SEARCH · Search PubMed

Results for “hierarchical classification”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

118 records · Page 2Linked to original sources

PubMed2026

Extent of resection as an independent predictor of survival for patients with glioblastoma as defined by the new WHO 2021 classification.

OBJECTIVE: Extent of resection (EOR) has previously been demonstrated to have an impact on survival in patients with glioblastoma (GBM). However, with the World Health Organization (WHO) 2021 reclassification of GBMs based on IDH-mutation status, patients with "IDH-mutant GBMs," who typically survive long term, were reclassified as WHO grade 4 IDH-mutant astrocytomas and removed from the GBM taxonomy. Therefore, it is unknown whether the previously reported impact of resection on survival was a false-positive result due to the inclusion of the less aggressive IDH-mutant tumors in previous datasets. This study aimed to determine the extent to which EOR remains an independent predictor of survival in patients with WHO 2021 GBM after the reclassification of IDH-mutant grade 4 astrocytomas. METHODS: All cases of GBM tumors (based on the pre-2021 GBM classification) that were newly diagnosed between 2005 and 2021 were identified in our institutional database and subsequently reclassified based on the updated WHO 2021 criteria using IDH status. Multivariable statistical analyses of demographic information, survival time, and EOR based on volumetric MRI were performed to determine the independent predictors of survival for the whole group of patients and for IDH-wildtype GBM patients exclusively. Additional analyses were performed to identify an EOR threshold for improvement in survival. RESULTS: Of the 523 tumors classified as GBM based on the pre-2021 taxonomy, 52 (9.9%) cases were reclassified as WHO grade 4 IDH-mutant astrocytomas, and the median survival of patients in this group was 7.9 years, whereas median survival of the IDH-wildtype GBM patients was 1.4 years. Multivariate analyses of the whole group demonstrated that IDH-mutant astrocytomas were associated with reduced hazard of death. In both the whole group (n = 523) and in IDH-wildtype GBMs (n = 471), higher EOR of the contrast-enhancing (CE) tumor was associated with reduced hazard of death, whereas older age or male sex was associated with increased hazard of death. Because most patients (90%) had high EOR values (> 81%), a statistically meaningful EOR threshold could not be established. CONCLUSIONS: These analyses demonstrated that EOR of the CE tumor is an independent predictor of survival and that greater EOR is associated with improved survival in WHO 2021 IDH-wildtype GBMs even after excluding grade 4 IDH-mutant astrocytomas. However, an absolute EOR threshold below which resection did not improve survival could not be established, raising concerns about prior cutoff assessments.

Humans↗

PubMed2026

Metabolic engineering of Candida yeasts for biotechnological applications.

Candida yeasts represent a versatile yet underexploited platform for industrial biotechnology. These yeasts utilize a remarkably broad range of carbon sources, particularly for hydrophobic carbon sources, coupled with robust growth and diverse biosynthetic capacities, making them promising hosts for sustainable production of chemicals, fuels, and proteins. Despite these advantages, industrial deployment of Candida species has been hindered by concerns regarding opportunistic pathogenicity and the historical lack of efficient genetic manipulation tools, leading to a substantial gap between metabolic potential and practical utilization. Recent advances in functional genomics, genome editing, and systems metabolic engineering are rapidly overcoming these barriers, enabling more precise and efficient strain development. In this review, we systematically summarize recent progress in the metabolic engineering of Candida species as microbial cell factories, with particular emphasis on expanding genetic toolkits, utilizting renewable and non-conventional carbon sources, and biosynthesizing high-value compounds. In addition, we propose a biosafety-oriented classification framework to support their safe industrial deployment. Finally, we discuss current challenges and emerging opportunities, emphasizing that the synergy of synthetic biology and artificial intelligence-driven design holds the key to unlocking the biotechnological potential of Candida yeasts.

Candida↗

PubMed2026

The impact of body mass index classification on operative characteristics and perioperative outcomes in lumbar microdiscectomy.

