Bootstrap Validation Research Articles

Distributed Bootstrap Simultaneous Inference for High-Dimensional Quantile Regression

Modern massive data with enormous sample size and tremendous dimensionality are usually impossible to process with a single machine. They are typically stored and processed in a distributed manner. In this paper, we propose a distributed bootstrap simultaneous inference for a high-dimensional quantile regression model using massive data. Meanwhile, a communication-efficient (CE) distributed learning algorithm is developed via the CE surrogate likelihood framework and ADMM procedure, which can handle the non-smoothness of the quantile regression loss and the Lasso penalty. We theoretically prove the convergence of the algorithm and establish a lower bound on the number of communication rounds ιmin that warrant statistical accuracy and efficiency. The distributed bootstrap validity and efficiency are corroborated by an extensive simulation study.

Predicting and analysing of the unfavourable outcomes of early applicated percutaneous endoscopic interlaminar discectomy for lumbar disc herniation: development and validation based nomogram.

To predict and analyse the unfavourable outcomes of early applicated percutaneous endoscopic interlaminar discectomy for lumbar disc herniation. Information of 426 patients treated by early applicated percutaneous endoscopic interlaminar discectomy (PEID) for lumbar disc herniation (LDH) at our hospital from June 2017 to October 2021 in addition to 17 other features was collected. The risk factors were selected by the least absolute shrinkage and selection operator method (LASSO) regression. Then, a prediction model (nomogram) was established to predict the unfavourable outcomes of using the risk factors selected from LASSO regression. Bootstrap (n = 1000) was used to perform the internal validation, and the performance of the model was assessed by the concordance index (C-index), receiver operating characteristic curve, and calibration curve. The decision curve analysis (DCA) and clinical impact curve (CIC) were used to assess the clinical utility of the model, respectively. Finally, 53 of 426 patients showed unfavourable outcomes. Five potential factors, Modic change, Calcification, Lumbar epidural steroid injection preoperative, Articular process hyperplasia and cohesion, and Laminoplasty technique, were selected according to the LASSO regression, that identified the predictors to establish nomogram model. Meanwhile, the C-index of the prediction nomogram was 0.847, the area under the receiver operating characteristic curve value was 0.847, and the interval bootstrapping (n = 1000) validation C-index was 0.809. The model has good practicability for clinics according to the DCA and CIC. This nomogram model has good predictive performance and clinical practicability, which could provide a certain basis for predicting unfavourable outcomes of early applicated PEID for LDH.

Predictive factors for lung metastasis in pediatric differentiated thyroid cancer: a clinical prediction study.

The objective of this study was to develop and evaluate the efficacy of a nomogram for predicting lung metastasis in pediatric differentiated thyroid cancer. The SEER database was utilized to collect a dataset consisting of 1,590 patients who were diagnosed between January 2000 and December 2019. This dataset was subsequently utilized for the purpose of constructing a predictive model. The model was constructed utilizing a multivariate logistic regression analysis, incorporating a combination of least absolute shrinkage feature selection and selection operator regression models. The differentiation and calibration of the model were assessed using the C-index, calibration plot, and ROC curve analysis, respectively. Internal validation was performed using a bootstrap validation technique. The results of the study revealed that the nomogram incorporated several predictive variables, namely age, T staging, and positive nodes. The C-index had an excellent calibration value of 0.911 (95 % confidence interval: 0.876-0.946), and a notable C-index value of 0.884 was achieved during interval validation. The area under the ROC curve was determined to be 0.890, indicating its practicality and usefulness in this context. This study has successfully developed a novel nomogram for predicting lung metastasis in children and adolescent patients diagnosed with thyroid cancer. Clinical decision-making can be enhanced by assessing clinicopathological variables that have a significant predictive value for the probability of lung metastasis in this particular population.

Identifying neonates at risk for post-discharge mortality in Dar es Salaam, Tanzania, and Monrovia, Liberia: Derivation and internal validation of a novel risk assessment tool

IntroductionThe immediate period after hospital discharge carries a large burden of childhood mortality in sub-Saharan Africa. Our objective was to derive and internally validate a risk assessment tool to identify...

