SEARCH · Search PubMed

Results for “biomarker prediction”

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.

517 records · Page 9Linked to original sources

PubMed2026

Artificial intelligence in treatment prediction for skeletal Class III malocclusion: A systematic review.

In skeletal Class III patients, treatment options range from orthodontics to orthognathic surgery. Choosing the optimal approach requires a comprehensive clinical evaluation, which may be supported by AI tools. The aim of this study was to assess the performance of AI models in predicting the need for orthognathic surgery and in identifying predictors influencing treatment decisions. A PRISMA-guided electronic database search (PubMed, Web of Science; 2009-2024; English/French) was performed to identify studies using machine learning (ML) or deep learning (DL) on cephalometric and clinical data. After screening and assessment for eligibility, 15 studies were critically appraised. Model performance was summarized using accuracy, sensitivity, specificity, and the area under the curve (AUC). ML algorithms (particularly Random Forest and XGBoost) and DL models (ResNet-based convolutional neural networks (CNNs)) achieved high accuracy for predicting surgical need. Frequently selected predictors included Wits appraisal, ANB angle, the maxillomandibular ratio (Mx/Md), overjet, and the divergence of the lower gonial angle. AI methods show promise for assisting treatment decisions in Class III malocclusion, with Random Forest and XGBoost performing well on tabular cephalometric data and CNNs on imaging. Larger, multicentre datasets and external validation are needed to improve reliability, address bias, and support clinical implementation.

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↗

PubMed2025

Impact of subthreshold troponin levels and temporal trends on short term adverse cardiovascular outcomes in patients discharged from the emergency department: a RACE-IT trial substudy.

BACKGROUND: High-sensitivity cardiac troponin I assays enable early exclusion of myocardial infarction in the emergency department. However, the clinical implications of detectable troponin values below the 99th percentile upper reference limit (4-18 ng/L) remain unclear. OBJECTIVE: To assess the association between subthreshold troponin levels and 30-day outcomes in patients from the RACE-IT trial, using exact troponin values when available. METHODS: This post-hoc analysis of the RACE-IT stepped-wedge randomized controlled trial included patients with troponin ≤ 18 ng/L across nine EDs. Patients were stratified by initial troponin, peak value, absolute change, and percent change. The primary outcome was a 30-day composite of all-cause death, acute MI, percutaneous coronary intervention, and coronary artery bypass grafting. Logistic regression analysis after adjusting for age, sex, race, and coronary artery disease was performed. RESULTS: Among 19,194 patients with troponin ≤ 18 ng/L, 117 (0.6%) experienced the composite outcome. Higher troponin levels were associated with increased event rates in unadjusted analyses. Adjusted analyses showed no independent associations overall, though patients whose highest troponin values fell within the ≥ 11- ≤ 18 ng/L range continued to demonstrate significantly worse outcomes than those with lower peak levels. Elevated troponin values correlated with older age, male sex, and greater comorbidity burden. CONCLUSION: In this post-hoc analysis of patients with troponin values below the 99th percentile URL, absolute levels and temporal changes were not independently associated with 30-day adverse outcomes. These findings support the use of subthreshold troponin values in rapid rule-out protocols, emphasizing the need to consider clinical context and comorbidities in risk assessment.

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↗

PubMed2026

Acute and chronic effects of mixed nuts on energy metabolism in women at cardiometabolic risk: a randomized clinical trial.