INTRODUCTION: Body mass index (BMI) stratification helps classify obesity severity. In patients undergoing microdiscectomy for symptomatic lumbar disc herniation, the effect of obesity on perioperative risk remains incompletely understood. This retrospective single-institution study evaluated whether BMI class influences perioperative risk in a large surgical cohort. METHODS: Adults older than 18 years who underwent primary, elective single-level lumbar microdiscectomy between June 2018 and March 2025 with at least 3 months of follow-up were included. Patients were grouped by BMI: without obesity (WO, BMI < 30), class I (CI, 30-34.9), class II (CII, 35-39.9), and class III (CIII, ≥40). Outcomes were analyzed separately for open microdiscectomy (OM), tubular microdiscectomy (TM), and endoscopic discectomy (ED). Continuous variables were compared using Kruskal-Wallis testing with Dunn post hoc analysis; categorical variables were compared with chi-square tests. Significance was set at p < 0.05. RESULTS: A total of 757 patients were included (OM 422, TM 190, ED 145). Higher obesity classes underwent ED more frequently (p = 0.038). In the OM cohort (WO 258, CI 97, CII 50, CIII 17), CI had a higher proportion of males and CII a lower proportion (p = 0.007). Operative time, length of stay, and estimated blood loss were greatest in CII and CIII patients (all p < 0.001). CII patients also had more emergency department visits within 1 year than other classes (p = 0.026). No differences were found in age, smoking status, disc herniation type, dural tears, intraoperative or postoperative complications, or revision presence/time. In the TM cohort (WO 117, CI 47, CII 21, CIII 5), WO patients were oldest and CIII youngest (p < 0.001), with no other significant differences. In the ED cohort (WO 79, CI 31, CII 20, CIII 15), WO patients were oldest and CIII youngest (p = 0.004). CIII patients had higher estimated blood loss (p = 0.028) and shorter time to revision (p < 0.001), while other variables were similar. CONCLUSIONS: ED was used more often in higher obesity classes. In OM, CII and CIII obesity were associated with longer operative time, longer hospital stay, and greater blood loss, likely due to increased exposure requirements. TM and ED showed few obesity-related differences in complications, suggesting minimally invasive approaches may mitigate obesity-related perioperative risk. However, the retrospective design and small number of CIII patients warrant further study.

Humans↗

PubMed2026

Methylation profiling in CNS tumor diagnostics: a single-centre real-world experience from Central Europe.

Genome-wide DNA methylation profiling has transformed neuro-oncology by providing an objective, machine learning-based taxonomy that mitigates interobserver variability and refines the histo-molecular criteria of the current WHO classification. We evaluate the real-world diagnostic performance and clinical utility of this modality in a prospective, consecutively accrued three-year cohort of 291 central nervous system (CNS) tumors across a mixed adult-pediatric population. Successful profiling was completed in 95.9% of cases. Using the Epignostix classifier, a high-confidence diagnostic match (calibrated score [CS] ≥ 0.84) was achieved in 70.3% of analyzable samples, while 26.5% returned lower-confidence scores (≥ 0.3 to < 0.84) and only 3.2% remained completely unclassifiable (CS < 0.3). When integrated into a comprehensive diagnostic framework, methylation profiling provided clinically useful results in 81.1% of cases, establishing diagnoses in 70 cases submitted for molecular subclassification and resolving diagnostic uncertainty or prompting major revisions in 149 histologically challenging tumors. Within truly ambiguous lesions, integration of methylome data dictated tumor grade modifications in 38.8% of cases (upgrading in 29.4% and downgrading in 9.4%), shifting patient risk stratification. Crucially, over half (52.7%) of the lower-confidence cases yielded meaningful clinical integration when supported by histomorphology and ancillary genetic or immunohistochemical markers, demonstrating that rigid score cutoffs should not dictate assay failure. Discrepant or misleading classifications occurred in 1.9%. Updating bioinformatic pipelines from version 11b4 to 12.8 rescued multiple ambiguous entries, increasing overall clinical utility to 84.1%. These findings demonstrate that integrating computational epigenomics with classical neuropathology enhances diagnostic precision, while highlighting the ongoing need for careful clinical-pathological correlation.

Central nervous system tumors↗

PubMed2026

Functional neuroimaging subtypes of obsessive-compulsive disorder: A systematic review and meta-analysis.

