Specific Scores Research Articles

Cardiac amyloidosis Echo-Score for patients with severe aortic stenosis. The AMY-TAVI trial

Abstract Background and Purpose A longitudinal strain pattern characterized by apical sparing (RELAPS>1) has been identified as indicative of cardiac amyloidosis (CA). We conducted an analysis to assess its diagnostic accuracy in individuals with severe aortic stenosis (AS) and explored novel echocardiographic variables for predictive value. Methods All patients diagnosed with AS who underwent transcatheter aortic valve implantation (TAVI) were prospectively enrolled. 2D-Speckle-tracking echocardiography was utilized to assess myocardial deformation parameters. Subsequent to TAVI, screening for CA was conducted through 99mTc-DPD scintigraphy and protein electrophoresis. Results A total of 324 patients were included (mean age 81.5 ± 5.8 years, 51% women). Among them, 38 individuals (11.7%) exhibited cardiac uptake in scintigraphy: 14 patients (4.3%) with grade 1, 13(4%) grade 2, and 11(3.4%) grade 3. RELAPS>1 was more prevalent in patients with AS-CA (74%vs.44%,p=0.006). We propose an echocardiographic prediction model (GRAM score) for CA in the context of AS that is more sensitive and specific than RELAPS >1 alone. This model incorporates the left ventricular mass index, maximum transvalvular aortic pressure gradient, RELAPS>1, and age (AUC: 0.85, 95% CI: 0.77–0.93). Conclusions While RELAPS >1 is more commonly observed in individuals with AS-CA, it is noteworthy that this pattern can also manifest in nearly half of AS patients without CA, diminishing its utility as a standalone screening tool. In light of this, our study puts forth a more sensitive and specific prediction score for identifying CA in patients with severe AS.GRAM score for predicting AS-CA

Machine learning identification of risk factors for ischaemic stroke in patients with heart failure with reduced ejection fraction but without atrial fibrillation: A report from WARCEF trial

Abstract Background The prediction of ischaemic stroke in patients with heart failure with reduced ejection fraction (HFrEF) but without atrial fibrillation (AF) remains challenging and risk stratification based clinical factors alone has limited discrimination. Our aim was to evaluate the performance of machine learning (ML) to identify risk factors for ischaemic stroke in such HFrEF patients. Methods We performed a post-hoc analysis of the WARCEF trial, only including patients without prior history of AF, and censoring those who developed incident AF during follow-up. We evaluated the performance of 9 ML models to identify risk factors for incident stroke, using metrics including area under the curve (AUC), precision, specificity, and decision curve analysis. The importance of each feature in constructing the model was quantified using SHapley Additive exPlanations (SHAP) values. Results We included 2,213 patients with HFrEF but without AF (mean age 59±12 years; 79% male). Of these, 74 (3.3%) had an ischaemic stroke during mean follow-up of 3.5±1.8 years. Out of the 29 patient-demographics variables, 12 were selected for the ML training. All ML models demonstrated high AUC values, outperforming the CHA2DS2-VASc score (Figure 1). The logistic regression model exhibited a slight deviation from this trend. The XGBoost model achieved an AUC of 0.86 (95% confidence interval [CI]: 0.72-0.99), while the Random Forest model showed an AUC of 0.85 (95% CI: 0.74-0.96). Similarly, the Support Vector Machine and Light Gradient Boosting Machine models recorded AUCs of 0.85 (95% CI: 0.72-0.97) and 0.83 (95% CI: 0.70-0.96), respectively. Each of these models attained precision and specificity scores of 1.0, indicating high performance. These models consistently provided significant net clinical benefits. Key risk factors for stroke identified across these models were creatinine clearance and blood urea nitrogen. The use of warfarin appears to be protective. Conclusions Machine-learning models can identify risk factors for ischaemic stroke in patients with HFrEF but without AF. Worsening renal failure was predictive of an incident stroke, while warfarin use was protective against stroke (Figure 2).Figure 1Figure 2

Associations between genotype, phenotype and behaviours measured by the Rett syndrome behaviour questionnaire in Rett syndrome

