{"doi":"10.5281/zenodo.19445092","title":"Unraveling the Complex Interplay of Neurobehavioral Disorders, Cardiometabolic Factors, and Genetics; An Integrative Analysis Using All of Us Datasets","abstract":"Objective - Cardiometabolic diseases (CMD), including chronic kidney disease (CKD), hypertension (HTN), diabetes mellitus (DM), and obesity, pose significant public health challenges due to their association with end-organ dysfunction and neurological complications. These disorders often coexist and synergistically contribute to accelerated cognitive decline, epilepsy, and dementia. Among the biomarkers used to track kidney function, estimated glomerular filtration rate (eGFR), calculated using serum creatinine, plays a central role. However, the links between creatinine levels, eGFR, and neurobehavioral disorders remain insufficiently characterized in large-scale populations. Methodology - This study aimed to develop a multi-organ, biomarker-driven predictive framework for identifying individuals at risk for epilepsy and dementia. Machine learning (ML) models—including decision trees, ordinal logistic regression, and XGBoost—were applied to two datasets: the Texas Kidney Foundation cohort (2010–2023; N = 14,561) and the All of Us dataset (N = 200,000) for external validation. Input variables included biometric, laboratory, and access-related features such as serum creatinine, eGFR, hemoglobin A1C, blood glucose, body mass index (BMI), systolic and diastolic blood pressure, age, and insurance status. Results - Models consistently demonstrated that objective biomarkers were the most influential predictors. Self-reported population group classification was included in the analysis but showed low feature importance and did not improve predictive accuracy for kidney function or neurological outcomes. Overreliance on such general classifications in traditional clinical algorithms may obscure physiologically relevant risk indicators. In nephrology, this could lead to inaccurate eGFR estimation in certain subgroups, potentially delaying diagnosis and treatment, increasing the risk of downstream neurological complications. Conclusion and Implications - XGBoost models yielded strong predictive performance, with area under the curve (AUC) values of 0.74 for epilepsy and 0.75 for dementia. Key predictive features for neurobehavioral disorders included creatinine, eGFR, blood glucose, A1C, BMI, age, and blood pressure—reinforcing the potential of integrated, biomarker-based models in early detection. These findings support a paradigm shift toward precision modeling that relies on measurable clinical variables rather than generalized demographic proxies. The adoption of such approaches may enhance early identification of patients at risk for both renal and neurological deterioration. Future research will employ Mendelian randomization to evaluate causality between kidney function and neurological outcomes and explore deployment of real-time AI/ML tools integrated into clinical workflows. This work holds translational value for preventative neurology, cardiometabolic care, and personalized medicine.","journal":"Zenodo (CERN European Organization for Nuclear Research)","year":2025,"id":587998,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":0,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9409,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2025-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":454107,"name":"Donald E. Wesson","orcid":null,"position":1,"is_corresponding":false},{"id":40591,"name":"Kevin Smith","orcid":"0000-0002-6163-191X","position":2,"is_corresponding":false},{"id":1504334,"name":"Tiffany Jones-Smith","orcid":null,"position":3,"is_corresponding":false},{"id":528289,"name":"James W. Baurley","orcid":"0000-0003-4116-7723","position":4,"is_corresponding":false},{"id":1504333,"name":"Adetoun A. Musa","orcid":null,"position":0,"is_corresponding":true}],"reference_count":0,"raw_metadata":null,"created_at":"2026-07-19T02:59:43.096742Z","pmid":null,"pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}