Machine Learning Models in Prediabetes Screening: A Systematic Review

Identifying high-risk individuals using a risk prediction model could be a crucial first stage of screening pathways to improve the early detection of pancreatic cancer. A systematic review was conducted to critically evaluate the published primary literature on the development or validation of clinical risk prediction models for pancreatic cancer risk. MEDLINE, Embase, and Web of Science were searched for relevant articles from the inception of each database up to November 2021. Study selection and data extraction were conducted by 2 independent reviewers. The Prediction model Risk Of Bias Assessment Tool (PROBAST) was applied to assess risk of bias. In total, 33 studies were included, describing 38 risk prediction models. Excluding studies with an overlapping population, this study consist of 15,848,100 participants, of which 58,313 were diagnosed with pancreatic cancer. Eight studies externally validated their model, and 13 performed internal validation. The studies described risk prediction models for pancreatic cancer in the general population (n = 14), patients with diabetes (n = 8), and individuals with gastrointestinal (and other) symptoms (symptoms included abdominal pain, unexplained weight loss, jaundice, and change in bowel habits and indigestion; n = 11). The commonly used clinical risk factors in the model were cigarette smoking (n = 27), age (n = 25), diabetes history (n = 22), chronic pancreatitis (n = 18), and body mass index (n = 14). In the 25 studies that assessed model performance, C-statistics ranged from 0.61 to 0.98. Of the 33 studies included, 6 were rated as being at a low risk of bias based on PROBAST. Many clinical risk prediction models for pancreatic cancer had been developed for different target populations. Although low risk-of-bias studies were identified, these require external validation and implementation studies to ensure that these will benefit clinical decision making.

Prediction models for COVID-19 clinical decision making

Introduction: Numerous prediction models have been developed to identify high-risk individuals for lung cancer screening, aimed at improving early detection and survival rates. However, there has been no comprehensive review and meta-analysis of model performance across different sociocultural contexts. Therefore, this review aims to systematically examine the performance of lung cancer risk prediction models in Western and Asian populations. Methods: PubMed and EMBASE were searched from inception to January 2023. Studies published in English and proposed a validated model on human populations with well-defined predictive performances were included. Two reviewers independently screened titles and abstracts and the Prediction Model Risk of Bias Assessment Tool (PROBAST) was used to appraise study quality. A random-effects meta-analysis was performed and a 95% confidence interval (CI) for model performance was provided. Between-study heterogeneity was adjusted for using the Hartung-Knapp-Sidik-Honkman test. Results: Fifty-four studies were included, with 42 from Western and 12 from Asian countries. Most Western studies were designed for ever-smokers (19/42; 45.2%) and the general population (17/42; 40.5%), and only two Asian studies developed models exclusively for never-smokers. Across both Western and Asian prediction models, the three most consistently included risk factors are age, sex, and family cancer history. In 45.2% (19/42) of Western and 50.0% (6/12) of Asian studies, models incorporated both traditional risk factors and biomarkers; 14.8% (8/54) directly compared their biomarker-based models with those incorporating only traditional risk factors, demonstrating better discrimination. Machine-learning algorithms were applied in eight Western models and two Asian models. External validation of PLCOM2012 (AUC=0.748; 95% CI: 0.719-0.777) outperformed prediction models such as Bach (AUC=0.710; 95% CI: 0.674-0.745) and Spitz models (AUC=0.698; 95% CI: 0.640-0.755). Conclusion: Despite showing promising results, the majority of Asian risk models in our study lack external validation. Our review also reveals a significant gap in prediction models for never-smokers. Future research should externally validate existing Asian models or incorporate relevant Asian risk factors in widely-used Western models (PLCOM2012) to account for their unique risk profiles and lung cancer progression patterns. Citation Format: Yah Ru Juang, Lina Ang, Wei Jie Seow. The predictive performance of risk prediction models in determining lung cancer incidence in Western and Asian countries: A systematic review and meta-analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 6210.

BackgroundEmergency department boarding and hospital exit block are primary causes of emergency department crowding and have been conclusively associated with poor patient outcomes and major threats to patient safety. Boarding occurs when a patient is delayed or blocked from transitioning out of the emergency department because of dysfunctional transition or bed assignment processes. Predictive models for estimating the probability of an occurrence of this type could be useful in reducing or preventing emergency department boarding and hospital exit block, to reduce emergency department crowding.ObjectiveThe aim of this study was to identify and appraise the predictive performance, predictor utility, model application, and model utility of hospital admission prediction models that utilized prehospital, adult patient data and aimed to address emergency department crowding.MethodsWe searched multiple databases for studies, from inception to September 30, 2019, that evaluated models predicting adult patients’ imminent hospital admission, with prehospital patient data and regression analysis. We used PROBAST (Prediction Model Risk of Bias Assessment Tool) and CHARMS (Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies) to critically assess studies.ResultsPotential biases were found in most studies, which suggested that each model’s predictive performance required further investigation. We found that select prehospital patient data contribute to the identification of patients requiring hospital admission. Biomarker predictors may add superior value and advantages to models. It is, however, important to note that no models had been integrated with an information system or workflow, operated independently as electronic devices, or operated in real time within the care environment. Several models could be used at the site-of-care in real time without digital devices, which would make them suitable for low-technology or no-electricity environments.ConclusionsThere is incredible potential for prehospital admission prediction models to improve patient care and hospital operations. Patient data can be utilized to act as predictors and as data-driven, actionable tools to identify patients likely to require imminent hospital admission and reduce patient boarding and crowding in emergency departments. Prediction models can be used to justify earlier patient admission and care, to lower morbidity and mortality, and models that utilize biomarker predictors offer additional advantages.