IntroductionRett syndrome (RTT) is a rare neurodevelopmental disorder with developmental impairments, comorbidities, and abnormal behaviours such as hand stereotypies and emotional features. The Rett Syndrome Behaviour Questionnaire (RSBQ) was developed to describe the behavioural and emotional features of RTT. Little is known how RSBQ scores are associated with genetic and clinical characteristics in RTT. This study investigated relationships between genotype, age, walking, hand function, sleep, and RSBQ total and subscale scores in RTT.MethodsThis is a cross-sectional analysis of data collected in the Australian Rett Syndrome Database and the International Rett Syndrome Phenotype Database. Parent caregivers completed the RSBQ and Sleep Disturbance Scale for Children [subscales for disorders of initiating and maintaining sleep (DIMS), disorders of excessive somnolence (DOES)], and provided information on age, variant type, functional abilities (mobility, hand function), seizure frequency and gastrointestinal problems. Associations between the RSBQ scores and the independent variables were modelled using linear regression.ResultsData were available for 365 individuals with RTT [median (range) age 17.8 (2.9–51.9) years, 2 males]. Compared to adults, 2- to 12-year-old children had higher mean Total, Night-time Behaviour and Fear/Anxiety scores. Compared to individuals with a C-terminal deletion, individuals with the p.Arg255* variant had higher mean Total and Night-time Behaviours scores, whereas the p.Arg294* variant had higher mean Mood scores. Individuals with intermediate mobility and hand function abilities had a higher mean Total score. Total RSBQ and subscale scores were similar across categories for seizures, constipation, and reflux, but were higher with abnormal DIMS and abnormal DOES scores.ConclusionExcept for associations with sleep, the RSBQ measures the behavioural phenotype rather than clinical severity in RTT, as traditionally conceptualised in terms of functional abilities and comorbidities. When designing clinical trials, the RSBQ needs to be complemented by other outcome measures to assess specific core functions and associated comorbidities in RTT.

Healthy and unhealthy dietary patterns and sperm quality from the Led-Fertyl study.

Dietary patterns may affect sperm quality, but the scientific evidence is limited. To evaluate the association between adherence to different a-priori dietary patterns and sperm quality parameters in healthy reproductive-age men. A cross-sectional analysis was conducted using data from 200 young men enrolled in the Led-Fertyl study. Tertiles of six a-priori dietary patterns were estimated: four healthy dietary patterns [Mediterranean Diet Adherence Screener (MEDAS), Dietary Approaches to Stop Hypertension (DASH), Healthful Plant-Based Diet Index (hPDI) and EAT-Lancet Score], and two unhealthy dietary patterns [Western Diet and Unhealthful Plant-Based Diet Index (uPDI)]. Sperm quality parameters (count, concentration, vitality, total and progressive motility, and normal morphology) were considered the main outcomes. Compared with the lowest tertile, participants in the highest MEDAS tertile had higher total sperm count (β=3.2;95%CI: 1.0, 5.5) and concentration (β=1.8;95%CI: 0.6, 3.0), and total (β=8.2;95%CI: 1.3, 15.1) and progressive motility (β=7.1;95%CI: 0.2, 14.0). Similarly, participants in the highest hPDI tertile had higher total sperm count (β=3.4;95%CI: 1.4, 5.5) and concentration (β=1.2;95%CI: 0.0, 2.3) compared with those in the lowest tertile. When these dietary patterns were modelled as continuous variables (for each 1-point increment in the specific score), an inverse association was found between the uPDI and Western and total sperm count [(β=-2.7;95%CI: -4.8, -0.7) and (β=-3.8;95%CI: -5.8, -1.7), respectively] and sperm concentration [(β=-1.2;95%CI: -2.4, -0.1) and (β=-1.7;95%CI: -2.8, -0.5), respectively]. Compared with participants in the lowest tertile, those in the highest uPDI tertile presented higher odds of abnormal sperm concentration (OR: 4.6;95%CI: 1.0, 19.9) and one or more seminogram abnormalities (OR: 2.3;95%CI: 1.1, 5.0). Our findings suggest that higher adherence to healthy dietary patterns (Mediterranean and healthful plant-based diet) was positively associated with better sperm quality parameters, in contrast, greater adherence to unhealthy dietary patterns was inversely associated.

Gestational organophosphate esters (OPEs) and executive function in adolescence: The HOME Study

BackgroundEvidence from toxicological studies indicate organophosphate esters (OPEs) are neurotoxic, but few epidemiological studies investigated associations between gestational OPEs and executive function. ObjectiveTo examine the associations between gestational concentrations of OPE urinary metabolites and executive function at 12 years MethodsWe used data from 223 mother-adolescent dyads from the Health Outcomes of Measures of the Environment (HOME) Study. Women provided spot urine samples at 16 weeks gestation, 26 weeks gestation, and at delivery for quantification of bis(1,3-dichloro-2-propyl) phosphate, bis-2-chloroethyl phosphate (BCEP), diphenyl phosphate (DPHP), and di-n-butyl phosphate (DNBP). Executive function was assessed at age 12 years using the parent- and self-report Behavior Rating Inventory of Executive Function (BRIEF2). Covariate-adjusted associations between specific gravity-corrected OPEs and BRIEF2 scores were estimated using multiple informant models. Bayesian Kernel Machine Regression (BKMR) was used to assess the impact of all OPEs simultaneously. ResultsParent- and self-report BRIEF2 indices and composite scores were weakly to moderately correlated (rs=0.32-0.41). A natural-log unit increase in BCEP at 26 weeks was associated with approximately a 1-point increase on the self-report Cognitive Regulation Index [CRI] (95% CI 0.4, 2.3), the Emotion Regulation Index [ERI] (95% CI 0.3, 2.2), and the Global Executive Composite [GEC] (95% CI 0.4, 2.2), indicating poorer performance. Higher DPHP at 16 weeks was associated with lower parent-report GEC score (β=-1.1, 95% CI -2.3, -0.003). BKMR identified BCEP and DNBP at 26 weeks as important contributors to CRI and ERI, respectively. ConclusionOPE metabolites during gestational development, particularly BCEP, may influence adolescent executive function. However, since the FDR p-values failed to reach statistical significance, additional studies would benefit from using larger cohorts.