BACKGROUND AND AIMS: The effect of consuming a mix of Brazilian nuts on energy metabolism has not been explored. Thus, the present study aimed to evaluate the effects of acute and chronic consumption of mixed nuts on markers of energy metabolism in women with overweight/obesity. METHODS AND RESULTS: This is a randomized, controlled, and parallel clinical trial with adult women. In an acute study, participants received a beverage containing mixed nuts (30 g of cashew nuts + 15 g of Brazil nuts) or a control beverage, and energy metabolism markers were assessed for up to 3 h postprandially. For the chronic study, participants received 45 g of a mix of nuts/day and a -500kcal energy-restricted diet (MNG) or only a -500kcal energy-restricted diet free of nuts (CTG) for 8 weeks, and energy metabolism was assessed before and after the intervention period. In the postprandial period, fat oxidation was higher in the MNG than in the CTG (piAUC: 47.53 ± 5.78 mg/min vs. 27.93 ± 6.98 mg/min; p = 0.048). After 8 weeks of the intervention, fasting fat oxidation increased in the MNG (+16.0 ± 7.0 mg/min) and decreased in the CTG (-5.0 ± 6.0 mg/min), with no significant difference between groups. Other acute and chronic markers also showed no significant changes between groups. CONCLUSION: The acute consumption of mixed nuts increased postprandial fat oxidation, whereas chronic intake within an energy-restricted diet did not affect energy metabolism markers in women at cardiometabolic risk. REGISTRATION NUMBER FOR BRAZILIAN REGISTRY OF CLINICAL TRIALS: RBR-3ntxrm.

Humans↗

PubMed2026

The effect of zalunfiban on high sensitivity cardiac troponin and the association with clinical outcomes in patients with STEMI.

BACKGROUND: Among individuals with ST-segment elevation myocardial infarction (STEMI), a single subcutaneous injection of the short-acting glycoprotein IIb/IIIa receptor blocker antagonist zalunfiban at first medical contact significantly improved the primary outcome including clinical endpoints. The impact of zalunfiban on Myocardial Infarction (MI) size and association with downstream outcomes remains unclear. METHODS: In a prespecified analysis, we studied results among study participants treated with 2 doses of zalunfiban who had core laboratory measurements concentrations of hs-cTnT. RESULTS: More elevated hs-cTnT concentrations at presentation were associated with less resolution of ST deviation (P = .006) and more frequent Q wave development (P < .001). At coronary angiography more elevated hs-cTnT at presentation was associated with higher thrombus grade and worse epicardial and myocardial perfusion (all P < .05). In multivariable analyses, higher hs-cTnT concentrations at 24 hours were associated with greater adjusted risk for all-cause death (odds ratio [OR] 1.83 per log unit increase; P = .03), cardiovascular death (OR 1.83 per log unit increase; P = .03), heart failure (OR 2.74 per log unit increase; P < .001) or the composite of death and heart failure (P < .001) by 30 days. At 24 hours, those treated with zalunfiban had lower hs-cTnT compared to placebo (P = .04) and across multiples ≥ 10 to ≥ 1,000 times elevation, treatment with zalunfiban resulted in smaller hs-cTnT determined MI size. CONCLUSIONS: Among patients with STEMI, more elevated concentrations of hs-cTnT are associated with worse measures of reperfusion and higher-risk for short-term death or heart failure. A single dose of zalunfiban at first medical contact reduced MI size. TRIAL REGISTRATION: A phase 3 study of zalunfiban in subjects with ST-elevation MI (CELEBRATE); NCT04825743.

Humans↗

PubMed2026

Gastrointestinal digestion governs insect protein hydrolysis and predicted bioactive peptide release: Species-dependent implications for functional food applications.

This study investigates the digestion of insect proteins and the release of predicted bioactive peptides during human gastrointestinal digestion. Using the Infogest in vitro model, mealworm, cricket, and black soldier fly larvae (BSFL) proteins were digested and analyzed through discovery proteomics and bioinformatics to identify predicted bioactive peptides. Sequential windowed acquisition of all theoretical fragment ion mass spectra (SWATH-MS) quantified insect proteins including predicted bioactive peptide precursor proteins, the precursors of predicted bioactive peptides. Results indicated that gastrointestinal digestion strongly influences peptide release, with the gastric phase exhibiting a richer predicted bioactive peptide profile than the small intestinal phase. Many predicted bioactive peptides were rapidly hydrolysed under small intestine conditions, which may lead to reduced stability or diminished activity in vivo, potentially explaining why certain peptides show strong bioactivity in vitro but limited effects in vivo. Additionally, predicted bioactive peptide release varied by insect species, influenced by genetic factors and peptide abundance. These findings highlight the importance of species selection and consideration of proteolytic digestion patterns in optimizing insect-derived bioactive peptides for functional foods and nutraceutical applications.