Artificial intelligence (AI) prediction models can accurately identify high-risk populations by integrating multi-dimensional clinical data, providing decision support for doctors in formulating individualized discharge plans and optimizing follow-up intervention strategies, thereby reducing the risk of readmission from the source. Currently, the number of AI prediction models for readmission of critically ill patients is increasing, but the quality and applicability of these models in clinical practice and future research remain uncertain. To systematically evaluate published studies on AI prediction models for critically ill patients. This study conducted a computerized search of the CNKI, Wanfang Data, VIP, SinoMed, PubMed, Web of Science, Cochrane, and Embase databases, with the time range from 2020 to June 25, 2025. Information such as study design, data sources, outcome definitions, sample size, predictors, model development, and performance was extracted from the selected studies. The Prediction Model Risk of Bias Assessment Tool (PROBAST) checklist was used to evaluate the risk of bias and applicability. A total of 387 studies were retrieved, and after screening, 31 studies with their 31 prediction models were included in this review. All studies developed risk prediction models for readmission of critically ill patients using artificial intelligence algorithms. The readmission risk of critically ill patients ranged from 1.3% to 13.7%. The most commonly used predictors were structured data. The reported area under the curve (AUC) ranged from 0.66 to 0.98. All studies had a high risk of bias, mainly due to poor reporting quality in the analysis domain and insufficient applicability. The pooled AUC of the 24 validation models was 0.82, with a 95% confidence interval of 0.77-0.87. These study results constitute a comprehensive set of high-quality evidence, demonstrating that AI prediction models exhibit moderate-to-high predictive performance and that their predictive performance is significantly higher than that of traditional prediction models. No Patient or Public Contribution. This Meta-analysis is based on the systematic review and statistical combination of the published clinical research data. The processes of research design, data extraction, and result interpretation did not involve the participation of patients or the public. The protocol for this study has been registered in PROSPERO (registration number: CRD42025637829).

The incidence of Alzheimer’s disease (AD) continues to rise, and predictive models combining artificial intelligence (AI) with wearable devices offer a new approach for its detection and diagnosis. Existing reviews remain focused on traditional biomarkers, making it necessary to supplement the evidence in this field, particularly given the rapid advancements in wearable and AI technologies. The scoping review protocol aims to systematically evaluate AI-based predictive models using wearable devices for AD, with a focus on their measurement outcomes and model development processes. The review will follow the Arksey and O’Malley framework and incorporate PRISMA-ScR guidelines. This study will search multiple databases, including Web of Science, Cochrane Library, and PubMed, covering relevant gray literature. The quality of the included studies will be rigorously assessed using the Prediction Model Risk of Bias Assessment Tool (PROBAST) and Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) checklists. Two independent reviewers will conduct title and abstract screening, retrieve and assess full-text evidence sources, and extract data. The results will be narratively synthesized and presented in tables and figures. The knowledge gained from this review is expected to provide systematic evidence supporting AI-based predictive models that combine wearable devices for AD, potentially offering insights into model construction details such as data collection and external validation.

Risk prediction models for postoperative delirium in elderly patients with fragility hip fracture: A systematic review and critical appraisal

Significance: Postoperative Pressure Injuries (PIs) present unique risks, requiring dedicated research for accurate assessment. Despite the increasing number of Intraoperative Acquired Pressure Injury (IAPI) prediction models, their risk of bias and clinical applicability remains unclear. Recent Advances: Adhered to the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement requirements, IAPI prediction models of adult inpatients (≥18 years) were systematically retrieved from eight databases. Bias risk and applicability were evaluated using the Prediction model Risk Of Bias Assessment Tool (PROBAST), followed by narrative synthesis. Critical Issues: From 837 studies, 25 were included, covering 32 prediction models. Most studies (88%) were single-center and conducted in China, Korea, the United States, or Singapore, spanning various surgical specialties. Among 26,142 participants, IAPI incidence ranged from 4.1% to 41.75%. Common predictors included surgery duration, age, and diabetes. Areas Under the Curve (AUC) values varied from 0.702 to 0.984, but calibration was underreported. All studies had high bias risk, with 22 models exhibiting applicability concerns. Future Directions: The development of IAPI models requires a clear definition of the timing and personnel responsible for assessing PIs, with a preference for prospective data collection and thorough internal and external validation. Adherence to the critical appraisal and data extraction for systematic reviews of prediction modeling studies checklist and PROBAST guidelines can improve reporting quality. Models should be user-friendly, clinically applicable, and rigorously validated. Precisely defining and rigorously selecting predictors is critical to reducing variability. Future research should adopt more stringent designs to develop high-quality models capable of effectively guiding clinical practice. PROSPERO registration number: CRD42024502726.