Echocardiographic Phenotype in Severe Aortic Stenosis With and Without Transthyretin Cardiac Amyloidosis: The AMY-TAVI Study.

The relative apical sparing pattern of left ventricular (LV) longitudinal strain (RELAPS]>1) has been described as a typical sign of cardiac amyloidosis (CA). The objective was to validate this pattern in concomitant CA and aortic stenosis (AS) and to identify new echocardiographic variables suggestive of CA in the presence of AS. 324 consecutive patients (age 81.5±5.8 years, 51% women) with AS who underwent transcatheter aortic valve implantation (TAVI) were prospectively included. 2D-Speckle-tracking echocardiography was performed. Following TAVI, 99mTc-DPD-scintigraphy and protein electrophoresis were performed to screen for CA. 38 patients (11.7%) showed cardiac uptake in scintigraphy: 14 patients (4.3%) with grade 1, 13 (4%) with grade 2, and 11 (3.4%) with grade 3. Patients with grades 2 and 3 (AS-CA group) had more LV hypertrophy (LV mass index: 188 vs.172 g/m2, p=0.032), lower transvalvular aortic pressure gradient (p<0.003), and higher prevalence of low-gradient AS (50% vs.19%, p=0.001), as well as greater diastolic and systolic dysfunction. Strain analysis was limited to 243 patients due to poor acoustic window and COVID-19 restrictions (81 lost: 79 in AS alone, 1 each in AS-DPD1 and AS-CA groups). RELAPS>1 was more prevalent in AS-CA group (74% vs.44%, p=0.006). An echocardiographic prediction model (GRAM score) for CA in the presence of AS, that is more sensitive and specific than RELAPS>1 alone, is proposed using the LV mass, maximum aortic gradient, and RELAPS>1, in addition to age (AUC:0.85, 95%CI: 0.77-0.93). RELAPS>1 is more prevalent in AS-CA but can occur in almost half of AS patients without CA, which reduces its value as a screening tool. A more sensitive and specific prediction score for CA in patients with severe AS is proposed.

A classification system for identifying persons with an unknown cardiovascular disease (CVD) status for a multiracial/ ethnic Caribbean population.

The need for classification systems for cardiovascular disease (CVD) that is population-specific is important towards understanding the clinical disease and diagnostics associated with the disease. This paper presents the form and validation results of this classification system. The survey data used was captured from 778 participants, 526 persons with no prior CVD, and 252 who reported prior CVD. Binomial logistic regression and Discriminant analysis were utilised to develop classification models. This classification system provided a general measure of severity of disease by utilising scores estimated from two algorithms developed from 13 routine physiologic measurements, along with demographic information of age and ethnicity, inter alia, and previous health status. For each model, specific score ranges were identified, which gave the best classification for those with a prior CVD incident (higher scores) and for others labelled as non-CVD (lower scores). The two classification models (Logistic Regression Model and Discriminant Analysis Model) developed had high area under the receiver-operating characteristic (AUROC) values (98% & 99%) and sensitivity (86 and 90%), which improved discrimination between Non-CVD and CVD participants and, more importantly, correctly classified a greater proportion of CVD participants. New to this type of research was the estimation and detailed evaluation of a range of scores, labelled non-differentiating, which fell in the middle of the spectrum and which contained the higher-end scores for the non-CVD individuals and the lower-end scores for CVD patients, all of whom were incorrectly classified, based on their prior history. The classification system of scores is able to differentiate the CVD status of individuals, with good predictability, and could assist physicians with recommending different treatment plans. The two models in this classification system each individually outperformed the three established models in terms of the strength of their correct classifications of individuals with or without prior reported CVD incidents. More importantly, they have smaller non-differentiating ranges than the three known models and, in that range, the two new models have lower CVD/non-CVD ratios suggesting they are more likely to misclassify non-CVD individuals compared to CVD patients, which is a more benign misclassification. Further, when used in combination, the two models increased the sensitivity, in classifying individuals of different ethnicities, beyond that of either one used independently or of any of the three standard European/North American models. These efforts will be instrumental in advancing personalised CVD management strategies and improving health outcomes across diverse populations.