Obsessive-compulsive disorder (OCD) exhibits substantial clinical heterogeneity that may reflect underlying neurobiological diversity. Neuroimaging-based subtyping may advance precision psychiatry by identifying biologically distinct subgroups with differential treatment responses. This study systematically synthesized evidence from functional neuroimaging subtyping studies in OCD to identify reproducible neurobiological subtypes, characterize their clinical profiles, and establish a consensus-based classification framework. We reviewed 40 original studies employing machine learning, clustering, normative modeling, or classification approaches, encompassing approximately 8,150 patients. Consensus clustering identified three reproducible neurobiological subtypes. The Limbic-Hyperactive subtype, comprising approximately 40% of patients, exhibited amygdala and insula hyperconnectivity, elevated anxiety levels, predominant contamination and washing symptoms, and favorable response to cognitive-behavioral therapy. The Fronto-Striatal-Hypoconnected subtype, comprising approximately 35% of patients, demonstrated reduced orbitofrontal-striatal connectivity, cognitive inflexibility, predominant checking and ordering symptoms, and a favorable response to selective serotonin reuptake inhibitors. The Global-Disrupted subtype, comprising approximately 25% of patients, exhibited widespread connectivity disruption, greater symptom severity, and poor treatment response. Support vector machine classification achieved 81.5% accuracy for subtype assignment, though classification of OCD versus healthy controls showed limited generalizability in multisite settings (AUC 0.567-0.673). These findings support a neuroimaging-based framework for personalized treatment selection but require prospective validation.

Humans↗

PubMed2026

The incidence and descriptive factors of calcaneal malunion after surgical fixation of intra-articular calcaneal fractures using the sinus tarsi approach: A retrospective cohort study with binary logistic regression analysis.

BACKGROUND: This study evaluated the incidence of calcaneal malunion after minimally invasive sinus tarsi approach (MIS-STA) in displaced intra-articular calcaneal fractures (I-ACFs) and identified related descriptive factors of calcaneal malunion. METHODS: A retrospective review of 99 displaced I-ACFs treated with MIS-STA was conducted. Demographic data, pre-operative radiographs, and operative details were analyzed. Outcomes included numerical rating scale (NRS) pain scores at rest and during activities of daily living (ADL), Foot and Ankle Ability Measure (FAAM) for ADL and radiographic parameters. Logistic regression was used to identify descriptive factors associated with malunion. RESULTS: Malunion occurred in 33/99 cases (33.3%). The significant descriptive factors were the initial Böhler angle < 0.5 °, time to surgery > 12.5 days, and Sanders type ≥ III. Malunion patients had significantly worse NRS and FAAM scores (p ≤ 0.001). CONCLUSION: Calcaneal malunion after MIS-STA occurred in one-third of cases, with three descriptive factors identified and poorer outcomes observed. LEVEL OF EVIDENCE: III, Comparative retrospective study with binary logistic regression analysis.

Humans↗

PubMed2026

King's Mill anterior-posterior and transverse versus paracervical block in operative outpatient hysteroscopy: a randomised trial.

BACKGROUND: Pain control during operative outpatient hysteroscopy remains a clinical challenge, with the paracervical block (PCB) showing inconsistent efficacy. The King's Mill Anterior-Posterior and Transverse (KAPT) block targets the Lee-Frankenhauser plexus within the uterosacral ligament complex, where the highest density of uterine and cervical sensory fibres lies. We compared the KAPT block with the PCB in operative outpatient hysteroscopy. METHODS: Single-centre, participant- and outcome assessor-blinded randomised controlled trial in a UK district general hospital. Forty-eight women undergoing operative outpatient hysteroscopy were randomised 1:1 to the KAPT block or the PCB, each delivered with 10 ml of prilocaine. Pain was measured on a 10-point Visual Analogue Scale (VAS) during the procedure and 10 minutes post-procedure (co-primary endpoints), and during cervical dilatation (exploratory). Analyses used the Mann-Whitney U test on an intention-to-treat basis, with Hodges-Lehmann median differences, percentile bootstrap 95% confidence intervals, and a hierarchical fixed-sequence testing strategy. RESULTS: Median intra-procedural VAS was 1 (IQR 0 to 3) with the KAPT block versus 2 (IQR 1 to 6) with the PCB (Hodges-Lehmann median difference -1.00, 95% CI -3.00 to 0.00; p = 0.040). At 10 minutes post-procedure, median VAS was 0 (IQR 0 to 1.25) versus 2 (IQR 0 to 3.25) (median difference -1.00, 95% CI -2.00 to 0.00; p = 0.011). Both co-primary endpoints reached significance under hierarchical testing. Cervical dilatation pain did not differ significantly (p = 0.146). Satisfaction was 100% in both arms; willingness to recommend was 100% (KAPT) versus 95.8% (PCB). No serious adverse events occurred. CONCLUSIONS: The KAPT block was associated with lower intra-procedural and post-procedural pain than the PCB, with effect sizes consistent with a clinically meaningful difference. These findings support the KAPT block as a superior, anatomically targeted alternative to the PCB and justify evaluation in a larger multicentre trial.Trial registration: ISRCTN15619382.

