METHODS: After 10 min of supine rest, the subject was tilted at a 70-degree angle on a tilt table for approximately a total of 35 min. 400 µg of glyceryl trinitrate (GTN) was administered sublingually after the first 20 min and monitoring continued for another 15 min. Mean imputation and K-nearest neighbors (KNN) imputation approaches to handle missing values. Next, feature selection techniques were implemented, including genetic algorithm, recursive feature elimination, and feature importance, to determine the crucial features. The Mann-Whitney U test was then performed to determine the statistical difference between two groups. Patients with VVS are categorized via machine learning models including Support Vector Machine (SVM), Gaussian Naïve Bayes (GNB), Multinomial Naïve Bayes (MNB), KNN, Logistic Regression (LR), and Random Forest (RF). The developed model is interpreted using an explainable artificial intelligence (XAI) model known as partial dependence plot.
RESULTS: A total of 137 subjects aged between 9 and 93 years were recruited for this study, 54 experienced clinical symptoms were considered positive tests, while the remaining 83 tested negative. Optimal results were obtained by combining the KNN imputation technique and three tilting features with SVM with 90.5% accuracy, 87.0% sensitivity, 92.7% specificity, 88.6% precision, 87.8% F1 score, and 95.4% ROC (receiver operating characteristics) AUC (area under curve).
CONCLUSIONS: The proposed algorithm effectively classifies VVS patients with over 90% accuracy. However, the study was confined to a small sample size. More clinical datasets are required to ensure that our approach is generalizable.
METHODS: This protocol was drafted in agreement with the ROBUST-statement, and is submitted for publication before database lock and primary data analysis. The primary outcome is health-related quality of life as measured by the EQ-5D-5L health utility score and is longitudinally assessed. Secondary outcomes comprise the 6-min walking test and handgrip strength over the entire follow-up period (longitudinal analyses), and 60-day mortality, duration of mechanical ventilation, and EQ-5D-5L health utility scores at 30, 90 and 180 days (cross-sectional). All analyses will primarily be performed under weakly informative priors. When available, informative priors elicited from contemporary literature will also be incorporated under alternative scenarios. In all other cases, objectively formulated skeptical and enthusiastic priors will be defined to assess the robustness of our results. Relevant identified subgroups were: patients with acute kidney injury, severe multi-organ failure and patients with or without sepsis. Results will be presented as absolute risk differences, mean differences, and odds ratios, with accompanying 95% credible intervals. Posterior probabilities will be estimated for clinically important benefit and harm.
DISCUSSION: The proposed secondary, pre-planned Bayesian analysis of the PRECISe trial will provide additional information on the effects of high protein on functional and clinical outcomes in critically ill patients, such as probabilistic interpretation, probabilities of clinically important effect sizes, and the integration of prior evidence. As such, it will complement the interpretation of the primary outcome as well as several secondary and subgroup analyses.
OBJECTIVE: To employ machine learning (ML) and stacked ensemble learning (EL) methods in predicting short- and long-term mortality in Asian patients diagnosed with NSTEMI/UA and to identify the associated features, subsequently evaluating these findings against established risk scores.
METHODS: We analyzed data from the National Cardiovascular Disease Database for Malaysia (2006-2019), representing a diverse NSTEMI/UA Asian cohort. Algorithm development utilized in-hospital records of 9,518 patients, 30-day data from 7,133 patients, and 1-year data from 7,031 patients. This study utilized 39 features, including demographic, cardiovascular risk, medication, and clinical features. In the development of the stacked EL model, four base learner algorithms were employed: eXtreme Gradient Boosting (XGB), Support Vector Machine (SVM), Naive Bayes (NB), and Random Forest (RF), with the Generalized Linear Model (GLM) serving as the meta learner. Significant features were chosen and ranked using ML feature importance with backward elimination. The predictive performance of the algorithms was assessed using the area under the curve (AUC) as a metric. Validation of the algorithms was conducted against the TIMI for NSTEMI/UA using a separate validation dataset, and the net reclassification index (NRI) was subsequently determined.