Environmental Justice and Sustainable Development: Cumulative Environmental Exposures and All-Cause Mortality in Colorado Counties

(1) Background: Colorado’s (CO) Environmental Justice mapping tool, CO EnviroScreen, quantifies environmental injustices through “EnviroScreen Scores”, highlighting areas likely affected by environmental health disparities. Identifying the specific scores most strongly associated with mortality could help prioritize interventions and allocate resources to address these issues. This study contributes to sustainable development goals by examining the relationship between environmental justice indicators and population health outcomes. By utilizing the CO EnviroScreen tool, we assess how cumulative environmental exposures and vulnerabilities impact mortality rates, providing insights for sustainable planning and public health policies. (2) Methods: We assessed the cross-sectional association between county-level all-cause mortality rates in CO, using 2019 data from the Colorado Department of Public Health and Environment, and three county-level component scores obtained from CO EnviroScreen: sensitive populations (i.e., health-related outcomes), environmental exposures (e.g., from air, water, noise), and climate vulnerability (i.e., risk of drought, flood, extreme heat, wildfire). A quasi-Poisson generalized linear model was utilized, incorporating covariates (county-level metrics for insufficient sleep, alcohol overconsumption, physical inactivity, and smoking) to explore associations adjusted for behavioral risk factors (n = 64 counties). (3) Results: The analysis revealed that a 10% increase in the “Environmental Exposures” component score was associated with a 3% higher all-cause mortality rate (95% CI: 1.00, 1.05), highlighting the importance of addressing environmental determinants for sustainable community health. No significant associations were observed for the “Sensitive Populations” or “Climate Vulnerability” component scores. (4) Conclusions: This study provides novel evidence of an association between the CO EnviroScreen score, particularly the environmental exposure component, and all-cause mortality rates at the county level in Colorado in 2019. The findings suggest that cumulative environmental exposures may contribute to geographic disparities in mortality risk, even after adjusting for key behavioral risk factors. These results underscore the importance of integrating environmental justice considerations into sustainable development strategies to promote equitable health outcomes and resilient communities. While our study demonstrates the utility of CO EnviroScreen in identifying areas at risk due to environmental factors, it does not establish a direct link to broader environmental justice outcomes. Further research is needed to explore specific environmental exposures and their direct impacts on health disparities to provide a more complete picture of environmental justice in Colorado.

FORTA Score and Negative Outcomes in Older Adults: Insights from Italian Internal Medicine Wards.

The study aimed to assess the relationship between the Fit fOR The Aged (FORTA) score-a classification system designed to evaluate medication appropriateness in older adults-and several negative outcomes, including impaired cognitive performance, functional status, adverse clinical events, and all-cause mortality at 3, 6, and 12 months after hospital discharge. This retrospective study utilized data from the ELICADHE cohort, a cluster-randomized trial conducted across 20 Italian internal medicine and geriatric wards. The study included patients aged 75 and older with complete FORTA score assessments. Demographics, medication history, and comorbidities were collected. The FORTA classification system assessed medication appropriateness. FORTA scores were calculated and FORTA score cut-offs (3 and 5) were applied. Statistical analyses included descriptive statistics, survival analysis with Cox regression, logistic regression, and negative-binomial regression using SAS 9.4 and RStudio 12.1. Ethical approval was obtained. Of the 506 patients included, 171 (33.8%) were fully assessable with complete FORTA scores. The study found no significant association between higher FORTA scores and impaired cognitive performance, functional status, or mortality. Additionally, no clear relationship was observed between FORTA scores and adverse clinical events or mortality. The analysis indicated that age was a significant factor associated with mortality and adverse clinical events. The study did not find a significant relationship between the FORTA score and negative outcomes in older patients discharged from internal medicine and geriatric wards. Further research is needed to define specific FORTA score cut-off values and expand the criteria to improve medication assessment in this population.

Action observation intervention using three-dimensional movies improves the usability of hands with distal radius fractures in daily life-A nonrandomized controlled trial in women.

