Analysis Model Research Articles

Tennis Player Momentum Analysis and Match Flow Prediction Model Based on PCA

Momentum is a fluctuating phenomenon that occurs during the course of a game when the seemingly dominant side scores consecutive points or influences the course of the entire game. And as tennis is a popular sport all over the world, exploring the effects and mechanisms of athletes' momentum on match changes and outcomes in tennis matches can provide reasonable match advice to athletes. To this end, this paper systematically establishes a prediction model based on tennis players' momentum analysis and match direction. In this paper, this text first cleaned the data for the 2023 Wimbledon tennis tournament obtained from the American Collegiate Mathematical Modeling Contest website and supplemented the missing values using cubic Hermite interpolation. The processed data is then analyzed using principal component analysis, where a linear combination of each principal component is defined as a momentum value, thus quantifying the momentum and creating a mathematical model between the indicator and the momentum. Finally, a momentum plot of the player is created to visualize the relationship between the player's momentum and the outcome of the race.

Association of lipid target achievement and cardiovascular outcomes following statin initiation: a population-based cohort study

Abstract Introduction Non-high density lipoprotein cholesterol (non-HDL-C) is increasingly incorporated into clinical practice guidelines along with low density lipoprotein cholesterol (LDL-C) to guide lipid-lowering therapy decisions. We aimed to examine patterns of LDL-C and non-HDL-C levels after statin initiation for primary prevention of cardiovascular disease (CVD), and their association with incident CV events. Methods We conducted a population-based cohort study of Ontario, Canada residents age ≥66 years who initiated a statin for primary prevention between January 1, 2012 and December 31, 2019. For those surviving 1-year after statin initiation, we determined the proportion with a lipid panel in the 1-year period after a statin was started. For those with a lipid panel, we used the lipid panel closest to the end of the 1-year period to categorize individuals based on the LDL-C and non-HDL-C thresholds in the 2021 Canadian Cardiovascular Society dyslipidemia guidelines (Table 1). We stratified by diabetes/chronic kidney disease (CKD) status. We evaluated the association between LDL-C/non-HDL-C category and a composite of all-cause mortality or hospitalization for myocardial infarction, stroke or revascularization (primary outcome). We also investigated the association between each category and each component of the primary outcome. The index date was the 365 days after a statin was initiated and follow-up was until December 31, 2020. We used a cause-specific hazards model for analysis, adjusted for clinical and sociodemographic confounders. Results Our cohort comprised 173,127 people. In the 1-year period after statin initiation, 72% of people had a follow-up lipid panel performed. Median follow-up after that was 2.5 years. Compared with those meeting both LDL-C and non-HDL-C targets, being above both targets was associated with an increased rate of the primary outcome for people without diabetes/CKD (HR 1.10, 95% CI 1.05 to 1.15) and those with diabetes/CKD (HR 1.16, 95% CI 1.09 to 1.23). Being at LDL-C target but above non-HDL-C target was associated with an increased rate of the primary outcome for people with diabetes/CKD (HR 1.16, 95% CI 1.03 to 1.30) but not for those without diabetes/CKD (HR 1.05, 95% CI 0.93 to 1.17). Conclusions In this population-based cohort, having an LDL-C and non-HDL-C above guideline-recommended targets after statin initiation was associated with an increased rate of adverse outcomes. These findings support the residual risk associated with incompletely controlled LDL-C or non-HDL-C levels after initiating statin therapy for primary prevention.Primary outcome associationsComponents of primary outcome

Economic evaluation of Polygenic Risk Scores in healthcare interventions: a systematic review