RESULTS: Using both complete and reduced features, the algorithm performance achieved an AUC ranging from 0.73 to 0.89. The top-performing ML algorithm consistently surpassed the TIMI risk score for in-hospital, 30-day, and 1-year predictions (with AUC values of 0.88, 0.88, and 0.81, respectively, all p < 0.001), while the TIMI scores registered significantly lower at 0.55, 0.54, and 0.61. This suggests the TIMI score tends to underestimate patient mortality risk. The net reclassification index (NRI) of the best ML algorithm for NSTEMI/UA patients across these periods yielded an NRI between 40-60% (p < 0.001) relative to the TIMI NSTEMI/UA risk score. Key features identified for both short- and long-term mortality included age, Killip class, heart rate, and Low-Molecular-Weight Heparin (LMWH) administration.
CONCLUSIONS: In a broad multi-ethnic population, ML approaches outperformed conventional TIMI scoring in classifying patients with NSTEMI and UA. ML allows for the precise identification of unique characteristics within individual Asian populations, improving the accuracy of mortality predictions. Continuous development, testing, and validation of these ML algorithms holds the promise of enhanced risk stratification, thereby revolutionizing future management strategies and patient outcomes.
DESIGN: Retrospective study.
SETTING: Malaysian National Cardiovascular Disease Database-Acute Coronary Syndrome (NCVD-ACS) registry years 2006-2013, which consists of 18 hospitals across the country.
PARTICIPANTS: 7180 male patients diagnosed with STEMI from the NCVD-ACS registry.
PRIMARY AND SECONDARY OUTCOME MEASURES: A graphical model based on the Bayesian network (BN) approach has been considered. A bootstrap resampling approach was integrated into the structural learning algorithm to estimate probabilistic relations between the studied features that have the strongest influence and support.
RESULTS: The relationships between 16 features in the domain of CVD were visualised. From the bootstrap resampling approach, out of 250, only 25 arcs are significant (strength value ≥0.85 and the direction value ≥0.50). Age group, Killip class and renal disease were classified as the key predictors in the BN model for male patients as they were the most influential variables directly connected to the outcome, which is the patient status. Widespread probabilistic associations between the key predictors and the remaining variables were observed in the network structure. High likelihood values are observed for patient status variable stated alive (93.8%), Killip class I on presentation (66.8%), patient younger than 65 (81.1%), smoker patient (77.2%) and ethnic Malay (59.2%). The BN model has been shown to have good predictive performance.
CONCLUSIONS: The data visualisation analysis can be a powerful tool to understand the relationships between the CVD prognostic variables and can be useful to clinicians.
METHOD: We developed a conjoint analysis survey evaluating patients' preferences for FDA-approved MOUDs. We recruited patients with OUD presenting to initiate treatment. This survey included five attributes: induction, location and route of administration, impact on mortality, side effects, and withdrawal symptoms with cessation. Participants performed 12 choice sets, each with two hypothetical profiles and a "none" option. We used Hierarchical Bayes to identify relative importance of each attribute and part-worth utility scores of levels, which we compared using chi-squared analysis. We used the STROBE checklist to guide our reporting of this cross-sectional observational study.
RESULTS: Five-hundred and thirty participants completed the study. Location with route of administration was the most important attribute. Symptom relief during induction and withdrawal was a second priority. Mortality followed by side effects had lowest relative importance. Attribute levels with highest part-worth utilities showed patients preferred monthly pick-up from a pharmacy rather than daily supervised dosing; and oral medications more than injection/implants, despite the latter's infrequency.
CONCLUSION: We measured treatment preferences among patients seeking to initiate OUD treatment to inform strategies to scale MOUD treatment uptake. Patients prioritize the route of administration in treatment preference-less frequent pick up, but also injections and implants were less preferred despite their convenience. Second, patients prioritize symptom relief during the induction and withdrawal procedures of medication. These transition periods influence the sustainability of treatment. Although health professionals prioritize mortality, it did not drive decision-making for patients. To our knowledge, this is the largest study on patients' preferences for MOUD among treatment-seeking people with OUD to date. Future analysis will evaluate patient preference heterogeneity to further target program planning, counseling, and decision aid development.