Humans↗

PubMed2026

Comparative effectiveness of torsemide vs furosemide in the management of heart failure patients: Win-ratio reanalysis of the TRANSFORM-HF trial.

BACKGROUND: Loop diuretics are widely used for managing congestion in patients with heart failure (HF). The TRANSFORM-HF trial is a multicenter randomized study that enrolled heart failure patients, comparing a strategy of torsemide vs furosemide. The time-to-event analysis demonstrated neutral effects on all-cause death at 30 months and the composite of all-cause death and first rehospitalization at 12 months. We evaluated whether a hierarchical win-ratio (WR) framework integrating mortality, recurrent hospitalization, and patient-reported health status provides additional interpretive insight. METHODS: This study is a secondary analysis of the pragmatic, multicenter, open-label, randomized TRANSFORM-HF trial, conducted across 60 US hospitals that randomized 2,859 patients hospitalized with HF to torsemide or furosemide. The primary 12-month hierarchical composite outcome was defined as (1) all-cause mortality, (2) recurrent all-cause hospitalizations, and (3) lack of improvement in the Kansas City Cardiomyopathy Questionnaire Clinical Summary Score (KCCQ-CSS). The primary statistical method was a WR analysis adjusting covariates via inverse probability weighting. Subgroup analyses evaluated potential heterogeneity across patient demographics and clinical characteristics. RESULTS: In the primary 12-month intention-to-treat analysis, the adjusted WR was 1.07 (95% CI, 0.98-1.16; P = .13), indicating no significant difference between torsemide and furosemide. A supplementary 30-month analysis with extended mortality follow-up yielded a similar estimate (adjusted WR, 1.06; 95% CI, 0.98-1.16; P = .14); hospitalization and KCCQ-CSS components were assessed through 12 months. As-treated sensitivity analyses were consistent with the neutral primary findings. Exploratory subgroup analyses were not adjusted for multiplicity and should be considered hypothesis-generating. CONCLUSIONS: The overall WR comparison between torsemide and furosemide showed no statistically significant difference in the primary 12-month analysis. The WR framework provided an interpretive decomposition across outcome domains but did not establish superiority of either loop diuretic strategy. All findings should be considered exploratory. TRIAL REGISTRATION: ClinicalTrials.gov, NCT03296813, https://clinicaltrials.gov/study/NCT03296813.

Aged↗

PubMed2026

Meta-PseU: A meta-classifier for robust prediction of RNA pseudouridine modification sites from long sequences.

BACKGROUND AND OBJECTIVES: Pseudouridine (Ψ) represents one of the most abundant and conserved RNA modifications. Ψ provides an additional hydrogen-bond donor that enhances RNA structural stability and modulates translation. It participates in diverse biological processes, including RNA-protein interactions, splicing, translational control, and stress responses. Aberrant pseudouridylation is implicated in cancer, neurodegenerative disorders, and autoimmune diseases. Despite its biological importance, experimental identification of Ψ sites remains time-consuming and costly, limiting the feasibility of transcriptome-wide profiling. Computational approaches have therefore become essential complements to experimental techniques. However, state-of-the-art machine-learning and deep-learning predictors often suffer from limited generalizability due to small training datasets. To overcome these issues, we aim at constructing new long-sequence datasets and developing a novel Ψ site predictor. METHODS: New long-sequence datasets were constructed as benchmarks for RNA Ψ-site prediction. The Ψ modification sites in RMBase 3.0 were mapped to the reference genomes across three species of human, mouse, and yeast, and the RNA sequences with a length of 201 were generated by extending the upstream and downstream from the mapped, central sites. To eliminate sequence redundancy, the sequences were clustered using CD-HIT with a 70% sequence identity threshold. We developed Meta-PseU, a logistic regression-based meta-classifier that considered 118 machine learning and deep learning classifiers. The datasets and programs are freely accessible at https://github.com/kuratahiroyuki/MetaPseU. RESULTS: By optimizing model configuration, we proposed the Meta-PseU model stacking 32 machine learning and deep learning classifiers out of 118 classifiers. Meta-PseU substantially improved model generalizability, overcoming a key limitation of existing approaches. It greatly outperformed state-of-the-art predictors and achieved increasing accuracy with increasing sequence length. CONCLUSIONS: Long-sequence datasets were newly constructed as benchmarks for RNA Ψ-site prediction. Meta-PseU offers a new framework for robust Ψ-site identification by using long sequences.

