Early identification of high-risk individuals is essential to guide infection-prevention strategies and optimize antibiotic stewardship in this vulnerable population. To identify independent risk factors associated with hospital-acquired infections in elderly ICU patients following antibiotic use and to develop and internally validate a clinical risk prediction model for early infection detection. A retrospective cohort study was conducted in the ICU of Nanjing First Hospital, Nanjing Medical University. A total of 120 patients aged ≥65 years, with ICU stay >48 hours, no documented infection at ICU admission and antibiotic exposure within 48 hours before or at ICU admission were included. Demographic data, comorbidities, Sequential Organ Failure Assessment (SOFA) scores, antibiotic exposure characteristics, invasive device use and nutritional support were collected from electronic health records. Hospital-acquired infections occurred in 46 patients (38.3%). Independent predictors included advanced age (odds ratio [OR] 1.08 per year), higher SOFA score (OR 1.25 per point), diabetes mellitus (OR 1.45), chronic kidney disease (OR 1.65), use of central venous catheters (OR 1.75), mechanical ventilation (OR 1.85), Foley catheterization (OR 1.55), broad-spectrum antibiotic use (OR 1.50), longer antibiotic duration (OR 1.20 per day) and prolonged ICU stay (all p<0.05). The prediction model demonstrated good discrimination (AUC-ROC = 0.82), which improved slightly after variable refinement (AUC-ROC = 0.83). Cross-validated performance remained robust (AUC = 0.80). A multivariable risk prediction model using routinely available clinical parameters demonstrated good internal validity and may assist clinicians in early identification of high-risk patients, enabling targeted infection prevention and improved antibiotic stewardship.

Construction of a risk prediction model for oral frailty in hospitalized elderly patients with chronic diseases.

Interpretable machine learning model for predicting refeeding syndrome after colorectal cancer surgery.

Multicentre development and validation of a risk model integrating immunotherapy and coagulation biomarkers for thrombosis in autoimmune neurological disorders.

Seizure risk prediction using machine learning following glioma resection surgery in seizure-naïve patients.

Tabular foundation models as a new portable standard in local surgical risk prediction.

Using principles of Mendelian genetics, probability theory, and mutation-specific knowledge, Mendelian risk prediction models identify those at high risk of carrying a heritable cancer susceptibility variant and assess future risk of cancer. Our previously-validated Fam3PRO model is a generalizable and computationally efficient Mendelian risk prediction framework that incorporates an arbitrary number of gene-cancer associations. In practice, from a model training perspective, there may be uncertainty in estimating the population-level model parameters necessary for rare gene-cancer associations. From a clinical perspective, it may be infeasible to obtain a detailed patient family history for many cancers. Motivated by the context of pre-screening for germline testing of a broad hereditary cancer gene panel, we propose a Mendelian model that aggregates information across genes and cancers, reducing patient burden and bypassing the need for robust parameter estimation for rare genes and syndromes. We evaluated this aggregate model through simulations and applied it to two independent clinical cohorts. We show that when the clinical goal is to assess patient risk of carrying a pathogenic variant for any cancer susceptibility gene, the aggregate model can give results comparable to a Mendelian model that considers many genes and cancers individually, while greatly simplifying model assumptions and user input.

Development and validation of a predictive model for severe complications following ovarian cancer cytoreductive surgery combined with hyperthermic intraperitoneal chemotherapy.

Association between red cell distribution width-albumin ratio and osteoarthritis in middle-aged and older adults: Analysis of NHANES data (1999-2018).

External validation and recalibration of cardiovascular risk scores for prediction of 10-year risk of fatal cardiovascular disease: a prospective, observational, population-based cohort analysis of adults in Mexico City.

Social determinants of health (SDoH), including poverty and social isolation, have emerged as important contributors to adverse outcomes of chronic diseases. However, their impact on patients with metabolic dysfunction-associated steatotic liver disease (MASLD) remains poorly characterised. This study aimed to assess the association between social vulnerability in MASLD and liver-related and cardiovascular outcomes. We conducted a population-based retrospective cohort study using the TriNetX network, which aggregates de-identified electronic health records from healthcare systems across the U.S. Patients with MASLD and at least one International Classification of Diseases, Tenth Revision (ICD-10) code for documented social vulnerability (Z59.5, Z59.6, Z56.0 and Z60.2) were compared to non-socially vulnerable individuals. Outcomes, including major adverse liver outcomes (MALO), major adverse cardiovascular events (MACE), hepatocellular carcinoma (HCC) and other cardiovascular complications, were assessed over a 5-year follow-up using Cox proportional hazards models. Individuals with MASLD and documented social vulnerability were at higher risk of MALO (8.3% vs. 4.4%; HR 1.69, 95% CI: 1.42-2.01). Cardiovascular morbidity was consistently elevated including MACE (22.7% vs. 12.5%; HR 1.64, 95% CI: 1.46-1.83), arrhythmias (34.4% vs. 17.8%; HR 1.81, 95% CI: 1.64-2.00) and heart failure (12.4% vs. 6.7%; HR 1.64, 95% CI: 1.42-1.89). The incidence of HCC did not differ between documented socially vulnerable and non-socially vulnerable individuals with MASLD. Documented social vulnerability is independently associated with higher risks of liver and cardiovascular complications in MASLD. These findings underscore the importance of integrating SDoH into MASLD management and risk prediction models to address disparities in long-term outcomes.