Animals↗

PubMed2026

Whole-Genome Deep Learning Predicts Chemotherapy Response in Colorectal Cancer.

Chemotherapy response in colorectal cancer (CRC) exhibits significant heterogeneity, with current clinical predictors failing to capture complex genomic determinants of resistance. We developed a hybrid deep learning framework integrating convolutional neural networks (CNNs) and bidirectional long short-term memory (BiLSTM) networks to analyze whole-genome somatic mutations, evolutionary conservation, chromatin accessibility, and 3D genome architecture in 2,546 TCGA patients. An attention mechanism identified predictive genomic regions. The model achieved an AUC of 0.92 (95% CI: 0.89-0.94) in cross-validation and 0.88 (95% CI: 0.85-0.91) in independent validation, outperforming clinical models (ΔAUC = +0.18, p < 0.001). Key predictors included non-coding variants in TP53, KRAS, and PIK3CA regulatory regions. Triple-positive patients (mutations in all 3 regions) had significantly worse progression-free survival (HR = 4.7, p < 0.001). Our framework enables accurate chemotherapy response prediction and reveals novel non-coding resistance mechanisms, advancing precision oncology in CRC.

Humans↗

PubMed2026

Molecular evaluation of residual disease following neoadjuvant chemotherapy in triple-negative breast cancer CALGB 40603 (Alliance).

BACKGROUNDDespite therapeutic advances in early-stage triple-negative breast cancer (TNBC), residual disease (RD) following neoadjuvant therapy remains a key predictor of a worse prognosis and obstacle to improving patient outcomes.METHODSTo better characterize RD and identify survival-associated features, we performed comprehensive transcriptomic profiling of 340 pretreatment stage II/III TNBCs and 70 matched posttreatment RD samples from the randomized CALGB 40603 (Alliance) phase II clinical trial. To explore preclinical treatment strategies for RD, patient-derived xenograft (PDX) mouse models mimicking RD were treated with antibody-drug conjugates (ADCs).RESULTSOur study shows prognostic genomic features measured pretreatment may differ from prognostic features measured posttreatment from RD specimens. Patients with a genomic PAM50 subtype of basal-like in RD specimens had a poor survival outcome, and their matching pretreatment tumors were characterized by elevated chromosomal amplifications of oncogenic drivers and significantly reduced B and T cell expression features. Paired analyses of basal-like RD and matched pretreatment tumors revealed further lymphocyte depletion in RD, along with lower expression of MHC class I and interferon signaling, indicating an immune-cold RD microenvironment. Treatment of a basal-like and conventional chemotherapy-resistant PDX model, resembling basal-like RD, with sacituzumab govitecan or trastuzumab deruxtecan produced a marked antitumor response.CONCLUSIONRD biology differs from pretreatment tumors, with basal-like subtype RD following neoadjuvant chemotherapy being immune cold and associated with poor survival. Preclinical modeling suggests this high-risk group may benefit from adjuvant ADC therapy.TRIAL REGISTRATIONClinicalTrials.gov NCT00861705.FUNDINGNIH NCI U10CA180821 (Alliance for Clinical Trials in Oncology), NCI U24CA176171 (Alliance for Clinical Trials in Oncology), NCI UG1CA233373 (Alliance for Clinical Trials in Oncology), NCI Breast SPORE program P50-CA058223; Susan G. Komen SAC-160074; Breast Cancer Research Foundation BCRF-23-127; NIH NCI R01-CA229409; UNC LCCC Triple Negative Breast Cancer Center.

