Search Results (217)

PCNL treatment is often associated with complications of hemorrhagic or infectious origin, which can result in prolonged hospitalization. This study aims to develop predictive models using machine learning (ML) techniques to anticipate these outcomes. Multiple ML algorithms—including Logistic Regression, Decision Tree, Random Forest, and Extreme Gradient Boosting—were evaluated on separate validation and test datasets. The Random Forest model achieved the highest predictive performance for hospitalization need (AUC 0.726/0.736) and infectious complications (AUC 0.799/0.735). Threshold adjustment was applied to increase sensitivity, reducing false negatives. The interpretability of the models was ensured through SHAP analysis, identifying clinically meaningful variables. Risk factors for both hospitalization and infectious complications models included nephrostomy drainage, a neutrophils percentage higher than 80, Guy’s score of grade 4, leukocytes level higher than 15 or lower than 4.5, and balloon dilation, while protective features included tubeless intervention, easy localization of a stone, negative culture, and microorganism results. However, no model achieved acceptable performance for predicting hemorrhagic complications, likely due to limited data. These results suggest that AI-based models can contribute to risk stratification after PCNL. Further experiments with larger, multi-center datasets are needed to confirm these findings and improve the generalizability of the models. Full article

Background: Myasthenia gravis (MG), a chronic autoimmune disorder with variable disease trajectories, presents considerable challenges for clinical stratification and acute care management. This systematic review evaluated machine learning models developed for prognostic assessment in patients with MG. Methods: Following PRISMA guidelines, we systematically searched PubMed, Embase, and Scopus for relevant articles published from January 2010 to May 2025. Studies using machine learning techniques to predict MG-related outcomes based on structured or semi-structured clinical variables were included. We extracted data on model targets, algorithmic strategies, input features, validation design, performance metrics, and interpretability methods. The risk of bias was assessed using the Prediction Model Risk of Bias Assessment Tool. Results: Eleven studies were included, targeting ICU admission (n = 2), myasthenic crisis (n = 1), treatment response (n = 2), prolonged mechanical ventilation (n = 1), hospitalization duration (n = 1), symptom subtype clustering (n = 1), and artificial intelligence (AI)-assisted examination scoring (n = 3). Commonly used algorithms included extreme gradient boosting, random forests, decision trees, multivariate adaptive regression splines, and logistic regression. Reported AUC values ranged from 0.765 to 0.944. Only two studies employed external validation using independent cohorts; others relied on internal cross-validation or repeated holdout. Of the seven prognostic modeling studies, four were rated as having high risk of bias, primarily due to participant selection, predictor handling, and analytical design issues. The remaining four studies focused on unsupervised symptom clustering or AI-assisted examination scoring without predictive modeling components. Conclusions: Despite promising performance metrics, constraints in generalizability, validation rigor, and measurement consistency limited their clinical application. Future research should prioritize prospective multicenter studies, dynamic data sharing strategies, standardized outcome definitions, and real-time clinical workflow integration to advance machine learning-based prognostic tools for MG and support improved patient care in acute settings. Full article

Background: Upper tract urothelial carcinoma (UTUC) is a rare and aggressive malignancy with limited prognostic tools to predict disease progression. Due to its low incidence, the molecular pathogenesis of UTUC remains poorly understood, and few studies have explored transcriptomic profiling in this setting. Identifying gene expression biomarkers associated with progression may help improve risk stratification and guide postoperative management. Methods: In this study, we applied a machine learning approach to gene expression data from radical nephroureterectomy (RNU) specimens of 17 consecutive patients with pT2 or pT3 UTUC treated at our institution. RNA was extracted from formalin-fixed paraffin-embedded tissues and sequenced using the Ion AmpliSeq™ Transcriptome Human Gene Expression Kit on an Illumina HiSeq 2500 platform. Differential gene expression was assessed using DESeq2, and results were visualized with volcano plots. Predictive power was evaluated through logistic regression and receiver operating characteristic (ROC) analysis. Gene Ontology enrichment analysis was used to explore biological pathways. Results: A total of 76 genes were differentially expressed between progressive and non-progressive patients. A random forest classifier identified ten key genes with prognostic potential. Validation with logistic regression yielded an area under the ROC curve (AUC) of 0.88, indicating high discriminative ability. These genes were associated with immune regulation, cell cycle control, and tumor progression. Conclusions: This pilot study demonstrates the potential of integrating machine learning with transcriptomic analysis to identify prognostic biomarkers in UTUC. Further validation in larger, independent cohorts is needed to confirm these findings and support their clinical application. Full article