METHODS: C0 were retrieved from a large neonatal vancomycin dataset. Individual estimates of AUC0-24 were obtained from Bayesian post hoc estimation. Various ML algorithms were used for model building to C0 and AUC0-24. An external dataset was used for predictive performance evaluation.
RESULTS: Before starting treatment, C0 can be predicted a priori using the Catboost-based C0-ML model combined with dosing regimen and nine covariates. External validation results showed a 42.5% improvement in prediction accuracy by using the ML model compared with the population pharmacokinetic model. The virtual trial showed that using the ML optimized dose; 80.3% of the virtual neonates achieved the pharmacodynamic target (C0 in the range of 10-20 mg/L), much higher than the international standard dose (37.7-61.5%). Once therapeutic drug monitoring (TDM) measurements (C0) in patients have been obtained, AUC0-24 can be further predicted using the Catboost-based AUC-ML model combined with C0 and nine covariates. External validation results showed that the AUC-ML model can achieve an prediction accuracy of 80.3%.
CONCLUSION: C0-based and AUC0-24-based ML models were developed accurately and precisely. These can be used for individual dose recommendations of vancomycin in neonates before treatment and dose revision after the first TDM result is obtained, respectively.
METHODS: We modelled meningitis mortality using vital registration, verbal autopsy, sample-based vital registration, and mortality surveillance data. Meningitis morbidity was modelled with a Bayesian compartmental model, using data from the published literature identified by a systematic review, as well as surveillance data, inpatient hospital admissions, health insurance claims, and cause-specific meningitis mortality estimates. For aetiology estimation, data from multiple causes of death, vital registration, hospital discharge, microbial laboratory, and literature studies were analysed by use of a network analysis model to estimate the proportion of meningitis deaths and cases attributable to the following aetiologies: Neisseria meningitidis, Streptococcus pneumoniae, Haemophilus influenzae, group B Streptococcus, Escherichia coli, Klebsiella pneumoniae, Listeria monocytogenes, Staphylococcus aureus, viruses, and a residual other pathogen category.
FINDINGS: In 2019, there were an estimated 236 000 deaths (95% uncertainty interval [UI] 204 000-277 000) and 2·51 million (2·11-2·99) incident cases due to meningitis globally. The burden was greatest in children younger than 5 years, with 112 000 deaths (87 400-145 000) and 1·28 million incident cases (0·947-1·71) in 2019. Age-standardised mortality rates decreased from 7·5 (6·6-8·4) per 100 000 population in 1990 to 3·3 (2·8-3·9) per 100 000 population in 2019. The highest proportion of total all-age meningitis deaths in 2019 was attributable to S pneumoniae (18·1% [17·1-19·2]), followed by N meningitidis (13·6% [12·7-14·4]) and K pneumoniae (12·2% [10·2-14·3]). Between 1990 and 2019, H influenzae showed the largest reduction in the number of deaths among children younger than 5 years (76·5% [69·5-81·8]), followed by N meningitidis (72·3% [64·4-78·5]) and viruses (58·2% [47·1-67·3]).
INTERPRETATION: Substantial progress has been made in reducing meningitis mortality over the past three decades. However, more meningitis-related deaths might be prevented by quickly scaling up immunisation and expanding access to health services. Further reduction in the global meningitis burden should be possible through low-cost multivalent vaccines, increased access to accurate and rapid diagnostic assays, enhanced surveillance, and early treatment.
FUNDING: Bill & Melinda Gates Foundation.