Humans↗

PubMed2026

Beyond predictive performance: A systematic review and critical methodological appraisal of AI/ML and conventional modelling strategies in breast, colorectal, and pancreatic Cancer.

BACKGROUND: Predictive modelling for cancer risk, treatment-related complications, and survival is central to precision oncology. Conventional logistic regression (LR) and Cox proportional hazards (CoxPH) regression remain widely used but are limited when modelling nonlinear interactions, high-dimensional imaging features, and multimodal clinical-metabolic predictors. Artificial intelligence (AI) and machine learning (ML) methods offer expanded capability through automated feature extraction, ensemble learning, and flexible survival modelling, but the evidence on when AI/ML adds value over conventional models across cancer sites and predictive tasks remains fragmented. OBJECTIVE: To systematically evaluate the methodological performance, validation strategies, and translational limitations of AI/ML models compared with conventional statistical models in published predictive-modelling studies for breast, colorectal, or pancreatic cancer. METHODS: PubMed, Scopus, and Web of Science were searched for studies published between January 2019 and March 2025. Two reviewers independently conducted title-and-abstract screening, full-text eligibility assessment, and PROBAST risk-of-bias assessment. Sixty-five studies (n = 907,567 participants) were narratively synthesised by cancer site, predictive task, model family, comparator, validation strategy, predictor modality, and calibration or explainability reporting. RESULTS: The 65 studies comprised breast cancer (n = 35), colorectal cancer (n = 21), and pancreatic cancer (n = 9). AI/ML superiority over LR and CoxPH was task- and data-dependent. CNN- and U-Net-based models predominated in imaging and body-composition tasks, tree-based ensembles consistently outperformed LR for tabular perioperative complication prediction, and CoxPH remained competitive, and in the largest pancreatic risk study, superior to XGBoost (C-index 0.802 vs 0.723) in well-structured datasets. PROBAST analysis-domain risk was moderate in 54 of 65 studies (83%), driven by limited external validation, sparse calibration reporting (11/65), and few decision-curve analyses (7/65). CONCLUSION: AI/ML adds the most methodological value in imaging-derived feature extraction and nonlinear perioperative prediction, while conventional regression remains preferable in large, structured datasets with linear predictors. Clinical translation requires standardised body-composition definitions, external validation, calibration assessment, decision-curve analysis, and explainability, in line with TRIPOD+AI and CLAIM standards.

Humans↗

PubMed2026

Predictive value of anal sphincter electromyography for sacral neuromodulation test-phase outcomes.

BACKGROUND: Sacral neuromodulation (SNM) is an established therapy for refractory pelvic organ dysfunction. Anal sphincter electromyography (EMG) is commonly used preoperatively to assess sacral and peripheral nerve integrity. However, the prognostic significance of chronic neurogenic EMG changes for SNM outcomes remains unclear. OBJECTIVE: To evaluate whether chronic neurogenic changes on preoperative anal sphincter EMG predict the outcome of the SNM test phase. METHODS: We retrospectively analysed 62 consecutive patients with bladder and/or bowel dysfunction or pelvic pain who were candidates for SNM treatment and who underwent preoperative anal sphincter EMG. EMG findings were classified as normal or showing chronic neurogenic changes. SNM test-phase success was defined as a ≥50% improvement of symptoms at 24 days. Outcomes were compared between EMG groups. RESULTS: Of the 62 patients (49 women, 13 men), 30 (48%) had normal EMG findings and 32 (52%) showed chronic neurogenic changes. Overall, the SNM test phase was successful in 47 patients (76%). Success rates were similar in patients with normal EMG (72%) and neurogenic EMG changes (79%), with no statistically significant difference (p = 0.878). Sex-stratified analyses revealed no significant association between EMG findings and test-phase success in women or men. CONCLUSIONS: Chronic neurogenic changes on anal sphincter EMG do not predict SNM test-phase outcomes. These findings suggest that abnormal sphincter EMG results should not be used as a standalone criterion to exclude patients from SNM therapy. SIGNIFICANCE: Signs of neurogenic damage on anal sphincter EMG are not an indicator of reduced neuromodulatory capacity or diminished clinical response to SNM.