Abstract Background Polygenic Risk Scores (PRS) quantify genetic susceptibility to diseases, promising personalized healthcare. This review explores current evidence regarding the economic evaluation of strategies based on PRS or other polygenic risk stratification approaches, scrutinizing their methodologies. Methods The study protocol was registered in PROSPERO (CRD42023442780). A systematic search in PubMed, Scopus, and Web of Science identified full economic evaluations of intervention based on polygenic risk stratification strategies. The quality of the included articles was assessed using the Drummond checklist. This study was supported by the EC and MUR under PNRR - M4C2-I1.3 Project PE_00000019 ‘HEAL ITALIA”. Results Nineteen articles were included in the analysis, with oncological conditions being the most frequently investigated (13), followed by cardiovascular conditions (3). In nearly 80% of the studies, PRS was employed for screening interventions, with the general population being the primary target in 14 out of the 19 studies. All the economic analysis models investigated cost-utility in terms of QALYs, with Markov models and microsimulations being the most common structures. The majority of these models were based on simulated cohorts derived mostly from North American or European data, with 9 adopting a healthcare system perspective and 6 a societal perspective. Although delivery strategies for PRS testing were rarely addressed, PRS costs were included in nearly all studies. Indirect costs were examined in fewer than half of the studies. In 12 out of 19 studies, the conclusions claimed the cost utility of PRS involving strategies. Conclusions Despite the technique’s potential, evaluations of various strategies based on PRS approaches yield heterogeneous conclusions regarding cost-utility. Our study highlights the factors contributing to this heterogeneity and underscores the need for further exploration, ideally prioritizing real-world data. Key messages • PRS strategies evaluations vary in cost-utility conclusions, highlighting the need for further investigation. • Real-world data and practical implications should be prioritised in PRS research.

Lipids as the link between central obesity and diabetes: perspectives from mediation analysis

BackgroundCentral obesity is a well-recognized risk factor for diabetes, yet the potential role of lipids in the diabetes risk associated with central obesity remains unclear. This study aimed to explore the possible mediating role of 11 lipid parameters [high-density lipoprotein cholesterol (HDL-C), total cholesterol (TC), non-high-density lipoprotein cholesterol (Non-HDL-C), low-density lipoprotein cholesterol (LDL-C), triglycerides (TG), remnant cholesterol (RC), and ratios of Non-HDL-C/HDL-C, RC/HDL-C, LDL/HDL-C, TG/HDL-C, TC/HDL-C] in the association of central obesity with diabetes risk.MethodsWe utilized data from 15,453 participants in the NAGALA longitudinal cohort to assess the association of baseline central obesity indicators [waist-height ratio (WHtR), waist circumference (WC)] and the 11 lipid parameters with diabetes risk. Mediation analysis models were constructed to explore the mediating role of lipid parameters in the association of WC/WHtR with diabetes.ResultsConfirmatory associative analysis using multivariable Cox regression showed that, except for Non-HDL-C, TC and LD-C, the remaining eight lipid parameters were significantly associated with WC/WHtR and diabetes risk. Mediation analysis indicated that TG, RC, HDL-C, and lipid ratios such as Non-HDL-C/HDL-C ratio, RC/HDL-C ratio, TG/HDL-C ratio, TC/HDL-C ratio and LDL/HDL-C ratio are potential lipids affecting the diabetes risk related to central obesity. Among these, the RC/HDL-C ratio seemed to contribute the most in the WC/WHtR-related diabetes risk association, with a mediation percentage of about 37%. Additionally, lipid ratio parameters appeared to play a more mediating role in the association of central obesity-related diabetes risk than individual lipids.ConclusionsIn central obesity-related diabetes risk, most lipids, especially lipid ratio parameters, play a significant mediating role. Given these findings, we advocate for increased efforts in multifactorial risk monitoring and joint management of diabetes. The evaluation of lipids, particularly lipid ratio parameters, may be holds substantial value in the prevention and management of diabetes risk under close monitoring of central obesity.

A Short-term PV Power Prediction and Uncertainty Analysis Model Based on CEEMDAN and AHA-BP

The stochastic and intermittent nature of photovoltaic (PV) generation brings a series of scheduling problems to the power system. An effective prediction of PV power is essential to minimize the impact of uncertainty. Therefore, this paper presents an integrated prediction model with complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), the artificial hummingbird algorithm (AHA), and the BP neural network (BPNN) for predicting power generation from PV power plants, and a methodology for uncertainty analysis by using the nonparametric kernel density estimation (NPKDE). First, one month of PV power is decomposed into an array of components using CEEMDAN. Then, the weights and thresholds of the BPNN are optimized by using AHA. These components are trained using the BPNN. Finally, the final prediction results are obtained by superimposing the components, and NPKDE is employed to compute the probability density and confidence interval of the prediction error. The proposed prediction method demonstrates superior predictive performance in comparison with other models. Also, the NPKDE approach better describes the accuracy of the probability density distribution.

