteomyelitis was OR 1.35 (95 % CI: 1.09, 1.67) after adjusting for NSAID medication. The proportion of mediation by NSAID medication was 23 % (95 % CI: 0.05 %, 38.6 %). CONCLUSION: This MR study provides evidence for a genetically predicted causal association between MDD, NSAID medication, and osteomyelitis. The findings emphasize the need for a comprehensive approach in managing individuals with comorbid depression and osteomyelitis, considering the potential risks and benefits of NSAID medication. Future research should address limitations and explore additional mediators and confounding factors to enhance understanding of this complex relationship.

Bidirectional two-sample mendelian randomization analysis identifies causal associations between age at menarche and sarcopenia-related traits.

BACKGROUND: Observational studies have shown that the age of menarche is associated with sarcopenia, but confounding factors make the causal relationship difficult to infer. OBJECTIVE: Therefore, we conducted a bidirectional two-sample Mendelian randomized (MR) analysis to evaluate the potential causal relationship between age at menarche and sarcopenia-related traits (hand grip strength, lean mass, walking pace). METHODS: We obtained the latest aggregate statistics from the Genome-wide association studies (GWAS) database on the age of menarche of 182,416 participants from ReproGen, the appendicular lean mass of 244,730 participants from EMBL's European Bioinformatics Institute, the left-hand grip strength of 401,026 participants, the right-hand grip strength of 461,089 participants and the usual walking pace of 459,915 participants from the UK Biobank. The inverse variance weighting (IVW) method and other MR methods were used to evaluate the bidirectional causal relationship between the age of menarche and sarcopenia. RESULTS: The forward MR results showed that the age of menarche predicted by the gene was positively correlated with left-hand grip strength (IVWß=0.041, P = 2.00 × 10-10), right-hand grip strength (IVWß=0.053, P = 1.97 × 10-18), appendicular lean mass (IVWß=0.012, P = 4.38 × 10-13) and usual walking pace (IVWß=0.033, P = 1.62 × 10-8).In the reverse MR analysis, we also found that the usual walking pace was positively correlated with the age of menarche predicted by genes (IVWß=0.532, P = 1.65 × 10-4). Still, there was no causal relationship between grip strength and appendicular lean mass and the age at menarche. CONCLUSION: Our results show that earlier menarche will increase the risk of sarcopenia. In addition, people with higher muscle function tend to have menarche later. These findings may provide a reference for prevention strategies and interventions for menarche in advance and sarcopenia.

Causal effects of tea intake on multiple types of fractures: A two-sample Mendelian randomization study.

Fracture is a global public health disease. Bone health and fracture risk have become the focus of public and scientific attention. Observational studies have reported that tea consumption is associated with fracture risk, but the results are inconsistent. The present study used 2-sample Mendelian randomization (MR) analysis. The inverse variance weighted method, employing genetic data from UK Biobank (447,485 cases) of tea intake and UK Biobank (Genome-wide association study Round 2) project (361,194 cases) of fractures, was performed to estimate the causal relationship between tea intake and multiple types of fractures. The inverse variance weighted indicated no causal effects of tea consumption on fractures of the skull and face, shoulder and upper arm, hand and wrist, femur, calf, and ankle (odds ratio = 1.000, 1.000, 1.002, 0.997, 0.998; P = .881, 0.857, 0.339, 0.054, 0.569, respectively). Consistent results were also found in MR-Egger, weighted median, and weighted mode. Our research provided evidence that tea consumption is unlikely to affect the incidence of fractures.

The metabolic score of insulin resistance is positively correlated with bone mineral density in postmenopausal patients with type 2 diabetes mellitus.

The prevalence of type 2 diabetes mellitus (T2DM) complicated with osteoporosis (OP) is increasing yearly. Early prevention, detection and treatment of OP are important in postmenopausal patients with T2DM. This study aimed to explore the correlation between insulin resistance and bone mineral density (BMD), and OP in postmenopausal patients with T2DM. In this study, postmenopausal patients with T2DM who visited our hospital from January 2021 to March 2022 were divided into the OP group (n = 91) and non-OP group (n = 119) according to whether they were complicated with OP or not. The general data of patients, BMD, blood routine, glucose metabolism, lipid metabolism, liver and kidney function indexes were collected, and the homeostatic model assessment for IR (HOMA-IR), the triglyceride-glucose (TyG) index and the metabolic score for IR (METS-IR) were calculated. A weighted multivariate linear regression model assessed the correlation between insulin resistance (IR) related indexes and lumbar spine, femoral neck, and hip BMD. A weighted logistic regression model assessed the odds ratios (ORs) and 95% confidence intervals (95% CIs) for the association between the IR-related indexes and OP risk. The nonlinear relationship was also evaluated by smooth curve fitting (SCF) and a weighted generalized additive model (GAM). Moreover, the Receiver-operating characteristics (ROC) curve was used to analyze the predictive efficiency of METS-IR in postmenopausal patients with T2DM with OP. HOMA-IR, TyG, and METS-IR in the OP group were lower than those in the non-OP group (all P < 0.05). Weighted multiple linear regression after adjusting covariates showed that METS-IR was positively correlated with the lumbar spine, femoral neck, and hip BMD (ßMETS-IR = 0.006,0.005,0.005, all P < 0.001). The results of weighted Logistic regression and GAM showed that when METS-IR < 44.5, each unit of increased METS-IR value was associated with a decreased OP risk of 12% (P = 0.002). When METS-IR ≥ 44.5, there was no significant correlation between METS-IR and the risk of OP (OR = 1.00, P = 0.934). Similar trends were not observed in HOMA-IR and TyG. The ROC suggested helpful discriminative power of the METS-IR index for T2DM. We confirmed that METS-IR, as a novel alternative marker of IR, had a positive association with BMD in postmenopausal patients with T2DM, and METS-IR was a protective factor for OP in a specific range.