Humans↗

PubMed2026

Externally validated risk prediction models for gestational diabetes mellitus: A systematic review and meta-analysis.

INTRODUCTION: Risk prediction models for gestational diabetes mellitus (GDM) offer potential for early identification and targeted prevention. External validation is crucial to assess model performance across diverse populations. Despite the availability of numerous GDM prediction models, limited evidence exists on their external validation frequency, methodological quality, and clinical applicability. This systematic review evaluated externally validated GDM prediction models, focusing on methodological rigor, reporting standards, and clinical relevance to inform future research and implementation. MATERIAL AND METHODS: Databases including Ovid MEDLINE, Embase, Scopus, Emcare, and CINAHL were searched up to May 1, 2025. Studies reporting external validation of GDM risk prediction models were included. Two reviewers independently screened studies. Data were extracted using the CHARMS framework, and risk of bias and applicability were assessed using PROBAST+AI. The study protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO; CRD420251125758). RESULTS: Twenty-six studies validated 33 models, with validation sample sizes ranging from 50 to 75 161. Over half used the IADPSG criteria to define GDM. Discrimination metrics were commonly reported, but calibration, overall performance, and clinical utility were often lacking. Meta-analysis was feasible for only four models: Teede et al., Nanda et al., Naylor et al., and Van Leeuwen et al., each showing fair discrimination. The Teede et al. model was the most widely validated, with 11 external validations across six continents and a pooled AUC of 0.72 (95% CI: 0.67-0.76). Despite fewer validations, the Nanda et al. model achieved the highest pooled discrimination (5 validations; pooled AUC 0.77, 95% CI: 0.74-0.80). The Naylor et al. and van Leeuwen et al. models also underwent meta-analysis, as sufficient external validation studies were available to support comparative performance assessment. Notably, 69.23% of studies had a high risk of bias. CONCLUSIONS: While many models showed acceptable predictive performance, most validations were methodologically weak. Future studies should follow best-practice guidelines and promote scalable validation strategies, such as algorithm sharing, to enhance clinical utility.

Humans↗

PubMed2026

Variability in β-human chorionic gonadotropin concentrations following evacuation of a hydatidiform mole pregnancy: A retrospective cohort study from Vietnam.

BackgroundGestational trophoblastic disease refers to a group of tumors defined by abnormal trophoblastic proliferation. This disease produces a distinct tumor marker, beta-human chorionic gonadotropin, which can be useful for diagnosis and follow-up. The objective of this study was to investigate the variations in serum beta-human chorionic gonadotropin levels after uterine evacuation as well and the progression of gestational trophoblastic neoplasia.Materials and methodsThis retrospective cohort study was conducted at Tu Du Hospital, Vietnam, between January 2019 and December 2020. All patients diagnosed with molar pregnancy were analyzed retrospectively based on serial serum beta-human chorionic gonadotropin levels following uterine evacuation. Post-evacuation outcomes, including relapsed molar pregnancy and gestational trophoblastic neoplasia, were also monitored.ResultsWe enrolled 560 patients with molar pregnancy, including 298 with complete hydatidiform mole and 262 with partial hydatidiform mole. Severe symptoms were more common in those with complete hydatidiform mole. Over the follow-up period, 97 cases of gestational trophoblastic neoplasia were noted. The data show that the median time to gestational trophoblastic neoplasia diagnosis was 8.75 ± 4.41 (4-26) weeks. In terms of variations in the serum beta-human chorionic gonadotropin levels, the generalized estimating equation model showed a faster decline in the complete hydatidiform mole group than in the partial hydatidiform mole group. Similarly, regression in serum beta-human chorionic gonadotropin levels was significantly more rapid in patients who progressed to gestational trophoblastic neoplasia than in those with relapsed molar pregnancy (-11,593 vs. -20,651.22 and -12,946.26 vs. -46,329.23 mUI/mL, p < 0.001).ConclusionsSurveillance of serum beta-human chorionic gonadotropin levels remains essential for gestational trophoblastic neoplasia monitoring in patients with molar pregnancy following surgical evacuation. The post-evacuation serum beta-human chorionic gonadotropin level regression curve helps distinguish gestational trophoblastic neoplasia from hydatidiform moles. Further evidence is required to strengthen these findings.