Abstract WP94: Forecasting Outcomes After Mechanical Thrombectomy: Predictive Modeling of Neurological Improvement in the SELECT Database

Introduction: Outcome prediction and prognostication are at the heart of determining next steps in stroke care. We aim to evaluate two variable selection strategies to provide optimal prediction of functional independence among post-EVT patients at admission, day 1, and time of discharge using data from the SELECT prospective cohort study. Methods: Prediction strategies were created using baseline demographics, clinical presentation and imaging characteristics for the admission model, procedure characteristics and outcomes, clinical status, and baseline factors for the day 1 model, and follow-up infarct volume, cerebral edema, clinical status, and previously mentioned factors for the discharge model. Model selection used backwards stepwise regression and least absolute shrinkage and selection operator (LASSO) strategy. Patients with unavailable data were excluded. Calibration and internal validation was performed using bootstrapping for 50 repetitions. Results: 149 EVT patients with all available data points were included. Functional independence (mRS 0-2) was observed in 89 (60%) patients. LASSO and stepwise selection methods selected different variables at baseline, with LASSO providing reasonably better model performance over stepwise model selection. Both stepwise and LASSO regression methods selected baseline ASPECTS and NIHSS at day 1 for day 1 model, and baseline ASPECTS, NIHSS at day 1 and NIHSS at day 7 for discharge model. Day 1 and discharge models demonstrated significantly better predictive performance in bootstrap validation over baseline models [Table]. Conclusions: Clinical symptoms in the post-acute phase were the strongest predictors of functional independence at 90-day follow-up and were selected over procedural factors or cerebral edema. Prediction of functional independence at baseline involved several clinical and imaging variables with lower predictive performance than post-acute phase models.

Machine learning approach to predict postpancreatectomy hemorrhage following pancreaticoduodenectomy: a retrospective study.

Postpancreatectomy hemorrhage (PPH) is a rare yet dreaded complication following pancreaticoduodenectomy (PD). This retrospective study aimed to explore a machine learning (ML) model for predicting PPH in PD patients. A total of 284 patients who underwent open PD at our institute were included in the analysis. To address the issue of imbalanced data, the adaptive synthetic sampling (ADASYN) technique was employed. The best-performing ML model was selected using the PyCaret library in Python and evaluated based on recall, precision, and F1 score metrics. In addition to assessing the model's performance on the test data, bootstrap validation (n = 1000) with the original dataset was conducted. PPH occurred in 11 patients (3.9%), with a median onset time of 22days postoperatively. These minority cases were oversampled to 85 using ADASYN. The extra trees classifier demonstrated superior performance with recall, precision, and F1 score of 0.967, 0.914, and 0.937, respectively. Both validation using the test data and bootstrap resampling consistently demonstrated recall, precision, and F1 score exceeding 0.9. The model identified the peak value of C-reactive protein during the first 7 postoperative days as the most significant feature, followed by the preoperative neutrophil-to-lymphocyte ratio. This study highlights the potential of the ML approach to predict PPH occurrence following PD. Vigilance and early interventions guided by such model predictions could positively impact outcomes for high-risk patients.

Individualized and Biomarker-Based Prognosis of Longitudinal Cognitive Decline in Early Symptomatic Alzheimer's Disease.

Although individualized models using demographic, MRI, and biological markers have recently been applied in mild cognitive impairment (MCI), a similar study is lacking for patients with early Alzheimer's disease (AD) with biomarker evidence of abnormal amyloid in the brain. We aimed to develop prognostic models for individualized prediction of cognitive change in early AD. A total of 421 individuals with early AD (MCI or mild dementia due to AD) having biomarker evidence of abnormal amyloid in the brain were included in the current study. The primary cognitive outcome was the slope of change in Alzheimer's Disease Assessment Scale-cognitive subscale-13 (ADAS-Cog-13) over a period of up to 5 years. A model combining demographics, baseline cognition, neurodegenerative markers, and CSF AD biomarkers provided the best predictive performance, achieving an overfitting-corrected R2 of 0.59 (bootstrapping validation). A nomogram was created to enable clinicians or trialists to easily and visually estimate the individualized magnitude of cognitive change in the context of patient characteristics. Simulated clinical trials suggested that the inclusion of our nomogram into the enrichment strategy would lead to a substantial reduction of sample size in a trial of early AD. Our findings may be of great clinical relevance to identify individuals with early AD who are likely to experience fast cognitive deterioration in clinical practice and in clinical trials.