Pseudouridine↗

PubMed2026

Integrated salivary proteomic and metabolomic analyses reveal molecular characterization and novel biomarker panels of chronic obstructive pulmonary disease.

Chronic obstructive pulmonary disease (COPD) is a respiratory disorder characterized by chronic inflammation, oxidative stress, and metabolic dysregulation. The lack of convenient and easily-accessible non-invasive diagnostic approaches remains a major clinical challenge. This study applied an integrated saliva-based proteomic and untargeted metabolomic strategy to identify potential biomarkers for COPD classification. Comprehensive multi-omics analyses identified 225 differentially abundant proteins and 60 differentially abundant metabolites between patients with COPD and healthy controls, including 24 biologically relevant endogenous metabolites. Functional enrichment analyses revealed pronounced dysregulation of mitochondrial energy metabolism, redox homeostasis, lipid remodeling, and inflammatory-related pathways in COPD. By integrating salivary proteomic and metabolomic biomarkers, a stepwise feature selection combined with LASSO logistic regression was used to construct diagnostic models, yielding an optimized biomarker panel consisting of 11 proteins and 2 endogenous metabolites. This integrated model achieved excellent diagnostic performance, with an area under the ROC curve of 0.96. Collectively, these findings demonstrate that integrated salivary proteomic and metabolomic profiling provides a robust, non-invasive approach for COPD classification and offers a promising foundation for the development of biosensor-based diagnostic platforms and early disease detection. SIGNIFICANCE: Chronic obstructive pulmonary disease (COPD) remains a major global health burden. Current diagnostic approaches rely largely on spirometry and clinical assessment, which are limited in sensitivity for early-stage disease and unsuitable for large-scale screening. This study employs an integrated saliva-based proteomic and metabolomic strategy to identify non-invasive biomarkers for COPD classification. Our findings reveal coordinated dysregulation of mitochondrial energy metabolism, redox homeostasis, and lipid remodeling in COPD, highlighting the interconnected roles of metabolic reprogramming, oxidative stress, and inflammation in disease pathophysiology. Notably, a robust diagnostic panel comprising 11 proteins and 2 endogenous metabolites was established, achieving excellent classification performance (AUC of 0.96). To our knowledge, the integrated application of salivary proteomics and metabolomics for COPD diagnosis remains largely unexplored, underscoring the significance and translational potential of our findings.

Humans↗

PubMed2026

Characteristics of p53 and Smad4 immunohistochemistry in pancreatic ductal adenocarcinoma and validation by next-generation sequencing.

BACKGROUND: Mutations in four major driver genes -KRAS, CDKN2A, TP53, and SMAD4- are central to the pathogenesis of pancreatic ductal adenocarcinoma (PDAC) and critically inform diagnosis, therapeutic decision-making, and prognostic assessment. Although next-generation sequencing (NGS) is widely regarded as the gold standard for detecting these mutations, its clinical application is often limited by suboptimal analytical efficiency and substantial economic cost. Among these genes, immunohistochemical (IHC) staining for the proteins encoded by TP53 and SMAD4 has been extensively adopted in routine pathology practice. However, standardized IHC pattern classification schemes and rigorous validation of their predictive accuracy for underlying genomic alterations remain lacking in PDAC. METHODS: We retrospectively enrolled 63 PDAC patients and systematically characterized the typical IHC expression patterns of p53 and Smad4. Targeted NGS was subsequently performed on all available tumor specimens, and the resulting mutational profiles were correlated with corresponding IHC findings. Diagnostic performance including sensitivity, specificity and accuracy of p53 IHC for predicting TP53 mutations and of Smad4 IHC for predicting SMAD4 mutations was rigorously evaluated. RESULTS: Among the four canonical driver genes, co-occurring double- or triple-gene mutations were prevalent; within TP53 and SMAD4, missense mutations constituted the most frequent variant type. Using NGS as the reference standard, we validated the diagnostic utility of a three-tiered p53 IHC classification system, particularly in fine-needle biopsy (FNB) specimens. Furthermore, we proposed a novel, refined Smad4 IHC pattern classification that incorporates an "intermediate" category, thereby expanding upon conventional binary interpretation. This new scheme achieved markedly improved mutation prediction accuracy (0.76) compared with traditional approaches (0.57). CONCLUSION: Our study highlights the complementary diagnostic value of p53 and Smad4 IHC relative to molecular testing in PDAC, especially when tissue is limited, as commonly encountered in FNB specimens. The newly established Smad4 IHC classification system, which integrates an intermediate expression category into the conventional two-tier framework, demonstrates superior clinical utility and enhances predictive accuracy for SMAD4 genomic alterations.