Designing a machine learning model for predicting cardiovascular events using the triglyceride-glucose index: a cohort study.

To compare predictive accuracy of 3-step theory of suicide (3ST) factor scores derived from natural language processing of Veterans Health Administration (VHA) clinical progress notes versus a model that underlies VHA's Recovery Engagement and Coordination for Health-Veterans Enhanced Treatment (REACH VET) program retrained to predict the combined outcome of suicide attempt or suicide death, and to compare characteristics of patients accurately predicted by both approaches. As health systems incorporate risk prediction models to guide suicide prevention efforts, it is important to evaluate their predictive accuracy and to consider the benefits of different modeling approaches. A comparative cohort design in which both risk prediction approaches were evaluated for the same random sample (n=162,132) of VHA patients alive on May 1, 2018, who had clinical encounters during the 4 weeks before that date. At the highest risks (top 1%-5%), the model based on REACH VET variables outperformed the 3ST approach in terms of positive predictive value and false-negative rate. Among patients who attempted or died by suicide, uniquely identified by the 3ST approach and not by the retrained REACH VET model, none had attempted suicide during the prior 6 months, emergency department visits during the prior month, discharges from mental health hospitalizations during the prior 12 months, or a diagnosis of bipolar disorder during the prior 24 months. Additional research is recommended to further prepare 3ST factor scores based on NLP of clinical progress notes for use in clinical decision-making.

Predictive value of machine learning for mortality risk in aortic dissection: a systematic review and meta-analysis.

Artificial Intelligence Applications for Automated Data Extraction and Secondary Use of Clinical Information in Uro-oncology: A Systematic Review.

Cognitive frailty risk prediction models in patients with chronic kidney disease in China: a systematic review and meta-analysis.

Osteoporotic fractures (OPF) and hypertension frequently co-occur in older adults, yet comprehensive datasets integrating clinical, pharmacological, and longitudinal outcome data remain scarce. We describe a longitudinal dataset derived from the Osteoporotic Fracture Registration System at Kunshan Hospital, Jiangsu University, including patients aged ≥ 50 years hospitalized for OPF between 2017 and 2024. A total of 4,782 patients were initially registered. After applying predefined eligibility criteria, 4,325 patients were included in the final analytical cohort. The dataset integrates demographic, clinical, and pharmacologic variables with long-term outcomes on mortality and refracture through deterministic linkage with regional health and mortality registries. Longitudinal antihypertensive prescription records (n = 42,367) were linked via the Kunshan Municipal Health Data Integration Platform, enabling detailed characterization of medication exposure patterns over time. Technical validation, including survival analysis, propensity score methods, and risk prediction modeling, was conducted to assess internal consistency and illustrate potential applications. This structured and de-identified dataset provides a quality-checked resource to support future research in osteoporosis, cardiovascular comorbidity, multimorbidity, and real-world comparative effectiveness studies.

A New Predictive Model for Radiation Necrosis Risk based on PTV-enriched Blood Inflammatory Biomarkers in Patients with Brain Metastases Treated with Stereotactic Radiosurgery.

Risk prediction models for community-acquired pneumonia (CAP) in hospitalised children are limited, especially in upper-middle-income countries. We developed a PaEdiatric Pneumonia Severity (PEPS) score to aid early risk stratification and examined its discriminatory ability. Prospective cohort study. Between April 2022 and April 2023, children aged 1 month to <12 years hospitalised with radiographic-confirmed CAP were enrolled in Sabah, Malaysia. Key clinical factors were collected to develop risk profiles for mild, moderate, and severe CAP. The PEPS score was derived from the final multivariable model and its ability to discriminate CAP severity assessed. Among 868 children, 639 (73.6%) had mild, 83 (9.6%) moderate, and 146 (16.8%) severe CAP. Independent factors associated with moderate/severe CAP were age <6 months (adjusted OR (ORadj)=6.81, 95% CI 4.65 to 9.98), partial/unvaccinated status (ORadj=2.66, 95% CI 1.47 to 4.81), vomiting/refusal of feeds (ORadj=1.77, 95% CI 1.19 to 2.61), hypoxaemia (oxygen saturation 90-93% (ORadj=2.38, 95% CI 1.22 to 4.64); <90% (ORadj=28.88, 95% CI 13.37 to 62.35)) and anaemia (ORadj=1.76, 95% CI 1.17 to 2.67). The model demonstrated excellent discrimination in identifying risk of children having moderate/severe CAP (c-statistic=0.81, 95% CI 0.78 to 0.84). The developed PEPS score (0-13) stratified children into low-risk (score 0-1), moderate-risk (score 2-4) and high-risk (score ≥5) groups with observed rates of moderate/severe CAP 10.9%, 29.1%, and 76.9%, respectively, and demonstrated strong predictive performance for moderate/severe CAP (c-statistic=0.81, 95% CI 0.77 to 0.84). The PEPS score, the first developed in an upper-middle-income country using radiographic-confirmed CAP, accurately discriminates mild from moderate/severe hospitalised CAP. It may support case management in similar settings, but requires external validation before broader implementation.

Development and validation of a multidimensional indicator-based risk prediction model for gestational diabetes mellitus: a nested case-control study.