A Comprehensive Prediction Model for Futile Recanalization in AIS Patients Post-Endovascular Therapy: Integrating Clinical, Imaging, and No-Reflow Biomarkers.

Our study aimed to construct a predictive model for identifying instances of futile recanalization in patients with anterior circulation occlusion acute ischemic stroke (AIS) who achieved complete reperfusion following endovascular therapy. We included 173 AIS patients who attained complete reperfusion, as indicated by a Modified Thrombolysis in Cerebral Infarction (mTICI) scale score of 3. Our approach involved a thorough analysis of clinical factors, imaging biomarkers, and potential no-reflow biomarkers through both univariate and multivariate analyses to identify predictors of futile recanalization. The comprehensive model includes clinical factors such as age, presence of diabetes, admission NIHSS score, and the number of stent retriever passes; imaging biomarkers like poor collaterals; and potential no-reflow biomarkers, notably disrupted blood-brain barrier (OR 4.321, 95% CI 1.794-10.405; p = 0.001), neutrophil-to-lymphocyte ratio (NLR; OR 1.095, 95% CI 1.009-1.188; p = 0.030), and D-dimer (OR 1.134, 95% CI 1.017-1.266; p = 0.024). The model demonstrated high predictive accuracy, with a C-index of 0.901 (95% CI 0.855-0.947) and 0.911 (95% CI 0.863-0.954) in the original and bootstrapping validation samples, respectively. Notably, the comprehensive model showed significantly improved predictive performance over models that did not include no-reflow biomarkers, evidenced by an integrated discrimination improvement of 8.86% (95% CI 4.34%-13.39%; p < 0.001) and a categorized reclassification improvement of 18.38% (95% CI 3.53%-33.23%; p = 0.015). This model, which leverages the potential of no-reflow biomarkers, could be especially beneficial in healthcare settings with limited resources. It provides a valuable tool for predicting futile recanalization, thereby informing clinical decision-making. Future research could explore further refinements to this model and its application in diverse clinical settings.

Ultrasound combined with Ki-67 to construct the prognostic model for radioactive iodine therapy outcomes in Graves' disease patients.

The aim of this study was to develop a prognostic model for radioactive iodine (RAI) therapy outcome in patients with Graves' disease. We enrolled 127 patients. Information on RAI therapy, ultrasound indexes of thyroid, and other lifestyle factors was collected. The competing risk model was used to estimate the multivariable-adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for nonhealing or recurrence of hyperthyroidism (NHRH). The performance of the model was assessed by receiver operator characteristic analysis and the Brier score and internally validated by bootstrap resampling. Then, a nomogram was developed. Forty-one cases (32.2%) of NHRH were documented. Positive Ki-67 expression, a higher dose of per-unit thyroid volume, and females showed lower risks of NHRH (all P < 0.05). The HR values (95% CI) were 0.42 (0.23, 0.79), 0.01 (0.00, 0.02), and 0.47 (0.25, 0.89), respectively. The bootstrap validation showed that the model had the highest accuracy and good calibration for predicting cumulative risk of NHRH at 180 days after RAI therapy (AUC = 0.772; 95% CI: 0.640-0.889, Brier score = 0.153). By decision curve analysis, the nomogram was shown to have a satisfactory net benefit between thresholds of 0.20 and 0.40. Ki-67, ultrasound volumetry, and scintigraphy techniques can play important roles in evaluating RAI therapy outcome in Graves' disease patients. The prediction nomogram shows reasonable accuracy in predicting NHRH.

Clinical prediction model for interdisciplinary biopsychosocial rehabilitation in osteoarthritis patients.