Prolonged immobilization of joints after distal radius fracture (DRF) causes cerebral disuse-dependent plasticity (DDP) and deterioration of upper extremity function. Action observation therapy (AOT) can improve DDP. This nonrandomized controlled trial (UMIN 000039973) tested the hypothesis that AOT improves hand-use difficulties during activities of daily living in patients with DRF. Right-handed women with volar locking plate fixation for DRF were divided into AOT and Non-AOT groups for a 12-week intervention. The primary outcome was difficulty in using the fractured hand, assessed with the Japanese version of the Patient-related Wrist Evaluation (PRWE). The secondary outcomes were range of motion (ROM) of the injured side and gap between measured ROM and patient-estimated ROM. The survey was administered immediately post operation and at postoperative weeks 4, 8, and 12. The AOT group used a head-mounted display and three-dimensional video during ROM exercises. The Non-AOT group used active ROM exercises alone. A generalized linear model (GLM) was used to confirm interactions and main effects by group and time period, and multiple comparisons were performed. Thirty-five patients were assigned to the AOT group (n = 18, median age, 74 years) or the Non-AOT group (n = 17, median age, 70 years). In the GLM, PRWE Total, PRWE Specific, and PRWE Usual scores revealed interactions between groups and periods. The post-hoc test revealed that the PRWE Specific scores (z = 3.43, p = 0.02) and PRWE Usual scores (z = 7.53, p<0.01) were significantly lower in the AOT group than in the Non-AOT group at 4 weeks postoperatively, whereas PRWE Total scores (z = 3.29, p = 0.04) were lower at 8 weeks postoperatively. These results suggested that AOT can improve hand-use difficulties in right-handed women after DRF surgery. AOT positively affects the motor imagery of patients with DRF and can reverse the patient's perceived difficulty in using the fractured hand during rehabilitation.

Loop-Mediated Isothermal Amplification Assay Using Gold Nanoparticles for Detecting Treponema pallidum subspp. pallidum.

Venereal syphilis in humans is caused by Trepenoma pallidum subspp. pallidum. A study has shown that 30,302 individuals in Thailand had syphilis in 2020, with a male-to-female ratio of 1:0.8 and the highest incidence rate at ages between fifteen and twenty-four. This research aimed to develop a loop-mediated isothermal amplification assay using gold nanoparticles (LAMP-AuNPs). Analytical sensitivity, diagnostic specificity, accuracy, and predictive values for each technique are provided. The diagnosis sensitivities of polymerase chain reaction using agarose gel electrophoresis (PCR-AGE), loop-mediated isothermal amplification assay using agarose gel electrophoresis (LAMP-AGE), and LAMP-AuNPs were 116 ng/µL, 11.6 ng/µL, and 11.6 ng/µL, respectively. We evaluated the analytical specificity using PCR and a LAMP-based assay, and there was no cross-reactivity to Leptospira interrogans, Staphylococcus aureus, Enterococcus faecalis, Escherichia coli, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, human immunodeficiency virus (HIV), and healthy humans. After analyzing 400 serum samples of patients suspected of syphilis, the LAMP-AGE and LAMP-AuNPs assays displayed 100% diagnostic sensitivity scores, 91% diagnostic specificity scores, 95.5% accuracy rates, 100% positive predictive values (PPVs), and 91% negative predictive values (NPVs), the positive likelihood ratio (LR+) was 11.11, while the negative likelihood ratio (LR-) was 0. Conversely, for PCR assays displayed 100% diagnostic sensitivity scores, 94.5% diagnostic specificity scores, 97.25% accuracy rates, 100% PPVs, and 94.5% NPVs, LR+ was 18.18, and LR- was 0. The LAMP-AuNPs technique demonstrates rapidity, affordability, and convenience, rendering it well-suited for point-of-care applications in the diagnosis, prevention, and management of pathogenic infections.

Machine learning's effectiveness in evaluating movement in one-legged standing test for predicting high autistic trait.

Research supporting the presence of diverse motor impairments, including impaired balance coordination, in children with autism spectrum disorder (ASD) is increasing. The one-legged standing test (OLST) is a popular test of balance. Since machine learning is a powerful technique for learning predictive models from movement data, it can objectively evaluate the processes involved in OLST. This study assesses machine learning's effectiveness in evaluating movement in OLST for predicting high autistic trait. In this study, 64 boys and 62 girls participated. The participants were instructed to stand on one leg on a pressure sensor while facing the experimenter. The data collected in the experiment were time-series data pertaining to pressure distribution on the sole of the foot and full-body images. A model to identify the participants belonging to High autistic trait group and Low autistic trait group was developed using a support vector machine (SVM) algorithm with 16 explanatory variables. Further, classification models were built for the conventional, proposed, and combined explanatory variable categories. The probabilities of High autistic trait group were calculated using the SVM model. For proposed and combined variables, the accuracy, sensitivity, and specificity scores were 1.000. The variables shoulder, hip, and trunk are important since they explain the balance status of children with high autistic trait. Further, the total Social Responsiveness Scale score positively correlated with the probability of High autistic trait group in each category of explanatory variables. Results indicate the effectiveness of evaluating movement in OLST by using movies and machine learning for predicting high autistic trait. In addition, they emphasize the significance of specifically focusing on shoulder and waist movements, which facilitate the efficient predicting high autistic trait. Finally, studies incorporating a broader range of balance cues are necessary to comprehensively determine the effectiveness of utilizing balance ability in predicting high autistic trait.