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

Comparative efficacy of LDL-C-lowering therapies in first-time vs. recurrent myocardial infarction prevention: a meta-analysis of large-scale randomized controlled trials.

AIMS: Reducing elevated low-density lipoprotein cholesterol (LDL-C) is central to global efforts to prevent myocardial infarction (MI). While many studies have evaluated LDL-C-lowering therapies in first-time and recurrent MI prevention, direct comparisons of their relative efficacy are lacking. Therefore, we conducted a systematic review and meta-analysis to compare the efficacy of LDL-C-lowering therapies in first-time vs. recurrent MI prevention. METHODS AND RESULTS: We searched three databases until 30 November 2024, for randomized controlled trials (RCTs) with at least 1000 patient-years of follow-up. Efficacy was quantified as relative risk (RR) with 95% confidence intervals (CIs). Differences in benefit magnitude were assessed using Cochran's Q test. Data were pooled with a random-effects model, and heterogeneity was measured using the I2 statistic. Additionally, we applied the Cochrane Risk of Bias Tool to evaluate study quality and utilized the GRADE method to assess the certainty of the evidence. This study included 22 large-scale RCTs involving 180 304 participants. In first-time MI prevention, LDL-C-lowering therapies achieved a remarkable 38% reduction in MI risk [12 RCTs; 79 604 participants; RR, 0.62 (95% CI, 0.55-0.69); P < 0.001]. In recurrent MI prevention, these therapies were associated with a more modest but significant 16% risk reduction [11 RCTs; 100 700 participants; RR, 0.84 (95% CI, 0.80-0.88); P < 0.001]. Importantly, the benefit magnitude between the two groups was significantly different (Q = 22.63; P < 0.001), highlighting the greater relative benefit in first-time MI prevention. Furthermore, the robustness of our findings was consistently supported by leave-one-out analyses, the absence of publication bias, high-quality GRADE evidence, and subgroup and sensitivity analyses. CONCLUSION: Our findings suggest that LDL-C-lowering therapies may offer a greater benefit in preventing first-time MI compared with recurrent MI.

Humans↗

PubMed2026

Predicting training outcomes for developmental dyslexia from EEG data.

Developmental dyslexia (DD) is characterised by lower-than-average reading abilities and is diagnosed in approximately 10% of individuals. The societal barriers may limit professional fulfilment and psychological wellbeing of individuals with DD, calling for the development of effective interventions to counteract them. As DD is associated with challenges in both phonological and visuo-attentional domains, different longitudinal training approaches were developed to strengthen them. However, they require a considerable amount of personal, social and economic resources and the outcomes may vary depending on individual differences in behavioural and neurophysiological functionality. Hence, predicting training outcomes might help in developing personalised treatment protocols and optimising the use of resources. In the present work we applied machine learning to resting-state EEG to predict longitudinal training outcomes in adults with DD enrolled in a randomized clinical trial. In particular, one group received a visuo-attentional training combined with transcranial alternating current stimulation (tACS), another group received visuo-attentional training with sham/placebo stimulation, and the third group received a phonological training with sham/placebo stimulation. The improvement in text reading speed was associated with spectral power in low-beta and individual frequencies in the alpha (IAF) and beta (IBF) bands, while the improvement in pseudoword reading was associated with IBF. The findings highlight the potential of capturing neural markers of treatment responsiveness in DD. Future studies should focus on the generalisability of predictive models to real-world settings, while investigating whether specific EEG markers predict responsiveness to distinct remediation protocols, thus supporting the development of personalised interventions.