Osteoarthritis (OA) is a heterogenous condition, in which different subgroups are present. Individualized interdisciplinary multimodal pain treatments (IMPT) based on the biopsychosocial model have resulted in positive improvement of pain, health and disability in OA patients. Moreover, predictive factors for treatment success of IMPT in different musculoskeletal pain populations have been examined, but a clinical prediction model which informs whether an OA patient is expected to benefit or not from IMPT is currently lacking. The aim was to develop and internally validate a clinical prediction model to inform patient-tailored care based on identified predictors for positive or negative outcomes of IMPT in patients with OA. Longitudinal prospective cohort study. Center for Integral Rehabilitation at six locations in the Netherlands. Chronic OA patients. Data in this study were collected during January 2019 until January 2022. Participants underwent a 10-week IMPT program based on the biopsychosocial model. Treatment success was defined by a minimal decrease from baseline of 9 points on the Pain Disability Index (PDI). Candidate predictors were selected by experts in IMPT and literature review. Backward logistic regression analysis was performed to develop the clinical predication model and bootstrap validation was performed for internal validation. Overall, 599 OA patients were included, of which 324 experienced treatment success. Thirty-four variables were identified as possible predictors for good IMPT outcome. Age, gender, number of pain locations, PDI baseline score, maximal pain severity, use of pain medication and alcohol, work ability, brief illness perceptions questionnaire subscales timeline, consequences, identity and treatment control, pain catastrophizing scale and self-efficacy questionnaire score were found as predictors for treatment success. The internally validated model has an acceptable discriminative power of 0.71. This study reports a specific clinical prediction model for good outcome of IMPT in patients with OA. The internally validated model has an acceptable discriminative power of 0.71. After external validation, this model could be used to develop a clinically useful decision tool.

Risk Factors for 30-Day Mortality in Patients with Bacteremic Pneumonia Caused by Escherichia coli and Klebsiella pneumoniae: A Retrospective Study.

Escherichia coli and Klebsiella pneumoniae are prevalent Gram-negative microorganisms responsible for pneumonia, as well as the primary Enterobacteriaceae pathogens causing bacteremic pneumonia. The objective of this research is to analyze the risk factors associated with bacteremic pneumonia caused by these pathogens and develop a predictive model. This retrospective investigation encompassed a cohort of 252 patients diagnosed with Escherichia coli or Klebsiella pneumoniae-induced bacteremic pneumonia between 2018 and 2022. The primary endpoint was 30-day mortality, which was analyzed using multifactorial logistic regression, nomogram construction, and Bootstrap validation. Among the 252 patients diagnosed with Escherichia coli and Klebsiella pneumoniae, 65 succumbed to the disease while 187 survived. The overall 30-day mortality was found to be 25.8%. A multifactorial logistic regression analysis revealed that diastolic blood pressure, cerebrovascular diseases/transient ischemic attacks (TIA), immunosuppression, blood urea nitrogen, Pitt score, and CURB-65 score were statistically significant factors. The Nomogram model demonstrated an AUC of 0.954, which closely aligns with the Bootstrap-derived mean AUC of 0.953 (95% CI: 0.952-0.954). In patients with bacteremic pneumonia caused by Escherichia coli and Klebsiella pneumoniae, Low diastolic blood pressure (≤61 mmHg), pre-existing cerebrovascular disease/ transient ischemic attacks (TIA), immunosuppression status, elevated blood urea nitrogen levels (≥8.39 mmol/L), high Pitt score (≥3), and a high CURB-65 score (≥2) are all independent risk factors for Escherichia coli and Klebsiella pneumoniae bacteremic pneumonia, among which the first three warrant particular attention.

Routine 4D Cardiac CT to Identify Concomitant Transthyretin Amyloid Cardiomyopathy in Older Adults with Severe Aortic Stenosis.