Leveraging Social Media and Deep Learning for Sentiment Analysis for Smart Governance: A Case Study of Public Reactions to Educational Reforms in Saudi Arabia

The Saudi government’s educational reforms aim to align the system with market needs and promote economic opportunities. However, a lack of credible data makes assessing public sentiment towards these reforms challenging. This research develops a sentiment analysis application to analyze public emotional reactions to educational reforms in Saudi Arabia using AraBERT, an Arabic language model. We constructed a unique Arabic dataset of 216,858 tweets related to the reforms, with 2000 manually labeled for public sentiment. To establish a robust evaluation framework, we employed random forests, support vector machines, and logistic regression as baseline models alongside AraBERT. We also compared the fine-tuned AraBERT Sentiment Classification model with CAMeLBERT, MARBERT, and LLM (GPT) models. The fine-tuned AraBERT model had an F1 score of 0.89, which was above the baseline models by 5% and demonstrated a 4% improvement compared to other pre-trained transformer models applied to this task. This highlights the advantage of transformer models specifically trained for the target language and domain (Arabic). Arabic-specific sentiment analysis models outperform multilingual models for this task. Overall, this study demonstrates the effectiveness of AraBERT in analyzing Arabic sentiment on social media. This approach has the potential to inform educational reform evaluation in Saudi Arabia and potentially other Arabic-speaking regions.

Dampak Istri Sebagai Pencari Nafkah Utama Terhadap Penerapan Fungsi Keluarga di Kelurahan Cupak Tangah Kota Padang

This study aims to describe the impact of wives as the main breadwinners on the implementation of family functions in Cupak Tangah Village, Padang City. The theory used in this research is structural functionalism proposed by Robert K. Merton. This research uses a qualitative research approach with a descriptive type. Selection of informants in this study is using purposive sampling technique with a total of 24 informants. The types of data used are primary data and secondary data. Data collection methods are carried out in three events, namely observation, interviews, and document studies. The unit of analysis in this research is the individual. Data analysis used with the interactive analysis model proposed by Milles and Huberman, which consists of (1) data collection (2) data reduction (3) data presentation (4) conclusion drawing. The results of this study indicate that when the wife is the breadwinner in this family, it will have an impact on family functions, which can be seen in three aspects, namely: (1) the impact on the husband consists of a) economic function: work, income / income. b) affective function: providing affection, emotional bonds (communication). (2) the impact on children consists of a) affective function does not work: less attention given, lack of interaction with children b) educational socialization function: less time to guide learning c) recreational function: joint activities at home, activities outside the home d) educating children to be independent (3) the impact on the wife herself: domestic work becomes neglected, less time to rest.

Cardio Policy Optimiser: an integrated public health approach to enhancing cardiovascular health

Abstract Background Cardiovascular diseases (CVDs) remain a leading cause of mortality in Europe, with atherosclerotic-related conditions accounting for 80% of CVD deaths - a third of them premature. Despite extensive efforts over the past five decades, the decline in these deaths over the last 20-25 years has plateaued, highlighting the need for innovative approaches. Additionally, the co-occurrence of CVDs with other chronic conditions such as diabetes, hypertension, and renal and hepatic diseases, adds complexity to their management and prevention. Methods In Portugal, the pilot phase implemented a novel integrated approach combining a Bayesian epidemiological model - which incorporates genetic, sociodemographic, comorbidity, and lifestyle data - with an adaptive conjoint analysis model focusing on atherosclerosis policy intervention. These models are part of a comprehensive response framework that incorporates health outcomes and associated costs resulting from better prevention and diagnosis. Results Currently in the development and calibration stage, the model is expected to yield initial results by early September. Conclusions This modeling approach offers new opportunities for public health to intervene at early stages in the natural history of CVDs. It emphasizes enhancing prevention and early diagnosis to potentially reduce the incidence of acute episodes and associated mortalities. Moreover, the model facilitates rapid evaluation of the economic benefits derived from preventive policies, thereby supporting more informed public health decision-making. Key messages • The CPO model combines epidemiological and policy models to address cardiovascular morbidity and mortality. • It boosts early diagnosis and prevention, and assesses economic benefits from policies.