Humans↗

PubMed2026

Transverse testicular ectopia with fused vas deferens: A systematic review.

BACKGROUND: Transverse testicular ectopia (TTE) with fused vas deferens is an extremely rare anomaly, often diagnosed intraoperatively. Current TTE classifications do not address internal ductal variations, limiting surgical guidance. OBJECTIVE: To systematically review cases of TTE with fused vas deferens, summarize presentation, operative strategies, outcomes and identify patterns that highlight the need for classification refinement. METHODS: A PRISMA 2020-compliant systematic review (PROSPERO; CRD420251247785) was performed across PubMed, ScienceDirect and citation of included articles through December 2025. Case reports and series confirming fused vas deferens were included. Data extracted comprised demographics, presentation, imaging, surgical approach, and outcomes. Quality assessment used JBI checklists. RESULTS: 12 studies (16 patients) were included. Most presented with unilateral inguinal hernia (62%) and contralateral undescended testis (68%); 81% were diagnosed intraoperatively. Anatomical patterns included common/proximal fused vas (87%), Y-shaped fusion (6%), and long-loop vas (6%). Trans-septal orchidopexy was the preferred approach, with preservation of vas integrity. Postoperative outcomes were favorable; long-term follow-up was limited. CONCLUSION: TTE with fused vas deferens represents a distinct variant requiring careful intraoperative recognition. We propose a Type IV TTE category for internal ductal fusion to guide surgical planning and classification refinement. Further accumulation of case-based evidence may help clarify its anatomical patterns and operative implications.

Humans↗

PubMed2026

Machine learning vs. traditional methods for predicting postoperative cardiac complications after non-cardiac surgery: a systematic review and Bayesian network meta-analysis.

INTRODUCTION: Accurate prediction of peri-operative cardiac complications is critical to optimise pre-operative decision-making. Traditional risk prediction scores, such as the Revised Cardiac Risk Index, show only modest discrimination. Machine learning can model complex, non-linear relationships but their predictive performance compared with traditional scores remains unclear. METHODS: We performed a systematic review and Bayesian network meta-analysis. The primary outcome was postoperative adverse cardiac events following non-cardiac surgery. Prediction models were assessed relative to the Revised Cardiac Risk Index. As many studies evaluated multiple versions of each model type, the highest performing ('best version') and lowest performing ('worst version') results were analysed. Models were ranked using the surface under the cumulative ranking curve (SUCRA). RESULTS: Thirteen studies evaluating 54 models and 927,113 patients were included. Machine learning approaches generally outperformed traditional risk scores. Automated machine learning ranked highest (SUCRA 96.6) showed the greatest improvement in the best version analysis (mean difference (MD) 0.28 (95%CrI 0.16-0.40)) and remained superior in the sensitivity analysis (MD 0.30 (95%CrI 0.14-0.45)). Gradient boosting models showed superior performance over the Revised Cardiac Risk Index across analysis (best version: MD 0.20 (95%CrI 0.14-0.26), worst version: MD 0.18 (95%CrI 0.12-0.25), SUCRA 82.4). The Gupta Perioperative Risk for Myocardial Infarction or Cardiac Arrest score outperformed the Revised Cardiac Risk Index in the best version analysis (MD 0.16 (95%CrI 0.01-0.32)). Between-study heterogeneity was low. None of the included studies externally validated their machine learning models and only six were judged to be at low risk of bias. DISCUSSION: Most machine learning models showed better discrimination than traditional risk scores, with automated machine learning and gradient boosting models ranking highest. However, study quality, calibration reporting and absence of external validation limit immediate clinical adoption. Prospective, multicentre evaluation is required before integration of these models into peri-operative practice.

Humans↗

PubMed2026

Predicting ACL injury risk in athletes: A systematic review of machine learning-based models.