Background Transthyretin amyloid cardiomyopathy (ATTR-CM) often coexists with severe aortic stenosis (AS). Although strain analysis from cardiac MRI and echocardiography was demonstrated to predict coexisting ATTR-CM, comparable data from four-dimensional (4D) cardiac CT are lacking despite wide availability. Purpose To evaluate the diagnostic performance of 4D cardiac CT-derived parameters in identifying ATTR-CM in older adults considered for transcatheter aortic valve implantation (TAVI). Materials and Methods This prospective single-center screening study for ATTR-CM included consecutive patients with severe AS considered for TAVI who underwent 4D cardiac CT between August 2019 and August 2021 approximately 1 day before technetium 99m (99mTc) 3,3-diphosphono-1,2-propanodicarboxylic-acid (DPD) scintigraphy. The diagnostic performance of CT-based left ventricular (LV), right ventricular, and left atrial dimensions, ejection fraction (EF), and myocardial strain were evaluated against 99mTc-DPD scintigraphy as the reference standard to identify ATTR-CM. Predictors and an unweighted cardiac CT score were validated with internal bootstrapping. The assignment of variables to the score was based on cutoff values achieving the highest Youden index J. Results Among 263 participants (mean age, 83 years ± 4.6 [SD]; 149 male and 114 female participants), 99mTc-DPD scintigraphy (Perugini grade 2 or 3) confirmed coexisting ATTR-CM in 27 (10.3%). CT-derived LV mass index, LV and LA global longitudinal strain (GLS), and relative apical longitudinal strain each predicted the presence of ATTR-CM with an area under the curve (AUC) of at least 0.70. Implementing these parameters with cutoff values of 81 g/m2 or higher, -14.9% or higher, less than 11.5%, and 1.7 or higher in the CT score, respectively, yielded high diagnostic performance (AUC = 0.89; 95% CI: 0.81, 0.94; P < .001) robust to internal bootstrapping validation (AUC = 0.88; 95% CI: 0.82, 0.94). If two criteria were fulfilled, the sensitivity and specificity in the detection of ATTR-CM were 96.3% (95% CI: 81.0, 99.9) and 58.9% (95% CI: 52.3, 65.2), respectively. Conclusion When compared against 99mTc-DPD scintigraphy as the reference standard, routine 4D cardiac CT in older adults considered for TAVI provided high diagnostic performance in the detection of concomitant ATTR-CM by assessing LV and left atrial GLS, relative apical longitudinal strain, and LV mass index. ClinicalTrials.gov registration no.: NCT04061213 © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Tavakoli and Onder in this issue.

Plasma proteomic analysis on neuropathic pain in idiopathic peripheral neuropathy patients.

Why only half of the idiopathic peripheral neuropathy (IPN) patients develop neuropathic pain remains unknown. By conducting a proteomics analysis on IPN patients, we aimed to discover proteins and new pathways that are associated with neuropathic pain. We conducted unbiased mass-spectrometry proteomics analysis on blood plasma from 31 IPN patients with severe neuropathic pain and 29 IPN patients with no pain, to investigate protein biomarkers and protein-protein interactions associated with neuropathic pain. Univariate modeling was done with linear mixed modeling (LMM) and corrected for multiple testing. Multivariate modeling was performed using elastic net analysis and validated with internal cross-validation and bootstrapping. In the univariate analysis, 73 proteins showed a p-value <.05 and 12 proteins showed a p-value <.01. None were significant after Benjamini-Hochberg adjustment for multiple testing. Elastic net analysis created a model containing 12 proteins with reasonable discriminatory power to differentiate between painful and painless IPN (false-negative rate 0.10, false-positive rate 0.18, and an area under the curve 0.75). Eight of these 12 proteins were clustered into one interaction network, significantly enriched for the complement and coagulation pathway (Benjamini-Hochberg adjusted p-value = .0057), with complement component 3 (C3) as the central node. Bootstrap validation identified insulin-like growth factor-binding protein 2 (IGFBP2), complement factor H-related protein 4 (CFHR4), and ferritin light chain (FTL), as the most discriminatory proteins of the original 12 identified. This proteomics analysis suggests a role for the complement system in neuropathic pain in IPN.