Validity of the Multigroup Ethnic Identity Measure for Māori, Pacific, Asian & European adolescents

Abstract Background Ethnic identity is important for positive health and wellbeing outcomes, especially for Indigenous youth and adolescents from minoritised ethnic groups. The Multigroup Ethnic Identity Measure (MEIM) is widely used to examine ethnic identity as a general phenomenon across ethnic groups. However, evidence regarding its validity for adolescents from different ethnic backgrounds is mainly limited to the US context. This study investigated psychometric properties of the MEIM, for the first time, within a large ethnically diverse population-based sample in New Zealand. Methods We used data from the Growing Up in New Zealand study. Participants were 4500 12-year-olds and included 22.4% Māori (the Indigenous people of New Zealand), 16.7% Pacific, 14.8% Asian and 51.9% European young people. 45.7% were cisgender boy, 37.4% were cisgender girl, and 16.4% were non-binary, trans or unsure. We conducted factor analysis for the 12-item MEIM. Results Confirmatory factor analysis model fit tended to be best for a solution with two factors representing two ethnic identity subcomponents of Exploration and Affirmation/Belonging. A single ‘ethnic identity’ factor showed a slightly weaker model fit. Exploratory factor analysis revealed a 2-factor structure with a slightly different item composition as compared to the original MEIM subscales. The findings were largely comparable across ethnic groups. Conclusions The MEIM appears to be a valid measure for Māori, Pacific, Asian and European young people. However, nuances may exist due to unique contexts including structural factors, societal norms and challenges, opportunities and access to cultural engagement. Ethnic identity is strongly linked to health and wellbeing including quality of life, self-esteem and life satisfaction. Therefore, we recommend acknowledging nuances of ethnic identity during an important time of development within diverse cultural contexts, by applying subscales and subgroup analyses where possible. Key messages • The Multigroup Ethnic Identity Measure is valid and appropriate to use at age 12-years among diverse ethnic groups including Māori, Pacific, Asian and European. • Nuances of ethnic identity should be acknowledged as an important factor for health and wellbeing, especially during critical times of transition for adolescents within diverse cultural contexts.

Pitfalls in the assessment of higher education learning gains in Brazil through multilevel regression: handling the ceiling effect.

This work assesses the validity of the Brazilian Indicator of the Difference between Observed and Expected Performances (IDD), which applies a multilevel regression model to rank university courses. It proposes a methodology to test for inconsistencies and research for the causes. With this objective, we develop an experimental methodology, applied to a database with 4,432 production engineering students distributed over 329 undergraduate courses, from the government’s National Institute for Educational Studies and Research. We display a graphical analysis of the expected and achieved grades in final graduation exams, the gap in which determines the learning gains for every course. Then we apply the multilevel linear regression model of the IDD in a particular scenario, to test for consistency in value-added learning measurements using the technique reductio ad absurdum. The result reveals that the present methodology for the calculation of the IDD is inadequate and inconsistent, and that this can be explained by the ceiling effect. It needs to be radically revised to prevent unfair judgements. Finally, we propose and implement a production frontier technique to correct for the ensuing ceiling effect, namely a variable returns to scale data envelopment analysis model for adjusting the measurements using evidence based benchmarks.

An Innovative Public Health Solution with European Features for an American Problem

Abstract Problem The Eastern Shore of Maryland is a beautiful region that boasts a multicultural community of 455,000 year-round residents. Despite its rural location, the coastal lifestyle attracts retirees, making it the state’s oldest population. Unfortunately, the region also has the highest socioeconomic disparities in the state, which creates difficulties in accessing healthcare. This is a significant public health issue that results in costly barriers to preventive care and timely, accounting for $320 billion in unnecessary healthcare spending. Description of the problem Maryland’s healthcare funding formulas favor larger health systems in urban areas, leading to health inequities. To address this, we propose an innovative scenario analysis model partially based on European public health approaches. This model will adjust funding formulas based on each region’s specific needs and challenges, promoting fairness and equity in healthcare funding. Results Adopting this population health management approach would have built-in efficiencies: In the United States, healthcare spending tends to be higher among underserved populations due to delayed care, access challenges, missed diagnoses, and limited access to the latest scientific advances as well as proper preventive services. Lessons The proposed scenario-based public health funding approach targeting the needs of the underserved population on the Eastern Shore of Maryland could yield significant positive health outcomes and health savings generating a ripple effect across the state, and perhaps even the country. The lessons learned in how other countries handle public health funding could lead to global collaboration. Key messages • Our approach aims to provide a practical solution to unnecessary healthcare spending. • We must adapt to the changing tides to stay on course.

