Background: Obstructive sleep apnea (OSA) is strongly associated with metabolic dysfunction and may contribute to the development of hepatic disorders such as metabolic dysfunction-associated steatotic liver disease (MASLD) and liver fibrosis. Early identification of hepatic involvement in OSA is important for preventing long-term complications. This study aimed to develop and compare artificial intelligence (AI)-based models using non-invasive clinical parameters to predict steatosis, MASLD, and liver fibrosis in patients with OSA. Methods: This prospective study included patients diagnosed with OSA based on polysomnography (PSG). Hepatic steatosis and fibrosis were assessed non-invasively using the controlled attenuation parameter (CAP) and liver stiffness measurement (LSM) obtained by FibroScan. A total of 700 patients were screened, and 517 patients who met the inclusion criteria were analyzed, including 42 demographic, anthropometric, laboratory, and sleep-related variables. A total of 42 variables were recorded, comprising 37 candidate predictors and 4 target outcomes. Following feature selection, 11 of the 37 candidates were retained for model training. Four supervised machine learning algorithms—logistic regression, support vector machine, random forest, and artificial neural network were trained to predict steatosis, MASLD, and liver fibrosis. Model performance was evaluated by accuracy, F1-score, and area under the ROC curve. Feature importance analysis identified the most influential predictors. Results: Among the four models, the Random Forest algorithm achieved the best performance, with an accuracy of 96% for binary classification; in an exploratory five-class analysis, overall accuracy reached 92%, although predominantly driven by the majority F0 class. The most influential predictors were apnea–hypopnea index (AHI), minimum oxygen saturation (minSpO ₂ ), and oxygen desaturation index. A sensitivity analysis excluding elastography-derived features (CAP and LSM) was performed using 9 predictor variables (Model 2). Model 2 achieved best AUC values of 0.700, 0.706, and 0.634 for steatosis, MASLD, and fibrosis, respectively, compared with 0.967, 0.969, and 0.985 in the original 11-feature model (Model 1). Conclusion: AI-based models, particularly Random Forest, demonstrated high predictive performance in predicting MASLD and liver fibrosis in OSA patients. These models may support detection and risk stratification of hepatic involvement in sleep-disordered breathing.

Background: Pulmonary involvement is one of the most serious extra-articular manifestations of rheumatoid arthritis (RA), frequently remaining clinically silent until irreversible lung damage develops. This study aimed to determine the prevalence of high-resolution computed tomography (HRCT)-detected pulmonary involvement in patients with RA; compare the clinical, laboratory, and pulmonary function characteristics of patients with and without pulmonary involvement; and identify factors associated with pulmonary abnormalities. Methods: In this single-center retrospective cross-sectional study, 141 patients fulfilling the 2010 ACR/EULAR classification criteria for RA who had undergone thoracic HRCT were included. Clinical characteristics, laboratory findings, disease activity scores, pulmonary function tests, and HRCT findings were compared between patients with and without pulmonary involvement. Logistic regression analyses were performed to identify independent predictors of pulmonary involvement, and receiver operating characteristic (ROC) analysis was used to evaluate the diagnostic performance of serum uric acid. Results: HRCT identified pulmonary involvement in 33 of 141 patients (23.4%), with nonspecific interstitial pneumonia being the most common radiological pattern (39.4%), followed by usual interstitial pneumonia (24.2%). Patients with pulmonary involvement had significantly higher serum uric acid concentrations (5.65 ± 1.62 vs. 4.51 ± 1.31 mg/dL, p < 0.001), lower serum albumin levels (38.8 ± 4.9 vs. 40.9 ± 3.5 g/L, p = 0.008), higher platelet counts (p = 0.006), and lower forced vital capacity and predicted FEV₁ values (p = 0.010 and p = 0.019, respectively). In multivariable logistic regression analysis, serum uric acid remained the only independent predictor of pulmonary involvement (odds ratio, 1.656; 95% confidence interval, 1.187–2.311; p = 0.003). ROC analysis demonstrated acceptable discriminatory performance for serum uric acid (AUC = 0.713), with an optimal cutoff value of >5 mg/dL yielding 68% sensitivity and 70% specificity. Conclusions: Nearly one-quarter of patients with RA demonstrated HRCT-detected pulmonary involvement despite having a similar prevalence of respiratory symptoms. Elevated serum uric acid was independently associated with pulmonary involvement and may represent an inexpensive, readily available biomarker to identify patients who may benefit from targeted pulmonary evaluation. These findings support the integration of simple laboratory markers with clinical assessment to facilitate earlier recognition of RA-associated pulmonary disease.

Background: Community-acquired pneumonia (CAP) is a common infectious disease associated with substantial morbidity and mortality. Acute kidney injury (AKI) is a frequent complication of CAP and may adversely affect clinical outcomes. This study aimed to determine the frequency of AKI and identify factors associated with AKI in hospitalized patients with CAP. Methods: This retrospective cohort study was conducted at a tertiary care university hospital in Türkiye and included patients aged >18 years hospitalized with CAP between July 2024 and July 2025. Only patients initially admitted to the general ward were included. Patients receiving dialysis for end-stage renal disease, those with active malignancy, those receiving immunosuppressive therapy, and those with incomplete data were excluded. Demographic characteristics, vital signs, laboratory findings, comorbidities, smoking status, and medication histories were recorded. AKI was defined according to KDIGO criteria using the closest clinically stable pre-admission serum creatinine value obtained within 7–365 days before hospitalization as baseline. Results: Among 161 patients, AKI was identified in 60 (37.3%), including 43 patients who fulfilled the KDIGO criteria at hospital admission and 17 who developed AKI during hospitalization. Eight patients with AKI (13.3%) required hemodialysis, although none became permanently dialysis-dependent. Patients with AKI had longer hospital stays (p = 0.038) and a greater need for intensive care (p = 0.028). They were older and had higher respiratory rates, blood urea nitrogen (BUN), and red cell distribution width (RDW) values and lower diastolic blood pressure. DM and CKD were more common in the AKI group. In multivariable analysis, confusion (adjusted OR 5.03, 95% CI 1.01–25.02; p = 0.048), CKD (adjusted OR 6.00, 95% CI 1.28–28.13; p = 0.023), and higher BUN levels (adjusted OR 1.095 per 1 mg/dL increase, 95% CI 1.046–1.146; p < 0.001) were independently associated with AKI. These associations were retained in a Firth penalized logistic regression sensitivity analysis. Conclusions: AKI was common among hospitalized patients with CAP and was associated with longer hospital stay and greater intensive care utilization. Confusion, pre-existing CKD, and elevated BUN levels were independently associated with AKI, whereas DM was not independently associated after multivariable adjustment. Given that most AKI cases were already present at hospital admission, these findings should be interpreted as associations rather than predictors of incident AKI.