A 46 Long Non-Coding RNAs Expression Signature Accurately Predicts Relapse and Acts As an Independent Prognostic Factor in Pediatric Acute Myeloid Leukemia

Risk stratification using cytogenetics, mutations and minimal residual disease (MRD) has allowed to increase the cure rates of pediatric acute myeloid leukemia (AML), reaching now up to 70% in contemporary protocols. Nevertheless, approximately 30% of patients still experience relapse, indicating that there is a need to optimize stratification strategies. Recently, long non-coding RNA (lncRNA) expression has been shown to hold prognostic power in multiple cancer types. Here, we built a 46 relapse-related lncRNAs prognostic signature, named AML lnc46, using 790 pediatric acute myeloid leukemia transcriptomes obtained from the Therapeutically Applicable Research To Generate Effective Treatments (TARGET) repository. Bootstrap validation (bootstrap resampling times = 1000) was used for internal validation and showed that the signature performed good in predicting the 1-, 2-, and 3-year relapse-free survival (RFS), with an area under the curve of 0.825, 0.824, and 0.822, respectively. Moreover, external validation using the TARGET repository validated the efficiency of the signature. Most importantly, we demonstrate that AML lnc46 not only is an independent predictor of RFS, but also identifies patients who were assigned a different risk-group by currently-used classification methods. In conclusion, the identified AML lnc46 might, after further prospective validation, provide additional information to guide management of pediatric AML patients. Figure 1. (A) Kaplan-Meier survival curve of relapse-free survival of high-and-low risk AML lnc46 pediatric AML patients, (B) Univariate and multivariate independent Cox regression analyses of known clinical and molecular characteristics and the AML lnc46 signature.

A bootstrap functional central limit theorem for time-varying linear processes

We provide a functional central limit theorem for a broad class of smooth functions for possibly non-causal multivariate linear processes with time-varying coefficients. Since the limiting processes depend on unknown quantities, we propose a local block bootstrap procedure to circumvent this inconvenience in practical applications. In particular, we prove bootstrap validity for a very large class of processes. Our results are illustrated by some numerical examples.

Risk Factors and Prognostic Scoring Model for 28-Day Mortality in Hematologic Patients with Stenotrophomonas M altophilia Bacteremia

Objective: Patients with hematologic diseases are at a higher risk for Stenotrophomonasmaltophilia(SM) bacteremia, which poses significant burden on morbidity and mortality. The aim of this study was to evaluate the early predictors of 28-day mortality in hematologic patients with SM bacteremia and to construct a reliable prediction model which can be used to identify patients at higher risk for mortality. Methods: This retrospective analysis of 131 hospitalized patients diagnosed with SM bacteremia was conducted in a tertiary blood disease center in Tianjin, China, from January 2012 to June 2023. Univariate analysis and multivariate logistic regression analysis were used to identify prognostic risk factors and develop a prognostic mortality scoring system and internally validated by bootstrap method (resampling = 1000). Results: The overall 28-day mortality rate in our study was 29.77% (39/131). The mean age was 45 (IQR 30.5-56.0) years and 70 individuals (53.44%) were diagnosed with myeloid malignancies. The existing condition severity-related scores did not perform well in predicting 28-day mortality for the patients in our population (AUROC values: SOFA: 0.842; ECOG: 0.831; APACHE Ⅱ: 0.709; SAPS Ⅱ: 0.675; aCCI: 0.654; Pitt score: 0.629). Through univariate and multivariate analysis, we identified four independent risk factors associated with 28-day mortality: SOFA score (OR = 3.67, 95% CI 1.99-6.78, P &amp;lt; 0.001; score = corresponding SOFA score), age ≥ 60 years (OR = 6.95, 95% CI 1.44-33.55, P = 0.0158; score = 2), bloody sputum or hemoptysis (OR = 5.32, 95% CI 1.08-26.29, P = 0.0402; score = 2) and neutropenia duration ≥ 25 days (OR = 3.92, 95% CI 1.01-15.26, P = 0.0487; score = 1). These factors were used to develop a prognostic scoring model, which yielded good discrimination (AUC = 0.889, 95% CI: 0.829-0.949) and calibration (Hosmer-Lemeshow goodness-of-fit test P = 0.998), with similar performance in the internal bootstrap validation (AUC = 0.888). Furthermore, all patients were divided into the high-risk group (≥ 7 points) and the low-risk group (&amp;lt; 7 points) based on the model and compared using Kaplan-Meier curves and log-rank tests. Additionally, we founded that the use of trimethoprim-sulfamethoxazole (TMP/SMX) after infection was associated with a better prognosis ( P = 0.012). Conclusions: The prediction model, utilizing SOFA score, age, bloody sputum or hemoptysis and neutropenia duration, proves to be an effective tool for early identification of hematologic patients with SM bacteremia at high risk of mortality.