Attractiveness of tourist destinations and collaborative tourism governance on Revisit Intention as a support for the tourism economy of the Paropo area of Dairi Regency with Tourist Experience as an Intervening Variable

The challenge of tourism is how to create tourists who are loyal to tourist destinations because every corner of tourism has an impact on the economy for the business community in the tourist destination area, if tourist visitors enjoy their tour and have the intention to return to visit the tour, it will have an impact on increasing the economy of the community located in tourist destinations. This research will be conducted in Sumatra Utaran precisely at the location of the paropo tourist spot in Dairi Regency with a population of paropo tourist visitors who have visited at least twice as many as 230 tourists, this research approach is quantitative with the Structural Equation Modeling (SEM) model. This approach is able to test a model that is most ideal in estimating endogenous variables, the sampling technique in this study uses Non-probability sampling where sampling does not provide opportunities or opportunities for each element or member to be selected as a sample The analysis model used to test the hypothesis in this study is multiple regression analysis using the Smart PLS application Directly that the attractiveness of tourist destinations, collaborative tourism governance and tourist experience have a significant effect on Revisit Intention as a support for the tourism economy of the Paropo area of Dairi Regency. Directly that the attractiveness of tourist destinations and collaborative tourism governance have a significant effect on tourist experience as a support for the tourism economy of the Paropo area of Dairi Regency. Indirectly, tourist experience has a significant role in mediating the attractiveness of tourist destinations and collaborative tourism governance on revisit intension as a support for the tourism economy of the Paropo area of Dairi Regency

Structural design and biomechanical analysis of a combined titanium and polyetheretherketone cage based on PE-PLIF fusion.

In lumbar spinal fusion, the titanium cage tends to cause stress shielding due to their high elastic modulus, which can lead to degenerative lesions in adjacent spinal segments. Furthermore, polyetheretherketone (PEEK) cages have certain material characteristics that do not promote bone ingrowth and fusion stability. In this study, a new cage was designed, and its biomechanical performance in percutaneous endoscopic posterior lumbar interbody fusion (PE-PLIF) was analyzed using the finite element (FE) method. A complete model of the L4-L5 lumbar spine was established, and static and harmonic vibration FE analysis models were developed based on it. The biomechanical properties of titanium, PEEK, and combined cage in PE-PLIF fusion were compared. The strain capacity of the combined fusion increased by 9.5% when compared to the titanium fusion. The surgical model under the combined fusion reduces the L5 endplate stress by 12% in the forward flexion condition and the fusion stress by 17% in the vibration condition compared to the model supported by the titanium fusion, and the differences in screw stress and mobility among the three models are not significant in multiple conditions. Consequently, the combined cage demonstrates a certain reduction in the stress-shielding effect when compared to the titanium cage; it has better fusion effect and provides theoretical support and guidance for the design of the clinical fusion cage.