Prediction and feature selection of low birth weight using machine learning algorithms

Background and aimsThe birth weight of a newborn is a crucial factor that affects their overall health and future well-being. Low birth weight (LBW) is a widespread global issue, which the World Health Organization defines as weighing less than 2,500 g. LBW can have severe negative consequences on an individual’s health, including neonatal mortality and various health concerns throughout their life. To address this problem, this study has been conducted using BDHS 2017–2018 data to uncover important aspects of LBW using a variety of machine learning (ML) approaches and to determine the best feature selection technique and best predictive ML model.MethodsTo pick out the key features, the Boruta algorithm and wrapper method were used. Logistic Regression (LR) used as traditional method and several machine learning classifiers were then used, including, DT (Decision Tree), SVM (Support Vector Machine), NB (Naïve Bayes), RF (Random Forest), XGBoost (eXtreme Gradient Boosting), and AdaBoost (Adaptive Boosting), to determine the best model for predicting LBW. The model’s performance was evaluated based on the specificity, sensitivity, accuracy, F1 score and AUC value.ResultsResult shows, Boruta algorithm identifies eleven significant features including respondent’s age, highest education level, educational attainment, wealth index, age at first birth, weight, height, BMI, age at first sexual intercourse, birth order number, and whether the child is a twin. Incorporating Boruta algorithm’s significant features, the performance of traditional LR and ML methods including DT, SVM, NB, RF, XGBoost, and AB were evaluated where LR, had a specificity, sensitivity, accuracy and F1 score of 0.85, 0.5, 85.15% and 0.915. While the ML methods DT, SVM, NB, RF, XGBoost, and AB model’s respective accuracy values were 85.35%, 85.15%, 84.54%, 81.18%, and 84.41%. Based on the specificity, sensitivity, accuracy, F1 score and AUC, RF (specificity = 0.99, sensitivity = 0.58, accuracy = 85.86%, F1 score = 0.9243, AUC = 0.549) outperformed the other methods. Both the classical (LR) and machine learning (ML) models’ performance has improved dramatically when important characteristics are extracted using the wrapper method. The LR method identified five significant features with a specificity, sensitivity, accuracy and F1 score of 0.87, 0.33, 87.12% and 0.9309. The region, whether the infant is a twin, and cesarean delivery were the three key features discovered by the DT and RF models, which were implemented using the wrapper technique. All three models had the identical F1 score of 0.9318. However, “child is twin” was recognized as a significant feature by the SVM, NB, and AB models, with an F1 score of 0.9315. Ultimately, with an F1 score of 0.9315, the XGBoost model recognized “child is twin” and “age at first sex” as relevant features. Random Forest again beat the other approaches in this instance.ConclusionsThe study reveals Wrapper method as the optimal feature selection technique. The ML method outperforms traditional methods, with Random Forest (RF) being the most effective predictive model for Low-Birth-Weight prediction. The study suggests that policymakers in Bangladesh can mitigate low birth weight newborns by considering identified risk factors.

A machine learning comparison of transportation mode changes from high-speed railway promotion in Thailand

Thailand's collaboration with China to develop High-Speed Rail (HSR) represents a crucial step in enhancing transportation infrastructure and promoting regional economic growth. While most research has focused on existing travel modes such as cars, buses, and planes, there is a notable lack of analysis on future transport choices, particularly in developing countries like Thailand. This study addresses this gap by analyzing the factors influencing High-Speed Railways adoption to inform future traveler decisions. The comparison of models for predicting High-Speed Railways usage revealed that CatBoost consistently outperformed the other models, with cross-validation confirming its superior performance across all key metrics. The Binary Logit Model (BL) demonstrated moderate effectiveness, achieving an accuracy of 0.7404 ± 0.006, sensitivity of 0.7655 ± 0.006, and a relatively lower AUC of 0.8161 ± 0.0067. Its specificity (0.7140 ± 0.0129), precision (0.7376 ± 0.0085), and F1 score (0.7513 ± 0.0055) were moderate, but it struggled with handling imbalanced data. In contrast, XGBoost delivered significantly stronger results, with an accuracy of 0.8846 ± 0.0068, sensitivity of 0.9210 ± 0.0094, and an AUC of 0.9583 ± 0.0036. XGBoost also achieved high specificity (0.8464 ± 0.0062), precision (0.8630 ± 0.0055), and an F1 score (0.8910 ± 0.0067). LightGBM also performed well, achieving an accuracy of 0.8763 ± 0.0077 and a relatively high sensitivity of 0.9358 ± 0.0071. However, it exhibited lower specificity (0.8139 ± 0.0107) and precision (0.8408 ± 0.0083), resulting in a slightly reduced F1 score of 0.8857 ± 0.0069 and an AUC of 0.9506 ± 0.0034. CatBoost demonstrated the highest overall performance, achieving an accuracy of 0.8853 ± 0.0061, sensitivity of 0.9245 ± 0.0072, specificity of 0.8441 ± 0.0068, precision of 0.8616 ± 0.0058, an F1 score of 0.8920 ± 0.0058, and an AUC of 0.9584 ± 0.0034. The decision to use High-Speed Railways in Thailand is influenced by several key factors that reflect the country's unique transportation context. These factors include travel time, access time, service frequency, cost, waiting time, household income, automobile ownership, gender, and the purpose of travel. This information is valuable for shaping policies to support High-Speed Railways adoption in Thailand.