BACKGROUND: Early ACL injury risk identification in athletes is essential. This systematic review examines machine learning (ML) models for predicting ACL injuries, evaluating their methodological quality, performance, and reliability. METHOD: A comprehensive electronic search was conducted across PubMed, Scopus, Web of Science, and IEEE Xplore databases, supplemented by Google Scholar for grey literature, covering articles published between January 1, 2015, and August 30, 2025. Eligible studies were appraised using the Prediction Model Study Risk of Bias Assessment Tool (PROBAST) for methodological quality and risk of bias, and the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) guidelines for quality of evidence. RESULTS: Ten studies were included. PROBAST showed eight studies had moderate risk of bias and two low risk. TRIPOD found only two studies met quality criteria. ML models included logistic regression (n = 5), support vector machines (n = 4), k-nearest neighbor (n = 3), decision trees (n = 3), random forests (n = 5), neural networks (n = 2), linear discriminant analysis (n = 1), and pre-trained CNNs (n = 1). AUC ranged from 0.63 to 0.98. Accuracy (reported in six studies) ranged from 26% to 95%; however, these values should be interpreted with caution due to the absence of confidence intervals, lack of class imbalance handling, and limited external validation across studies. Tree-based ensemble methods such as random forest achieved competitive accuracy (74-86%), while SVM, a non-ensemble classifier, reported accuracy ranging from 71% to 95%; however, the highest values were obtained in studies with notably small sample sizes (n = 12 to n = 39), raising concerns about overfitting and generalizability. CONCLUSION: Current ML algorithms show promise for identifying athletes at high ACL injury risk and detecting relevant risk factors. Although study quality was generally satisfactory, future research should prioritize external validation and model interpretability to support clinical translation.

Humans↗

PubMed2026

Reinforcement learning-based dynamic ensemble for missense variant effect prediction and tiered prioritization of VUS.

BACKGROUND: Accurate classification of missense variants remains a challenging task despite major advances in genomics. Numerous computational models have been developed to assist in variant classification, but often require repeated integration and benchmarking efforts. Ensemble methods have been proposed to overcome the limitations of single predictors, but mostly rely on fixed, predefined weights that constrain their ability to capture interactions among predictive signals. METHODS: We present GenixRL, a dynamic ensemble framework that reformulates model fusion as a reinforcement learning optimization problem. GenixRL uses a Q-learning agent to learn a policy that dynamically weights the probabilistic outputs of complementary predictors, including BayesDel (addAF and noAF), ClinPred, and MetaRNN. Replacing static weighting with policy learning allows GenixRL to adaptively identify optimal weightings and substantially improve classification accuracy. RESULTS: In benchmark evaluation against 25 state-of-the-art predictors, GenixRL achieved an AUROC of 0.9644 on an independent ClinVar dataset. On saturation genome editing assays for BRCA1 and BRCA2, GenixRL achieved the best performance and ranked highest on 14 of 17 clinically significant genes in a zero-shot evaluation. Applied to uncertain and conflicting ClinVar variants, GenixRL enabled tiered, evidence-based prioritization of hundreds of thousands of variants as likely pathogenic or pathogenic with high confidence, supported by orthogonal population evidence from gnomAD. CONCLUSION: GenixRL advances pathogenicity prediction for missense variants and provides an adaptive ensemble that sorts variants of uncertain significance into tiered candidates for expert curation and functional validation.

Mutation, Missense↗

PubMed2025

Impact of stromal maturity and proportion on prognosis and immune landscape in colorectal cancer.

BACKGROUND: Tumour microenvironment and cancer cells have constant interaction affecting cancer progression. Tumour-stroma ratio (TSR) in the tumour centre and desmoplastic reaction (DR) classification at the invasive margin are prognostic factors based on stroma evaluation on H&E slides. However, their combined value and immunological associations remain poorly defined. This study examines the prognostic and immunological value of TSR, DR, and their combination in two large colorectal cancer cohorts. METHODS: Two colorectal cancer cohorts (N = 1,876) were analyzed. We introduced a three-tiered Stromal Maturity and Proportion Score (SMAPS) based on the presence of high (>50%) TSR and myxoid stroma (immature DR classification). Alcian blue staining was used to further quantify myxoid stroma. Multiplex immunohistochemistry combined with digital image analyses, was utilized to study immune cell densities associated with SMAPS, TSR, DR, and Alcian blue intensity. RESULTS: In the study cohort (N = 1,100), SMAPS was a stronger predictor of cancer-specific mortality [HR for high (vs. low) SMAPS 2.01 (95% CI 1.47-2.75), p < 0.0001] compared to TSR [HR for stroma-high (vs. stroma-low) 1.49 (95% CI 1.15-1.93), p = 0.003] and DR classification [HR for immature (vs. mature) 1.84 (95% CI 1.39-2.45), p < 0.0001]. High SMAPS, stroma-high TSR, and immature DR correlated with lower densities of CD3+ T cells, B cells, M1-like macrophages, CD66B+ granulocytes, and mast cells. Alcian blue staining was associated with immature DR and corresponding immune cells. The validation cohort (N = 776) confirmed the association of SMAPS with survival and T cell densities. CONCLUSIONS: TSR and DR are independent prognostic factors for cancer-specific survival. SMAPS is a promising prognostic tool that integrates stromal maturity at the invasive margin and stromal proportion in the tumour centre. SMAPS has stronger prognostic value compared to TSR and DR classifications alone. A high stromal proportion and myxoid content are associated with an immunosuppressive microenvironment characterized by lower densities of antitumourigenic immune cells.