The change of paradigm: high HDL-c levels could not mean a long lifespan

Abstract Background It is known that a low HDL-c level is associated with a high risk of overall mortality. Consequently, it was presumed that achieving higher levels of high-density lipoprotein cholesterol (HDL-c) would reduce the risk or even extend the lifespan. However, clinical trials with medical intervention failed to find clinical benefits. Aim We aimed to investigate the association between (HDL-c) levels and overall mortality using a prospective cohort from Portugal. Methods A prospective sub-study (2001-2022) was performed in a cohort of 1718 participants without coronary artery disease (CAD) of the control arm of the ongoing GENEMACOR study. The participants <65 years were followed during a mean follow-up of 7.0±5.7 years. All demographic, biochemical, and clinical data were collected. Participants were stratified into six levels of HDL-c. Kaplan-Meier estimated differences in the survival probability of each HDL-c level. Cox proportional analysis models (unadjusted and adjusted) were used to assess the association between HDL-c levels and overall mortality. Results The HDL-c levels with the lowest overall mortality rate were 60 to 70 mg/dl, set as the reference level. Total HDL-c strata were associated with overall mortality with a U-shaped relationship: in the control group, the first level (<40 mg/dl) was independently associated with increased risk of overall mortality (HR of 3.533; 95% CI: 1.738-7.178; p<0.0001). This risk drops and falls to the reference level (the lower risk) and rises again to an HR of 3.875 (95% CI: 1.472-10.201; p=0.006) in the very high HDL-c levels (≥80 mg/dl). Conclusions Our results, based on a large Portuguese cohort, suggest that very high HDL-c levels in the normal population may not represent a good prognosis, especially in women. Future research may confirm these findings and investigate the underlying mechanisms.

Lack of incremental prognostic value of triglyceride glucose index beyond coronary computed tomography angiography features for major events

This study was aim to determine the prognostic value of triglyceride-glucose (TyG) index and coronary computed tomography angiography (CTA) features for major adverse cardiovascular events (MACE). In addition, we investigate the incremental prognostic value of TyG index beyond coronary CTA features in patients with suspected or known coronary artery disease (CAD). The present study ultimately includes 3528 patients who met the enrollment criteria. The TyG index was calculated based on measured levels of triglycerides and fasting blood glucose. Primary combined endpoint consisted of MACE, which defined as myocardial infraction (MI), all-cause mortality and stroke. Three multivariate Cox proportional hazard regression models were performed to assess the association between TyG index and MACE. C-statistic was performed to assess the discriminatory value of models. 212 (6.0%) patients developed MACE during a median follow-up of 50.4 months (IQR, 39.4–55.1). TyG index remained to be a significantly and independent risk factors for predicting MACE after adjusting by different models (clinical variables alone or plus coronary CTA features) in multivariable analysis. Both the addition of TyG index to clinical model plus Coronary Artery Disease Reporting and Data System (CAD-RADS) and to clinical model plus CAD-RADS 2.0 slightly but not significantly increased the C-statistic index (0.725 vs. 0.721, p = 0.223; 0.733 vs. 0.731, p = 0.505). TyG index was associated with an increased risk of MACE. However, no incremental prognostic benefit of TyG index over CAD-RADS or CAD-RADS 2.0 was detected for MACE in patients with suspected or known CAD.

Sentiment analysis of Algerian Arabic dialect on social media Using Bi-LSTM recurrent neural networks

This paper presents a sentiment analysis approach using Bidirectional Long Short-Term Memory (Bi-LSTM) Recurrent Neural Networks to train predictive models for sentiment analysis on social media, particularly focusing on Algerian Arabic Dialect. The method leverages word-to-vector embedding for word representation and incorporates natural language understanding of emojis to improve semantic interpretation. The model achieves a high accuracy of 94%, demonstrating its effectiveness in analyzing sentiments in online discussions. The originality lies in applying Bi-LSTM to handle multilingual challenges on social platforms. The findings have practical implications for business, policymaking, and public sentiment evaluation, while also contributing positively to fostering informed online discourse.

Progressive Functional Analysis and Function-Based Intervention Via Telehealth: A Replication and Extension.

The purpose of this study was to evaluate a progressive functional analysis (FA) model and function-based intervention delivered by caregivers with coaching via telehealth. Children diagnosed with autism and at least one caregiver (e.g., parent) participated in the study. We conducted three assessments prior to and following intervention: a researcher-developed 10-min observation, the Parental Stress Index, and the externalizing section of the Vineland Adaptive Behavior Rating Scales, Third Edition (VABS-3). We included 47 participants in the present evaluation of the progressive FA model. We identified the function of challenging behavior for 36 participants. A function was not identified for nine participants who exhibited low or no challenging behavior during the assessment; the results were inconclusive for two participants. For the 17 participants who participated in the intervention evaluation phase, each of the participants achieved the mastery criterion, which was an 80% reduction in challenging behavior for most participants. Additionally, there was a statistically significant decrease in the VABS-3 externalizing behavior measure from pre- to post-assessment, although neither of the other pre-post measures resulted in statistically significant changes. This study replicates and extends previous research, supporting the use of progressive FA model and function-based interventions to improve challenging behavior.