Species and phylogenetic diversity play vital roles in sustaining the structure, function, and resilience of plant communities, particularly in tropical rainforests. However, the mechanisms according to which environmental filtering and competitive exclusion influence diversity across forest layers remain insufficiently understood. In this study, we investigated the species and phylogenetic diversity patterns in two representative tropical rainforest sites—Bawangling and Jianfengling—within Hainan Tropical Rainforest National Park, China, focusing on communities associated with the endangered species Hopea hainanensis. We employed a one-way ANOVA and Pearson’s correlation analyses to examine the distribution characteristics and interrelationships among diversity indices and used Mantel tests to assess the correlations with environmental variables. Our results revealed that the plant community in Jianfengling exhibited a significantly higher species richness at the family, genus, and species levels (a total of 288 plant species have been recorded, belonging to 82 families and 183 genera) compared to that in Bawangling (a total of 212 plant species, belonging to 75 families and 162 genera). H. hainanensis held the highest importance value in the middle tree layer across both sites (IV(BWL) = 12.44; IV(JFL) = 5.73), while dominant species varied notably among other forest layers, indicating strong habitat specificity. Diversity indices, including the Simpson index, the Shannon–Wiener index, and Pielou’s evenness, were significantly higher in the large shrub layer of Jianfengling, whereas Bawangling showed a relatively higher Shannon–Wiener index in the middle shrub layer. Phylogenetic diversity (PD) and the phylogenetic structure indices (NRI and NTI) displayed distinct vertical stratification patterns between sites. Furthermore, the PD in Bawangling’s large shrub layer was positively correlated with total phosphorus in the soil, while community evenness was influenced by soil organic carbon and total nitrogen. In Jianfengling, species richness was significantly associated with soil bulk density, altitude, and pH. These findings enhance our understanding of the ecological and evolutionary processes shaping biodiversity in tropical rainforests and highlight the importance of incorporating both species and phylogenetic metrics into the conservation strategies for endangered species such as Hopea hainanensis. Full article

Background/Objectives: ISUP grade group discordance between prostate biopsy and radical prostatectomy (RP) impacts treatment decisions in over a third (~25–40%) of prostate cancer (PCa) patients. We aimed to identify ISUP grade migration predictors and assess the impact of preoperative imaging (MRI) in a contemporary Romanian PCa cohort. Methods: We retrospectively analyzed 142 PCa patients undergoing RP following biopsy between January 2021 and December 2024 at Pius Brinzeu County Hospital, Timișoara: 90 without and 52 with preoperative MRI. Clinical parameters, MRI findings (PI-RADS), and biopsy characteristics were evaluated. Machine learning models (gradient boosting, random forest) were developed with SHAP analysis for interpretability. Results: Grade migration occurred in 69/142 patients (48.6%): upstaging in 55 (38.7%) and downstaging in 14 (9.9%). In the non-MRI cohort, 37/90 (41.1%) were upstaged and 9/90 (10.0%) were downstaged, versus 18/52 (34.6%) upstaged and 5/52 (9.6%) downstaged in the MRI cohort. The MRI group showed a 6.5% absolute reduction in upstaging (34.6% vs. 41.1%), a promising non-significant trend (p = 0.469) that requires further investigation. Grade 1 patients showed the highest upstaging (69.4%), while Grades 3–4 showed the highest downstaging (11/43, 25.6%). PI-RADS 4 lesions had the highest upstaging (43.5%). PSA density > 0.20 ng/mL² emerged as the strongest predictor. Gradient boosting achieved superior performance (AUC = 0.812) versus logistic regression (AUC = 0.721), representing a 13% improvement in discrimination. SHAP analysis revealed PSA density as the most influential (importance: 0.287). Grade migration associated with adverse pathology: extracapsular extension (52.7% vs. 28.7%, p = 0.008) and positive margins (38.2% vs. 21.8%, p = 0.045). Conclusions: ISUP grade migration affects 48.6% of Romanian patients, with 38.7% upstaged and 9.9% downstaged. The 69.4% upstaging in Grade 1 patients emphasizes the need for enhanced risk stratification tools, while 10% downstaging suggests potential overtreatment. Machine learning with SHAP analysis provides superior predictive performance (13% AUC improvement) while offering clinically interpretable risk assessments. PSA density dominates risk assessment, while PI-RADS 4 lesions warrant closer scrutiny than previously recognized. Full article