Association of genetic and sulcal traits with executive function in congenital heart disease.

Persons with congenital heart disease (CHD) are at increased risk of neurodevelopmental disabilities, including impairments to executive function. Sulcal pattern features correlate with executive function in adolescents with single-ventricle heart disease and tetralogy of Fallot. However, the interaction of sulcal pattern features with genetic and participant factors in predicting executive dysfunction is unknown. We studied sulcal pattern features, participant factors, and genetic risk for executive function impairment in a cohort with multiple CHD types using stepwise linear regression and machine learning. Genetic factors, including predicted damaging de novo or rare inherited variants in neurodevelopmental disabilitiesrisk genes, apolipoprotein E genotype, and principal components of sulcal pattern features were associated with executive function measures after adjusting for age at testing, sex, mother's education, and biventricular versus single-ventricle CHD in a linear regression model. Using regression trees and bootstrap validation, younger participant age and larger alterations in sulcal pattern features were consistently identified as important predictors of decreased cognitive flexibility with left hemisphere graph topology often selected as the most important predictor. Inclusion of both sulcal pattern and genetic factors improved model fit compared to either alone. We conclude that sulcal measures remain important predictors of cognitive flexibility, and the model predicting executive outcomes is improved by inclusion of potential genetic sources of neurodevelopmental risk. If confirmed, measures of sulcal patterning may serve as early imaging biomarkers to identify those at heightened risk for future neurodevelopmental disabilities.

An Explainable Artificial Intelligence Model Proposed for the Prediction of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome and the Identification of Distinctive Metabolites

Background: Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a complex and debilitating illness with a significant global prevalence, affecting over 65 million individuals. It affects various systems, including the immune, neurological, gastrointestinal, and circulatory systems. Studies have shown abnormalities in immune cell types, increased inflammatory cytokines, and brain abnormalities. Further research is needed to identify consistent biomarkers and develop targeted therapies. This study uses explainable artificial intelligence and machine learning techniques to identify discriminative metabolites for ME/CFS. Material and Methods: The model investigates a metabolomics dataset of CFS patients and healthy controls, including 26 healthy controls and 26 ME/CFS patients aged 22–72. The dataset encapsulated 768 metabolites into nine metabolic super-pathways: amino acids, carbohydrates, cofactors, vitamins, energy, lipids, nucleotides, peptides, and xenobiotics. Random forest methods together with other classifiers were applied to the data to classify individuals as ME/CFS patients and healthy individuals. The classification learning algorithms’ performance in the validation step was evaluated using a variety of methods, including the traditional hold-out validation method, as well as the more modern cross-validation and bootstrap methods. Explainable artificial intelligence approaches were applied to clinically explain the optimum model’s prediction decisions. Results: The metabolomics of C-glycosyltryptophan, oleoylcholine, cortisone, and 3-hydroxydecanoate were determined to be crucial for ME/CFS diagnosis. The random forest model outperformed the other classifiers in ME/CFS prediction using the 1000-iteration bootstrapping method, achieving 98% accuracy, precision, recall, F1 score, 0.01 Brier score, and 99% AUC. According to the obtained results, the bootstrap validation approach demonstrated the highest classification outcomes. Conclusion: The proposed model accurately classifies ME/CFS patients based on the selected biomarker candidate metabolites. It offers a clear interpretation of risk estimation for ME/CFS, aiding physicians in comprehending the significance of key metabolomic features within the model.