METHODS: Estimates of diabetes prevalence and burden were computed in 204 countries and territories, across 25 age groups, for males and females separately and combined; these estimates comprised lost years of healthy life, measured in disability-adjusted life-years (DALYs; defined as the sum of years of life lost [YLLs] and years lived with disability [YLDs]). We used the Cause of Death Ensemble model (CODEm) approach to estimate deaths due to diabetes, incorporating 25 666 location-years of data from vital registration and verbal autopsy reports in separate total (including both type 1 and type 2 diabetes) and type-specific models. Other forms of diabetes, including gestational and monogenic diabetes, were not explicitly modelled. Total and type 1 diabetes prevalence was estimated by use of a Bayesian meta-regression modelling tool, DisMod-MR 2.1, to analyse 1527 location-years of data from the scientific literature, survey microdata, and insurance claims; type 2 diabetes estimates were computed by subtracting type 1 diabetes from total estimates. Mortality and prevalence estimates, along with standard life expectancy and disability weights, were used to calculate YLLs, YLDs, and DALYs. When appropriate, we extrapolated estimates to a hypothetical population with a standardised age structure to allow comparison in populations with different age structures. We used the comparative risk assessment framework to estimate the risk-attributable type 2 diabetes burden for 16 risk factors falling under risk categories including environmental and occupational factors, tobacco use, high alcohol use, high body-mass index (BMI), dietary factors, and low physical activity. Using a regression framework, we forecast type 1 and type 2 diabetes prevalence through 2050 with Socio-demographic Index (SDI) and high BMI as predictors, respectively.
FINDINGS: In 2021, there were 529 million (95% uncertainty interval [UI] 500-564) people living with diabetes worldwide, and the global age-standardised total diabetes prevalence was 6·1% (5·8-6·5). At the super-region level, the highest age-standardised rates were observed in north Africa and the Middle East (9·3% [8·7-9·9]) and, at the regional level, in Oceania (12·3% [11·5-13·0]). Nationally, Qatar had the world's highest age-specific prevalence of diabetes, at 76·1% (73·1-79·5) in individuals aged 75-79 years. Total diabetes prevalence-especially among older adults-primarily reflects type 2 diabetes, which in 2021 accounted for 96·0% (95·1-96·8) of diabetes cases and 95·4% (94·9-95·9) of diabetes DALYs worldwide. In 2021, 52·2% (25·5-71·8) of global type 2 diabetes DALYs were attributable to high BMI. The contribution of high BMI to type 2 diabetes DALYs rose by 24·3% (18·5-30·4) worldwide between 1990 and 2021. By 2050, more than 1·31 billion (1·22-1·39) people are projected to have diabetes, with expected age-standardised total diabetes prevalence rates greater than 10% in two super-regions: 16·8% (16·1-17·6) in north Africa and the Middle East and 11·3% (10·8-11·9) in Latin America and Caribbean. By 2050, 89 (43·6%) of 204 countries and territories will have an age-standardised rate greater than 10%.
INTERPRETATION: Diabetes remains a substantial public health issue. Type 2 diabetes, which makes up the bulk of diabetes cases, is largely preventable and, in some cases, potentially reversible if identified and managed early in the disease course. However, all evidence indicates that diabetes prevalence is increasing worldwide, primarily due to a rise in obesity caused by multiple factors. Preventing and controlling type 2 diabetes remains an ongoing challenge. It is essential to better understand disparities in risk factor profiles and diabetes burden across populations, to inform strategies to successfully control diabetes risk factors within the context of multiple and complex drivers.
FUNDING: Bill & Melinda Gates Foundation.
RESULTS: The spatial model with split random effects and a common intercept has the lowest Deviance and Watanabe Information Criteria. There was evidence of a spatial pattern in the prevalence of childhood overweight across districts. An increasing trend in smoothed prevalence of overweight was observed when moving from the east to the west of the Peninsular and Borneo regions. The proportion of Bumiputera ethnicity in the district had a significant negative association with childhood overweight: the higher the proportion of Bumiputera ethnicity in the district, the lower the prevalence of childhood overweight.
CONCLUSION: This study illustrates different available techniques for mapping prevalence across districts in disconnected regions using survey data. These techniques can be utilized to produce reliable subnational estimates for any areas that comprise of disconnected regions. Through the example, we learned that the best-fit model was the one that considered the separate variations of the individual regions. We discovered that the occurrence of childhood overweight in Malaysia followed a spatial pattern with an east-west gradient trend, and we identified districts with high prevalence of overweight. This information could help policy makers in making informed decisions for targeted public health interventions in high-risk areas.