Background/Objective: This study explores a novel approach for diagnosing common middle ear pathologies using Pressure-Less Acoustic Immittance (PLAI™), a non-invasive alternative to conventional tympanometry. Methods: A total of 516 ear measurements were collected and stratified into three age groups: 0–3, 3–12, and 12+ years, reflecting key developmental stages. PLAI™-derived acoustic parameters, including resonant frequency, peak admittance, canal volume, and resonance peak frequency boundaries, were analyzed using Random Forest classifiers, with SMOTE addressing class imbalance and SHAP values assessing feature importance. Results: Age-specific models demonstrated superior diagnostic accuracy compared to non-stratified approaches, with macro F1-scores of 0.79, 0.84, and 0.78, respectively. Resonant frequency, ear canal volume, and peak admittance consistently emerged as the most informative features. Notably, age-based stratification significantly reduced false negative rates for conditions such as Otitis Media with Effusion and tympanic membrane retractions, enhancing clinical reliability. These results underscore the relevance of age-aware modeling in pediatric audiology and validate PLAI™ as a promising tool for early, pressure-free middle ear diagnostics. Conclusions: While further validation on larger, balanced cohorts is recommended, this study supports the integration of machine learning and acoustic immittance into more accurate, developmentally informed screening frameworks. Full article

Urban expansion fragments once-contiguous forest patches, generating pronounced edge gradients that modulate soil physicochemical properties and biodiversity. We quantified how fragmentation reshaped the soil microbiome continuum and its implications for soil carbon storage in a temperate urban mixed deciduous forest. A total of 18 plots were considered in this study, with six plots for each fragment type. Intact interior forest (F), internal forest path fragment (IF), and external forest path fragment (EF) soils were sampled at 0–15, 15–30, and 30–45 cm depths and profiled through phospholipid-derived fatty acid (PLFA) chemotyping and amino sugar proxies for living microbiome and microbial-derived necromass assessment, respectively. Carbon fractionation was performed through the chemical oxidation method. Diversity indices (Shannon–Wiener, Pielou evenness, Margalef richness, and Simpson dominance) were calculated based on the determined fatty acids derived from the phospholipid fraction. The microbial biomass ranged from 85.1 to 214.6 nmol g⁻¹ dry soil, with the surface layers of F exhibiting the highest values (p < 0.01). Shannon diversity declined systematically from F > IF > EF. The microbial necromass varied from 11.3 to 23.2 g⋅kg⁻¹. Fragmentation intensified the stratification of carbon pools, with organic carbon decreasing by approximately 14% from F to EF. Our results show that EFs possess a declining microbiome continuum that weakens their carbon sequestration capacity in urban forests. Full article

Chromosomal microarray analysis (CMA) was performed for 40 patients with B-ALL undergoing treatment according to the ALL-2016 protocol to investigate the copy number alterations (CNAs) and copy neutral loss of heterozygosity (cnLOH) associated with minimal residual disease (MRD)-positive remission. Aberrations involving over 20,000 genes were identified, and a random forest approach was applied to isolate a subset of genes whose CNAs and cnLOH are significantly associated with poor therapeutic response. We have assembled the triple matched healthy population data and used that data as a reference, but not as a matched control. We identified a recurrent cluster of cnLOH in the 19q13.2–19q13.31 region, significantly enriched in MRD-positive patients (70% vs. 47% in the reference group vs. 16% in MRD-negative patients). This region includes the pregnancy-specific glycoprotein (PSG) gene family and the oncogene ERF, suggesting a potential role in leukemic persistence and treatment resistance. Additionally, we observed significant deletions involving 7p22.3 and 16q13, often as part of large-scale losses affecting almost the entire chromosomes 7 and 16, indicative of global chromosomal instability. These findings highlight specific genomic regions potentially involved in therapy resistance and may contribute to improved risk stratification in B-ALL. Our findings emphasize the value of high-resolution CMA in diagnostics and risk stratification and suggest that PSG genes and other candidate genes could serve as biomarkers for predicting treatment outcomes. Full article