A dual-labeled dataset and fusion model for automatic teeth segmentation, numbering, and state assessment on panoramic radiographs

BackgroundRecently, deep learning has been increasingly applied in the field of dentistry. The aim of this study is to develop a model for the automatic segmentation, numbering, and state assessment of teeth on panoramic radiographs.MethodsWe created a dual-labeled dataset on panoramic radiographs for training, incorporating both numbering and state labels. We then developed a fusion model that combines a YOLOv9-e instance segmentation model with an EfficientNetv2-l classification model. The instance segmentation model is used for tooth segmentation and numbering, whereas the classification model is used for state evaluation. The final prediction results integrate tooth position, numbering, and state information. The model’s output includes result visualization and automatic report generation.ResultsPrecision, Recall, mAP50 (mean Average Precision), and mAP50-95 for the tooth instance segmentation task are 0.989, 0.955, 0.975, and 0.840, respectively. Precision, Recall, Specificity, and F1 Score for the tooth classification task are 0.943, 0.933, 0.985, and 0.936, respectively.ConclusionsThis fusion model is the first to integrate automatic dental segmentation, numbering, and state assessment. It provides highly accurate results, including detailed visualizations and automated report generation.

Assessment of Ensemble-Based Machine Learning Algorithms for Exoplanet Identification

This paper presents a comprehensive assessment procedure for evaluating Ensemble-based Machine Learning algorithms in the context of exoplanet classification. Each of the algorithm hyperparameter values were tuned. Deployments were carried out using the cross-validation method. Performance metrics, including accuracy, sensitivity, specificity, precision, and F1 score, were evaluated using confusion matrices generated from each implementation. Machine Learning (ML) algorithms were trained and used to identify exoplanet data. Most of the current research deals with traditional ML algorithms for this purpose. The Ensemble algorithm is another type of ML technique that combines the prediction performance of two or more algorithms to obtain an improved final prediction. Few studies have applied Ensemble algorithms to predict exoplanets. To the best of our knowledge, no paper that has exclusively assessed Ensemble algorithms exists, highlighting a significant gap in the literature about the potential of Ensemble methods. Five Ensemble algorithms were evaluated in this paper: Adaboost, Random Forest, Stacking, Random Subspace Method, and Extremely Randomized Trees. They achieved an average performance of more than 80% in all metrics. The results underscore the substantial benefits of fine tuning hyperparameters to enhance predictive performance. The Stacking algorithm achieved a higher performance than the other algorithms. This aspect is discussed in this paper. The results of this work show that it is worth increasing the use of Ensemble algorithms to improve exoplanet identification.

Analysis and comparison of post-translational modifications of α-synuclein filaments in multiple system atrophy and dementia with Lewy bodies

Our previous cryogenic electron microscopy (cryoEM) analysis showed that the core structures of α-synuclein filaments accumulated in brains of patients diagnosed with dementia with Lewy bodies (DLB) and multiple system atrophy (MSA) patients are different. We analyzed the post-translational modifications (PTMs) in these filaments , and examined their relationship with the core filament structures and pathological features. Besides the common PTMs in MSA and DLB filaments, acetylation, methylation, oxidation and phosphorylation were frequently detected in MSA filaments, but not in DLB filaments. Furthermore, in DLB filament cases, the processing occurred at the C-terminal side of Asp at 119 residue and Asn at 122 residue, while in MSA cases, the processing occurred at multiple sites between residues 109–123. We have previously reported that PTMs in tau filaments depend on the filament core structure. This was considered to apply to α-synuclein filaments as well. As an example, PTMs including processing sites detected in α-synuclein filaments in early-onset DLB (an atypical form, now named juvenile-onset α-synucleinopathy) brain also supported this idea. These suggests that PTMs appeared to be closely related to the specific filament core structures.