Humans↗

PubMed2026

Pilot Distractions and Interruptions in Airlines: Ranking of Sources by Analytic Hierarchy Process.

ObjectiveThis work establishes a methodological framework for sources of pilot distraction and interruptions in a structured model that can be used as a tool for cockpit design/procedure assessment.BackgroundPilots must complete complex tasks, and distractions can impair performance and lead to errors that can cause aircraft accidents. Although various cockpit distractors are examined individually, there is no integrated approach.MethodDistraction and interruption sources were identified through a literature review and confirmed/extended by interviews with airline pilots. Associated weights were determined through pairwise comparisons, yielding a hierarchical model using the Analytic Hierarchy Process.Results26 sources of pilot distraction and interruptions were quantified and categorized into four categories: communication, head-down time, responding to abnormal conditions & unexpected situations, and searching for traffic.ConclusionA taxonomic structure for assessment is achieved with the top 5 sources identified as communications, technical interruptions, experience in type, environmental factors, operational irregularities, and airspace high terrain, accounting for 63.07%.ApplicationThe structured system is a flexible assessment scale that provides a taxonomic framework for airline risk management, supports future research, and cockpit design efforts.

Humans↗

PubMed2026

A systematic approach to standardizing the visual appearance of endometriotic lesions for artificial intelligence recognition.

INTRODUCTION: Numerous studies have shown that the diagnostic performance and reproducibility of visual recognition of endometriosis during laparoscopy are poor. The use of artificial intelligence (AI) seems relevant for exhaustive lesion recognition. Standardization of the visual classification of lesions, in the form of an ontology, is an essential prerequisite to enable medical experts to annotate surgical data consistently and subsequently allow engineers to train and build an artificial intelligence tool for endometriosis recognition. MATERIAL AND METHODS: A systematic search was conducted in the MEDLINE (via PubMed), EMBASE, and the Cochrane Library databases up to May 2022, aiming to identify studies describing the laparoscopic visual appearance of superficial endometriosis, endometriomas, and deep infiltrating endometriosis. The accumulated data in the literature concerning the visual appearance of the different forms of endometriosis were used to create an ontology that could be used for artificial intelligence applications. RESULTS: Out of 932 articles screened, 35 studies were selected based on the inclusion criteria of human subjects with histologically confirmed endometriosis lesions visualized via laparoscopy. The selected studies were reviewed to develop a visual ontology of endometriosis lesions observed via laparoscopy. The lesions were categorized into 4 classes and further subdivided into 11 subclasses: superficial (black, red, white, or subtle), adhesions (dense or filmy), deep (obliteration, retraction, or deformation), and ovarian (endometrioma or chocolate fluid). The positive predictive value (PPV) varied across lesion types: black lesions (PPV 47%-97%), red lesions (PPV 33%-100%), white lesions (PPV 20%-81%), and ovarian endometriosis (PPV 42%-98%). Nonspecific lesions such as adhesions (PPV 16%-50%) and subtle superficial lesions (PPV 0%-67%) presented lower PPVs. Deep endometriosis lesions, often buried within organs, required indirect signs (obliteration, retraction, deformation) for identification. CONCLUSIONS: The visual ontology proposed in this systematic search could facilitate the detection and classification of endometriosis lesions using artificial intelligence. This study highlights the challenges of reaching a consensus on lesion recognition and classification in AI projects due to the diverse visual presentations of endometriosis.

Humans↗