Cancer metabolic reprogramming plays a critical role in tumor progression and therapeutic resistance, underscoring the need for advanced analytical strategies. Metabolomics, leveraging mass spectrometry and nuclear magnetic resonance (NMR) spectroscopy, offers a comprehensive and functional readout of tumor biochemistry. By enabling both targeted metabolite quantification and untargeted profiling, metabolomics captures the dynamic metabolic alterations associated with cancer. The integration of metabolomics with machine learning (ML) approaches further enhances the interpretation of these complex, high-dimensional datasets, providing powerful insights into cancer biology from biomarker discovery to therapeutic targeting. This review systematically examines the transformative role of ML in cancer metabolomics. We discuss how various ML methodologies—including supervised algorithms (e.g., Support Vector Machine, Random Forest), unsupervised techniques (e.g., Principal Component Analysis, t-SNE), and deep learning frameworks—are advancing cancer research. Specifically, we highlight three major applications of ML–metabolomics integration: (1) cancer subtyping, exemplified by the use of Similarity Network Fusion (SNF) and LASSO regression to classify triple-negative breast cancer into subtypes with distinct survival outcomes; (2) biomarker discovery, where Random Forest and Partial Least Squares Discriminant Analysis (PLS-DA) models have achieved >90% accuracy in detecting breast and colorectal cancers through biofluid metabolomics; and (3) prognostic modeling, demonstrated by the identification of race-specific metabolic signatures in breast cancer and the prediction of clinical outcomes in lung and ovarian cancers. Beyond these areas, we explore applications across prostate, thyroid, and pancreatic cancers, where ML-driven metabolomics is contributing to earlier detection, improved risk stratification, and personalized treatment planning. We also address critical challenges, including issues of data quality (e.g., batch effects, missing values), model interpretability, and barriers to clinical translation. Emerging solutions, such as explainable artificial intelligence (XAI) approaches and standardized multi-omics integration pipelines, are discussed as pathways to overcome these hurdles. By synthesizing recent advances, this review illustrates how ML-enhanced metabolomics bridges the gap between fundamental cancer metabolism research and clinical application, offering new avenues for precision oncology through improved diagnosis, prognosis, and tailored therapeutic strategies. Full article

Background/Objectives: Heart rate variability (HRV) has been widely investigated as a predictor of disease and mortality across diverse patient populations; however, there remains no consensus on the optimal set or combination of time and frequency domain nor on nonlinear features for reliable prediction across clinical contexts. Given the relevance of the COVID-19 pandemic and the unique clinical profiles of these patients, this retrospective observational study explored the potential of HRV analysis for early prediction of in-hospital mortality using ECG signals recorded during the initial moments of ICU admission in COVID-19 patients. Methods: HRV indices were extracted from four ECG leads (I, II, III, and aVF) using sliding windows of 2, 5, and 7 min across observation intervals of 15, 30, and 60 min. The raw data posed significant challenges in terms of structure, synchronization, and signal quality; thus, from an original set of 381 records from 321 patients, after data pre-processing steps, a final dataset of 82 patients was selected for analysis. To manage data complexity and evaluate predictive performance, two feature selection methods, four feature reduction techniques, and five classification models were applied to identify the optimal approach. Results: Among the feature aggregation methods, compiling feature means across patient windows (Method D) yielded the best results, particularly for longer observation intervals (e.g., using LDA, the best AUC of $0.82\pm 0.13$ was obtained with Method D versus $0.63\pm 0.09$ with Method C using 5 min windows). Linear Discriminant Analysis (LDA) was the most consistent classification algorithm, demonstrating robust performance across various time windows and further improvement with dimensionality reduction. Although Gradient Boosting and Random Forest also achieved high AUCs and F1-scores, their performance outcomes varied across time intervals. Conclusions: These findings support the feasibility and clinical relevance of using short-term HRV as a noninvasive, data-driven tool for early risk stratification in critical care, potentially guiding timely therapeutic decisions in high-risk ICU patients and thereby reducing in-hospital mortality. Full article