Effects of vitamin D deficiency on blood lipids and bone metabolism: a large cross-sectional study.

To investigate the relationship between serum high-density lipoprotein (HDL-C) and spinal bone mineral density (BMD) under different serum 25-hydroxyvitamin D (25 (OH) D) levels in adults over 40 years old and to explore its mechanism. We include participants over the age of 40 with data on HDL-C, 25 (OH) D, spinal BMD, and other variables in the National Health and Nutrition Examination Survey 2007-2010 in the analysis. A weighted multiple linear regression model was used to evaluate the association between serum HDL-C and spinal BMD in different gender, ages, and serum 25 (OH) D levels. A total of 3599 subjects aged ≥ 40 years old were included in this study. Univariate analysis of the complete correction model showed a negative correlation between serum HDL-C and spinal BMD. In the two subgroups of serum 25 (OH) D, we found that the higher the serum HDL-C in the female with serum 25 (OH) D < 75 nmol/L aged 40-59 years old, the lower the total spinal BMD, and a similar relationship was found in the lumbar spine. However, no similar relationship was found in all populations with serum 25 (OH) D ≥ 75 nmol/L and males with serum 25 (OH) D < 75 nmol/L. These results suggest that among Americans over the age of 40, the increase in serum HDL-C is related to decreased BMD of spine only in women aged 40-59 years with vitamin D insufficiency or deficiency.

HN-PPISP: a hybrid network based on MLP-Mixer for protein-protein interaction site prediction.

MOTIVATION: Biological experimental approaches to protein-protein interaction (PPI) site prediction are critical for understanding the mechanisms of biochemical processes but are time-consuming and laborious. With the development of Deep Learning (DL) techniques, the most popular Convolutional Neural Networks (CNN)-based methods have been proposed to address these problems. Although significant progress has been made, these methods still have limitations in encoding the characteristics of each amino acid in protein sequences. Current methods cannot efficiently explore the nature of Position Specific Scoring Matrix (PSSM), secondary structure and raw protein sequences by processing them all together. For PPI site prediction, how to effectively model the PPI context with attention to prediction remains an open problem. In addition, the long-distance dependencies of PPI features are important, which is very challenging for many CNN-based methods because the innate ability of CNN is difficult to outperform auto-regressive models like Transformers. RESULTS: To effectively mine the properties of PPI features, a novel hybrid neural network named HN-PPISP is proposed, which integrates a Multi-layer Perceptron Mixer (MLP-Mixer) module for local feature extraction and a two-stage multi-branch module for global feature capture. The model merits Transformer, TextCNN and Bi-LSTM as a powerful alternative for PPI site prediction. On the one hand, this is the first application of an advanced Transformer (i.e. MLP-Mixer) with a hybrid network for sequence-based PPI prediction. On the other hand, unlike existing methods that treat global features altogether, the proposed two-stage multi-branch hybrid module firstly assigns different attention scores to the input features and then encodes the feature through different branch modules. In the first stage, different improved attention modules are hybridized to extract features from the raw protein sequences, secondary structure and PSSM, respectively. In the second stage, a multi-branch network is designed to aggregate information from both branches in parallel. The two branches encode the features and extract dependencies through several operations such as TextCNN, Bi-LSTM and different activation functions. Experimental results on real-world public datasets show that our model consistently achieves state-of-the-art performance over seven remarkable baselines. AVAILABILITY: The source code of HN-PPISP model is available at https://github.com/ylxu05/HN-PPISP.

A Hybrid Two-Stage Teaching-Learning-Based Optimization Algorithm for Feature Selection in Bioinformatics.

The "curse of dimensionality" brings new challenges to the feature selection (FS) problem, especially in bioinformatics filed. In this paper, we propose a hybrid Two-Stage Teaching-Learning-Based Optimization (TS-TLBO) algorithm to improve the performance of bioinformatics data classification. In the selection reduction stage, potentially informative features, as well as noisy features, are selected to effectively reduce the search space. In the following comparative self-learning stage, the teacher and the worst student with self-learning evolve together based on the duality of the FS problems to enhance the exploitation capabilities. In addition, an opposition-based learning strategy is utilized to generate initial solutions to rapidly improve the quality of the solutions. We further develop a self-adaptive mutation mechanism to improve the search performance by dynamically adjusting the mutation rate according to the teacher's convergence ability. Moreover, we integrate a differential evolutionary method with TLBO to boost the exploration ability of our algorithm. We conduct comparative experiments on 31 public data sets with different data dimensions, including 7 bioinformatics datasets, and evaluate our TS-TLBO algorithm compared with 11 related methods. The experimental results show that the TS-TLBO algorithm obtains a good feature subset with better classification performance, and indicates its generality to the FS problems.