Shrinkage estimation of gene interaction networks in single-cell RNA sequencing data

BackgroundGene interaction networks are graphs in which nodes represent genes and edges represent functional interactions between them. These interactions can be at multiple levels, for instance, gene regulation, protein-protein interaction, or metabolic pathways. To analyse gene interaction networks at a large scale, gene co-expression network analysis is often applied on high-throughput gene expression data such as RNA sequencing data. With the advance in sequencing technology, expression of genes can be measured in individual cells. Single-cell RNA sequencing (scRNAseq) provides insights of cellular development, differentiation and characteristics at the transcriptomic level. High sparsity and high-dimensional data structures pose challenges in scRNAseq data analysis.ResultsIn this study, a sparse inverse covariance matrix estimation framework for scRNAseq data is developed to capture direct functional interactions between genes. Comparative analyses highlight high performance and fast computation of Stein-type shrinkage in high-dimensional data using simulated scRNAseq data. Data transformation approaches also show improvement in performance of shrinkage methods in non-Gaussian distributed data. Zero-inflated modelling of scRNAseq data based on a negative binomial distribution enhances shrinkage performance in zero-inflated data without interference on non zero-inflated count data.ConclusionThe proposed framework broadens application of graphical model in scRNAseq analysis with flexibility in sparsity of count data resulting from dropout events, high performance, and fast computational time. Implementation of the framework is in a reproducible Snakemake workflow https://github.com/calathea24/ZINBGraphicalModel and R package ZINBStein https://github.com/calathea24/ZINBStein.

Clinically applicable optimized periprosthetic joint infection diagnosis via AI based pathology

Periprosthetic joint infection (PJI) is a severe complication after joint replacement surgery that demands precise diagnosis for effective treatment. We enhanced PJI diagnostic accuracy through three steps: (1) developing a self-supervised PJI model with DINO v2 to create a large dataset; (2) comparing multiple intelligent models to identify the best one; and (3) using the optimal model for visual analysis to refine diagnostic practices. The self-supervised model generated 27,724 training samples and achieved a perfect AUC of 1, indicating flawless case differentiation. EfficientNet v2-S outperformed CAMEL2 at the image level, while CAMEL2 was superior at the patient level. By using the weakly supervised PJI model to adjust diagnostic criteria, we reduced the required high-power field diagnoses per slide from five to three. These findings demonstrate AI’s potential to improve the accuracy and standardization of PJI pathology and have significant implications for infectious disease diagnostics.

Mosaicking based optimal threshold image enhancement for violence detection with deep quadratic attention mechanism

Violence is a prevalent societal issue that poses significant threats to individuals and communities. To address this challenge, researchers have developed various machine learning models to detect and prevent violent incidents. The violence detection problem statement is a critical challenge that requires innovative and collaborative solutions. By leveraging advancements in artificial intelligence, machine learning, and data analytics, systems are developed to effectively detect violence, prevent its occurrence, and protect vulnerable individuals and communities. Therefore proposed framework utilizes Differential Evolution Algorithm (DEA) along with mosaicking process which helps in enhancing the quality of the image in order to detect the presence of violence. Hockey fight dataset consist of 40 video clips out of which 20 are labelled as fight videos and remaining 20 is labelled as non-fight videos. The image is split as 30:20 ratio and the image will undergo either brightness enhancement or contrast enhancement according to the threshold value obtained by the fitness function. In mosaicking process, Ruled KNN based SURF algorithm is employed in the proposed model as it works faster and helps in providing high accuracy which assist in detecting the violence present in the images. The classification technique employed in the proposed method is Deep Quadratic Attention Mechanism (DQAM). It helps in providing better results for image classification, reducing the convergence rate and helps in improving the normalization. The performance of the proposed method is evaluated using different performance metrics which includes PSNR, SSIM and proposed method is compared with previous phase model in comparative analysis for comparing the efficiency of the proposed and existing model.