Sea clutter radar target detector based on autoregressive sieve bootstrap

The autoregressive sieve (ARS) bootstrap method can capture the correlation of time series and expand the original data set to the required volume. In this study, we introduce the ARS bootstrap into the matrix constant false alarm rate (CFAR) method to counter the inability of the Monte Carlo methods to extract the detection threshold due to the limited number of pulses under a certain false alarm probability. We assume the echo data of each range cell in a coherent processing interval (CPI) to be a stationary time series, and regard target detection as a two-sample hypothesis testing problem. We propose a statistic to test the equivalence of the autocovariance corresponding to two potential time series by using the correlation between pulses in a short pulse dataset, and analyze the validity of the ARS bootstrap for this statistic. The statistic distribution is obtained according to the bootstrap procedure and the threshold is obtained under fixed false alarm probabilities. Numerical experiments on the simulated and real sea clutter data were conducted to compare the detection results of the partial matrix CFAR detectors for the short and long pulse cases. With the introduction of the bootstrap method, we do not need to model the clutter amplitude distribution and the related statistical characteristics during target detection. To improve the detection probability with low signal-to-clutter ratio (SCR), we propose the two-sample autocovariance vector (ACV) detector based on the proposed statistic and use the correlation properties between pulses. We observed a 28% increase in detection probability when the SCR was 0 dB compared to other matrix CFAR methods, such as detectors based on Logarithmic Euclidean (LE) distance, Kernel Function-based Symmetric Kullback-Leibler (KSKL) divergence, and Total Bregman divergence (TBD) for the short pulse case.

External Validation and Updating of a Statistical Civilian-Based Suicide Risk Model in US Naval Primary Care

Suicide remains an ongoing concern in the US military. Statistical models have not been broadly disseminated for US Navy service members. To externally validate and update a statistical suicide risk model initially developed in a civilian setting with an emphasis on primary care. This retrospective cohort study used data collected from 2007 through 2017 among active-duty US Navy service members. The external civilian model was applied to every visit at Naval Medical Center Portsmouth (NMCP), its NMCP Naval Branch Health Clinics (NBHCs), and TRICARE Prime Clinics (TPCs) that fall within the NMCP area. The model was retrained and recalibrated using visits to NBHCs and TPCs and updated using Department of Defense (DoD)-specific billing codes and demographic characteristics, including expanded race and ethnicity categories. Domain and temporal analyses were performed with bootstrap validation. Data analysis was performed from September 2020 to December 2022. Visit to US NMCP. Recorded suicidal behavior on the day of or within 30 days of a visit. Performance was assessed using area under the receiver operating curve (AUROC), area under the precision recall curve (AUPRC), Brier score, and Spiegelhalter z-test statistic. Of the 260 583 service members, 6529 (2.5%) had a recorded suicidal behavior, 206 412 (79.2%) were male; 104 835 (40.2%) were aged 20 to 24 years; and 9458 (3.6%) were Asian, 56 715 (21.8%) were Black or African American, and 158 277 (60.7%) were White. Applying the civilian-trained model resulted in an AUROC of 0.77 (95% CI, 0.74-0.79) and an AUPRC of 0.004 (95% CI, 0.003-0.005) at NBHCs with poor calibration (Spiegelhalter P < .001). Retraining the algorithm improved AUROC to 0.92 (95% CI, 0.91-0.93) and AUPRC to 0.66 (95% CI, 0.63-0.68). Number needed to screen in the top risk tiers was 366 for the external model and 200 for the retrained model; the lower number indicates better performance. Domain validation showed AUROC of 0.90 (95% CI, 0.90-0.91) and AUPRC of 0.01 (95% CI, 0.01-0.01), and temporal validation showed AUROC of 0.75 (95% CI, 0.72-0.78) and AUPRC of 0.003 (95% CI, 0.003-0.005). In this cohort study of active-duty Navy service members, a civilian suicide attempt risk model was externally validated. Retraining and updating with DoD-specific variables improved performance. Domain and temporal validation results were similar to external validation, suggesting that implementing an external model in US Navy primary care clinics may bypass the need for costly internal development and expedite the automation of suicide prevention in these clinics.