Background/Objectives: Patients with stage IV gastric cancer who develop chronic intestinal failure require home parenteral nutrition (HPN). This study aimed to evaluate the prognostic relevance of nutritional and immune–inflammatory biomarkers and to construct an individualised survival prediction model using machine learning techniques. Methods: A secondary analysis was performed on a cohort of 410 patients with TNM stage IV gastric adenocarcinoma who initiated HPN between 2015 and 2023. Nutritional and inflammatory indices, including the Controlling Nutritional Status (CONUT) score and lymphocyte-to-monocyte ratio (LMR), were assessed. Independent prognostic factors were identified using Cox proportional hazards models. A Random Survival Forest (RSF) model was constructed to estimate survival probabilities and quantify variable importance. Results: Both the CONUT score and LMR were independently associated with overall survival. In multivariate analysis, higher CONUT scores were linked to increased mortality risk (HR = 1.656, 95% CI: 1.306–2.101, p < 0.001), whereas higher LMR values were protective (HR = 0.632, 95% CI: 0.514–0.777, p < 0.001). The RSF model demonstrated strong predictive accuracy (C-index: 0.985–0.986) and effectively stratified patients by survival risk. The CONUT score exerted the greatest prognostic influence, with the LMR providing additional discriminatory value. A gradual decline in survival probability was observed with an increasing CONUT score and a decreasing LMR. Conclusions: The application of machine learning to immune–nutritional data offers a robust tool for predicting survival in patients with advanced gastric cancer requiring HPN. This approach may enhance risk stratification, support individualised clinical decision-making regarding nutritional interventions, and inform treatment intensity adjustment. Full article

Background: Glioblastoma (GBM) is one of the most challenging malignancies in all of neoplasms. These malignancies are associated with unfavorable clinical outcomes and significantly compromised patient wellbeing. The immunological landscape within the tumor microenvironment (TME) plays a critical role in determining GBM prognosis. By mining data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases and correlating them with immune responses in the TME, genes associated with the immune microenvironment with potential prognostic value were obtained. Method: We selected GSE16011 as the training set. Gene expression profiles were substrates scored by both ESTIMATE and xCell, and immune cell subpopulations in GBM were analyzed by CIBERSORT. Gene expression profiles associated with low immune scores were performed by lasso regression, Cox analysis and random forest (RF) to identify a prognostic model for the multiple genes associated with immune infiltration in GBM. Then we constructed a nomogram to optimize the prognostic model using GSE7696 and TCGA-GBM as validation sets and evaluated these data for gene mutation and gene enrichment analysis. Result: The prognostic correlation between the six genes (MEOX2, PHYHIP, RBBP8, ST18, TCF12, and THRB) and GBM was finally found by lasso regression, Cox regression, and RF, and the online database obtained that all six genes were differentially expressed in GBM. Therefore, a prognostic correlation model was constructed based on the six genes. Kaplan–Meier (KM) survival analysis showed that this prognostic model had excellent prognostic ability. Conclusions: Prognostic models based on tumor microenvironment and immune score stratification and the construction of related genes have potential applications for prognostic analysis of GBM patients. Full article

Background and Objectives: Investigating the adherence to antihypertensive medication and identifying patients with low adherence allows targeted interventions to improve therapeutic outcomes. Artificial intelligence (AI) offers advanced tools for analyzing medication adherence data. This study aimed to develop and validate several predictive models for non-adherence, using patient-reported data collected via a structured questionnaire. Materials and Methods: A cross-sectional, multi-center study was conducted on 3095 hypertensive patients from community pharmacies. A structured questionnaire was administered, collecting data on sociodemographic factors, medical history, self-monitoring behaviors, and informational exposure, alongside medication adherence measured using the Romanian-translated and validated ARMS (Adherence to Refills and Medications Scale). Five machine learning models were developed to predict non-adherence, defined by ARMS quartile-based thresholds. The models included Logistic Regression, Random Forest, and boosting algorithms (CatBoost, LightGBM, and XGBoost). Models were evaluated based on their ability to stratify patients according to adherence risk. Results: A total of 79.13% of respondents had an ARMS Score ≥ 15, indicating a high prevalence of suboptimal adherence. Better adherence was statistically associated (adjusted for age and sex) with more frequent blood pressure self-monitoring, a reduced salt intake, fewer daily supplements, more frequent reading of medication leaflets, and the receipt of specific information from pharmacists. Among the ML models, CatBoost achieved the highest ROC AUC Scores across the non-adherence classifications, although none exceeded 0.75. Conclusions: Several machine learning models were developed and validated to estimate levels of medication non-adherence. While the performance was moderate, the results demonstrate the potential of AI in identifying and stratifying patients by adherence profiles. Notably, to our knowledge, this study represents the first application of permutation and SHapley Additive exPlanations feature importance in combination with probability-based adherence stratification, offering a novel framework for predictive adherence modelling. Full article