A YOLOX-based Deep Instance Segmentation Neural Network for Cardiac Anatomical Structures in Fetal Ultrasound Images.

Echocardiography is an essential procedure for the prenatal examination of the fetus for congenital heart disease (CHD). Accurate segmentation of key anatomical structures in a four-chamber view is an essential step in measuring fetal growth parameters and diagnosing CHD. Currently, most obstetricians perform segmentation tasks manually, but the pixel-level operation is labor-intensive and requires extensive anatomical knowledge and clinical experience. As such, efficiently and accurately detecting structures from real-world fetal ultrasound images is a key challenge. In this paper, we propose a YOLOX-based deep instance segmentation neural network (i.e., IS-YOLOX) for cardiac anatomical structure location and segmentation in fetal ultrasound images. Specifically, we reconstruct a new instance segmentation branch based on a multi-task deep learning framework. We then design a new multi-level non-maximum suppression (NMS) mechanism to further improve the segmentation performance that consists of three levels of selection. Moreover, unlike two-stage instance segmentation approaches, our method does not rely on object detection results. To the best of our knowledge, this is the first study regarding instance segmentation on 13 types of anatomical structures in the fetal four-chamber view. Extensive experiments were carried out on clinical datasets, and the experimental results show that our method outperforms nine competitive baselines.

Self-reported and genetically predicted effects of coffee intake on rheumatoid arthritis: Epidemiological studies and Mendelian randomization analysis.

Background and aims: Causal research concerning coffee intake and rheumatoid arthritis (RA) risk is controversial. The objective of this study was to further explore the causal relationship between coffee intake and RA risk. Methods: The 4,310 participants from NHANES 2003-2006 were included in an epidemiological study to assess the association between coffee intake and RA by weighted multivariate logistic regression. The inverse variance weighted (IVW) method of two-sample Mendelian randomization (MR), employing genetic data from UK Biobank (428,860 cases) of coffee intake and MR-Base platform (14,361 cases and 43,923 controls) of RA, was performed to estimate the causal relationship between coffee intake and RA. Results: Weighted multivariate logistic regression suggested no significant correlation between coffee intake and RA. Compared to the no-coffee group, the odds ratio for RA in the <1, 1-3, ≥4 cups/day group were 1.297, 1.378, and 1.125 (P = 0.204, 0.098, and 0.698, respectively). In the IVW of MR analysis, there was no causal relationship between coffee intake and RA (OR = 1.47, P = 0.218). Conclusion: Our study did not support a causal association between coffee intake and RA risk. However, it is necessary to consider valid information on coffee intake, including brewing method, type of coffee, and quantity, in further analysis of coffee intake and RA.

Comparison of dynamic and static spacers for the treatment of infections following total knee replacement: a systematic review and meta-analysis.

BACKGROUND: Revision surgery is the most common treatment for patients who develop infection after total knee arthroplasty (TKA). Two types of spacers are often used in revision surgery: dynamic spacers and static spacers. The comparative efficacy of these two types of spacers on knee prosthesis infections is not well established. Therefore, we carried out a systematic evaluation and meta-analysis with the aim of comparing the difference in efficacy between dynamic and static spacers. METHODS: We conducted the literature search in PubMed, Web of Science, Cochrane Library, and Embase databases. The articles searched were clinical study comparing the difference in efficacy between dynamic spacers and static spacers for the treatment of prosthetic infections occurring after total knee arthroplasty. RESULTS: We conducted a literature search and screening based on the principles of PICOS. Ultimately, 14 relevant clinical studies were included in our current study. We use infection control rate as the primary evaluation indicator. The KSS knee scores (KSSs), KSS functional scores, bone loss and range of motion (ROM) are secondary indicators of evaluation. Thirteen of these included studies reported the infection control rates, with no significant difference between dynamic and static shims (RR: 1.03; 95% Cl 0.98, 1.09; P = 0.179 > 0.05). The KSSs were reported in 10 articles (RR: 5.98; 95% CI 0.52, 11.43; P = 0.032 < 0.05). Six articles reported the KSS functional scores (RR: 13.90; 95% CI 4.95, 22.85; P = 0.02 < 0.05). Twelve articles reported the ROM (RR: 17.23. 95% CI 10.18, 24.27; P < 0.0001). Six articles reported the bone loss (RR: 2.04; 95% CI 1.11, 3.77; P = 0.022 < 0.05). CONCLUSION: Current evidence demonstrates that dynamic spacers are comparable to static spacers in controlling prosthetic joint infection. In terms of improving the functional prognosis of the knee joint, dynamic spacers are more effective than static spacers.