Humans↗

PubMed2026

Machine learning-based prediction of unplanned readmission and construction of an online calculator for elderly patients with mild ischemic stroke.

OBJECTIVE: To screen for independent risk factors for unplanned readmission in elderly patients with mild ischemic stroke, and to construct and validate an online risk prediction calculator based on an interpretable machine learning model, thereby providing a promising practical tool for accurate clinical assessment of 30‑day all‑cause unplanned readmission risk in this population. METHODS: A prospective cohort study was conducted, including 1050 patients aged ≥ 60 years with mild ischemic stroke admitted between August 2023 and September 2024. Participants were randomly divided into a training set (840 cases) and a test set (210 cases) at a ratio of 8:2. Risk factors were screened by univariate analysis and multivariable Logistic regression. Four machine learning models, namely LightGBM, XGBoost, Random Forest, and K‑Nearest Neighbors (KNN), were developed and their performance was evaluated using AUC, accuracy, sensitivity, and specificity as metrics. The SHAP framework was used for interpretability analysis, and an online calculator was subsequently developed based on the optimal model. RESULTS: Univariate analysis showed significant differences (P < 0.05) in 13 factors including age, smoking, AIP, TyG index, HALP score, etc. Multivariable Logistic regression identified age (OR = 9.752), smoking (OR = 5.171), AIP (OR = 6.691), TyG index (OR = 4.393), HALP score (OR = 2.831), and ≥ 2 comorbidities (OR = 3.664) as independent risk factors. All four machine learning models demonstrated good predictive performance. Based on a comprehensive evaluation of multiple metrics and computational efficiency, the LightGBM model exhibited the best predictive performance (AUC = 0.884, accuracy = 0.829, sensitivity = 0.812, specificity = 0.875). SHAP analysis showed that age, AIP, TyG index, smoking, and HALP score were key predictors. An online calculator developed based on this model enables individualized risk predictions. CONCLUSION: Key risk factors associated with 30‑day unplanned readmission in elderly patients with mild ischemic stroke were identified. The LightGBM model demonstrated high predictive accuracy, and together with the interpretability analysis and online calculator, offers a practical tool to support clinical risk assessment. However, this tool requires future external validation.

Humans↗

PubMed2026

Risk prediction models for blood transfusion in patients undergoing total hip and knee arthroplasty: a systematic review and meta-analysis.

OBJECTIVE: To systematically review and evaluate published risk prediction models for perioperative blood transfusion in patients undergoing total hip or knee arthroplasty (THA/TKA). METHODS: We systematically searched PubMed, Web of Science, the Cochrane Library, and Embase from inception to May 31, 2025. Two researchers independently screened the literature, extracted data, and assessed the risk of bias and applicability using the Prediction model Risk Of Bias Assessment Tool (PROBAST). The area under the receiver operating characteristic curve (AUC) values were pooled via a meta-analysis using Stata 18.0. RESULTS: d Fourteen studies containing 36 prediction models were included. The incidence of blood transfusion among THA/TKA patients ranged from 3.2% to 30.8%. Preoperative hemoglobin (Hb) level, tranexamic acid (TXA) use, operative duration, intraoperative blood loss, and age were the most frequently incorporated predictors. Model sensitivity ranged from 58% to 94.5%, and specificity ranged from 71.3% to 94%. Meta-analysis showed that the pooled AUC value of the 13 validated models was 0.87 (95% CI: 0.85-0.90), suggesting good discriminatory performance. All models were rated as having a high risk of bias. The applicability of four studies was rated as unclear. CONCLUSION: Although the included studies demonstrated promising discriminative ability of prediction models for blood transfusion in THA/TKA, all were assessed as having a high risk of bias using the PROBAST tool. Therefore, future research should prioritize the development of models with larger sample sizes, rigorous study designs, and multicenter external validation.

Humans↗