Accurate estimation of aboveground biomass (AGB) in tree–shrub communities is critical for quantifying forest ecosystem productivity and carbon sequestration potential. Although generalized allometric equations offer expediency in natural forest AGB estimation, their neglect of interspecific variability introduces methodological pitfalls. Precise AGB prediction necessitates resolving two biological constraints: phylogenetic conservation of allometric coefficients and ontogenetic regulation of scaling relationships. This study establishes an integrated framework combining the following: (1) phylogenetic signal detection (Blomberg’s K/Pagel’s λ) across 157 species’ allometric equations, revealing weak but significant evolutionary constraints (λ = 0.1249, p = 0.0027; K ≈ 0, p = 0.621); (2) hierarchical error decomposition of 9105 stems in a Mt. Wuyishan forest dynamics plot (15 species), identifying family-level error stratification (e.g., Theaceae vs. Myrtaceae, Δerror > 25%); (3) ontogenetic trajectory analysis of Castanopsis eyrei between Mt. Wuyishan and Mt. Huangshan, demonstrating significant biomass deviations in small trees (5–15 cm DBH, p < 0.05). Key findings resolve the following hypotheses: (1) absence of strong phylogenetic signals validates generalized models for phylogenetically diverse communities; (2) ontogenetic regulation dominates error magnitude, particularly in early developmental stages; (3) differential modeling is recommended: species-specific equations for pure forests/seedlings vs. generalized equations for mixed mature forests. This work establishes an error hierarchy: ontogeny > taxonomy > phylogeny, providing a mechanistic basis for optimizing forest carbon stock assessments. Full article

Background/Objectives: Postoperative deep vein thrombosis (DVT) is a common and serious complication after tibial fracture surgery. This study aimed to develop and evaluate machine learning (ML) models to predict the occurrence of DVT following tibia fracture surgery. Methods: A retrospective analysis was conducted on patients who had undergone surgery for isolated tibial fractures. A total of 42 predictive models were developed using combinations of six ML algorithms—logistic regression, support vector machine, random forest, extreme gradient boosting, Light Gradient Boosting Machine (LightGBM), and neural networks—and seven feature selection methods, including SHapley Additive exPlanations (SHAP), Least Absolute Shrinkage and Selection Operator (LASSO), Boruta, recursive feature elimination, univariate filtering, and full-variable inclusion. Model performance was assessed based on discrimination, quantified by the area under the receiver operating characteristic curve (AUC-ROC), and calibration, measured using Brier scores, with internal validation performed via bootstrapping. Results: Of 471 patients, 80 (17.0%) developed postoperative DVT. The ML models achieved high overall accuracy in predicting DVT. Twenty-four models showed similarly excellent discrimination (pairwise AUC comparisons, p > 0.05). The top-performing model (random forest with RFE) attained an AUC of ~0.99, while several others (including LightGBM and SVM-based models) also reached AUC values in the 0.97–0.99 range. Notably, support vector machine models paired with Boruta or LASSO feature selection demonstrated the best calibration (lowest Brier scores), indicating reliable risk estimation. The final selected SVM models achieved high specificity (≥95%) with moderate sensitivity (~75–80%) for DVT detection. Conclusions: ML models demonstrated high accuracy in predicting postoperative DVT following tibial fracture surgery. Support vector machine-based models showed particularly favorable discrimination and calibration. These results suggest the potential utility of ML-based risk stratification to guide individualized prophylaxis, warranting further validation in prospective clinical settings. Full article