High Somatization Rates, Frequent Spontaneous Recovery, and a Lack of Organic Biomarkers in Post-Covid-19 Condition.

Many patients report neuropsychiatric symptoms after SARS-CoV-2 infection. Data on prevalence of post-COVID-19 condition (PCC) vary due to the lack of specific diagnostic criteria, the report of unspecific symptoms, and reliable biomarkers. PCC patients seen in a neurological outpatient department were followed for up to 18 months. Neurological examination, SARS-CoV-2 antibodies, Epstein-Barr virus antibodies, and cortisol levels as possible biomarkers, questionnaires to evaluate neuropsychiatric symptoms and somatization (Patient Health Questionnaires D [PHQ-D]), cognition deficits (Montreal Cognitive Assessment [MoCA]), sleep disorders (ISS, Epworth Sleepiness Scale [ESS]), and fatigue (FSS) were included. A total of 175 consecutive patients (78% females, median age 42 years) were seen between May 2021 and February 2023. Fatigue, subjective stress intolerance, and subjective cognitive deficits were the most common symptoms. Specific scores were positive for fatigue, insomnia, and sleepiness and were present in 95%, 62.1%, and 44.0%, respectively. Cognitive deficits were found in 2.3%. Signs of somatization were identified in 61%, who also had an average of two symptoms more than patients without somatization. Overall, 28% had a psychiatric disorder, including depression and anxiety. At the second visit (n=92), fatigue (67.3%) and insomnia (45.5%) had decreased. At visit three (n=43), symptom load had decreased in 76.8%; overall, 51.2% of patients were symptom-free. Biomarker testing did not confirm an anti-EBV response. SARS-CoV-2-specific immune reactions increased over time, and cortisol levels were within the physiological range. Despite high initial symptom load, 76.8% improved over time. The prevalence of somatization and psychiatric disorders was high. Our data do not confirm the role of previously suggested biomarkers in PCC patients.

Development of a carbohydrate-binding protein prediction algorithm using structural features of stacking aromatic rings

Carbohydrate-protein interactions play fundamental roles in numerous aspects of biological activities, and the search for new carbohydrate (CHO)-binding proteins (CBPs) has long been a research focus. In this study, through the analysis of CBP structures, we identified significant enrichment of aromatic residues in CHO-binding regions. We further summarized the structural features of these aromatic rings within the CHO-stacking region, namely “exposing” and “proximity” features, and developed a screening algorithm that can identify CHO-stacking Trp (tryptophan) residues based on these two features. Our Trp screening algorithm can achieve high accuracy in both CBP (specificity score 0.93) and CBS (Carbohydrate binding site, precision score 0.77) prediction using experimentally determined protein structures. We also applied our screening algorithm on AlphaGO pan-species predicted models and observed significant enrichment of carbohydrate-related functions in predicted CBP candidates across different species. Moreover, through carbohydrate arrays, we experimentally verified the CHO-binding ability of four candidate proteins, which further confirms the robustness of the algorithm. This study provides another perspective on proteome-wide CBP and CBS prediction. Our results not only help to reveal the structural mechanism of CHO-binding, but also provide a pan-species CBP dataset for future CHO-protein interaction exploration.

Using machine learning model for predicting risk of memory decline: A cross sectional study

BackgroundMemory decline is the earliest symptom of various neurodegenerative disease, such as Alzheimer's disease (AD). However, accurately the prediction and identification of risk factors leading up to memory decline has remained limited. ObjectiveThe objective of this study is to create and verify a machine learning model that can accurately predict risk factors for memory decline among US adults. MethodsA total of 9971 individuals were enrolled from the National Health and Nutrition Examination Survey (NHANES) 2015–2016 database. The least absolute shrinkage and selection operator (LASSO) was used to screen for characteristic predictors. Five machine learning (ML) algorithms: including Logistic Regression, ExtraTrees classifier, Bagging classifier, eXtreme Gradient Boosting (XGBoost), and Random Forest (RF) were employed. The performance of each model was evaluated by confusion matrix, area under curve (AUC), accuracy, precision, specificity, Recall and F1 scores. ResultsThe ultimate sample comprised 4525 subjects, of whom 7.7 % (N = 347) exhibited memory deterioration. The ExtraTrees classifier model and the XGBoost model demonstrated superior prediction performance and clinical value compared to other independent machine learning models, based on the AUC value of 0.915 and 0.911. Additionally, they consistently demonstrated accurate predicting ability for memory decline in the external datasets, with an AUC of 0.851 and 0.843, respectively. ConclusionThe ExtraTrees classifier and the XGBoost models were the two outperformed models in predicting memory decline. Nevertheless, it is necessary to conduct future investigations to confirm the accuracy of our findings.