ObjectiveThis study aimed to systematically evaluate published predictive models for dental caries in children and adolescents.DesignA systematic review and meta-analysis of observational studies.Data sourcesComprehensive searches were conducted in PubMed, Web...

BackgroundFailure to extubate successfully from mechanical ventilation is a critical event associated with poor prognosis in ICU patients, significantly prolonging hospital stays and increasing mortality rates. It is widely accepted in academic circles that developing prediction models for extubation failure can facilitate precise extubation decisions. Despite the rapid proliferation of relevant prediction models, their methodological quality and bedside applicability remain ambiguous.ObjectiveThis study aims to outline the predictive factors associated with the risk of extubation failure in patients undergoing mechanical ventilation in the Intensive Care Unit (ICU) and to summarize the existing predictive models.MethodsWe searched the China National Knowledge Infrastructure (CNKI), Wanfang Database, VIP Database, China Biomedical Database, PubMed, Embase, Web of Science, and Cochrane Library. We included both prospective and retrospective studies that developed or validated risk prediction models for extubation failure in patients undergoing mechanical ventilation in the ICU. The Prediction Model Risk of Bias Assessment Tool (PROBAST) was used to assess the bias and applicability of the models.ResultsThis analysis includes 14 studies. Frequency analysis of the predictors revealed that there are 15 predictors that appeared at least twice, among which mechanical ventilation duration, GCS score, APACHE II score, age, and hemoglobin were the most common predictors. From the perspective of the models, only 2 studies conducted both internal and external validation, 3 studies ultimately employed machine learning, while 11 studies utilized traditional modeling methods. However, we found that many studies faced issues such as insufficient sample sizes, missing crucial methodological information, and all models being rated as having a high risk of bias.ConclusionMost published predictive models lack methodological rigor, leading to a heightened risk of bias. Future research should prioritize the enhancement of methodological rigor and the external validation of risk prediction models for extubation failure in ICU patients receiving mechanical ventilation. Additionally, it is essential to emphasize adherence to scientific methods and transparent reporting to improve the accuracy and generalizability of research findings.Systematic review registrationhttps://www.crd.york.ac.uk/PROSPERO/recorddashboard, Registration number:CRD420251124371.

Risk prediction model for chemotherapy-induced nausea and vomiting in cancer patients: a systematic review.

This systematic review evaluated the performance and risk of bias in Systemic Lupus Erythematosus (SLE) disease manifestations and case fatality prediction models, based on a search of PubMed, Embase, and Cochrane Library up to October 17, 2025. Risk of bias was assessed using the Prediction model Risk Of Bias Assessment Tool (PROBAST). A random-effects meta-analysis pooled the Area Under the Curve (AUC) values with 95% Confidence Intervals (CIs), with sensitivity and subgroup analyses. The study included 35 studies comprising 89 prediction models, primarily from China (92.3%). Designs were mainly cross-sectional (46.2%) or retrospective cohort (42.3%). Common predictor categories included immunologic/autoantibody profile s (n = 148) and biochemical parameters (n = 126). During development, pulmonary (AUC = 0.92, 95% CI: 0.78-0.87), perinatal (0.92, 0.83-1.03), and case fatality models (0.91, 0.89-0.95) performed highly, while cardiovascular models scored lower (0.78, 0.74-0.81). Upon validation, pulmonary models remained superior (0.86, 0.81-0.91); perinatal (0.82, 0.77-0.87) and cardiovascular models (0.80, 0.76-0.83) remained robust, whereas case fatality models declined markedly. Machine learning models showed greater potential for pulmonary outcomes (0.83, 0.78-0.89). Predictor number improved renal model performance but reduced accuracy for case fatality. All 89 models were rated high risk of bias and mostly low applicability per PROBAST.

ObjectiveThe prognosis for acute exacerbation of chronic obstructive pulmonary disease (AECOPD) is not optimistic, and severe AECOPD leads to an increased risk of mortality. Prediction models help distinguish between high‐ and low‐risk groups. At present, many prediction models have been established and validated, which need to be systematically reviewed to screen out more suitable models that can be used in the clinic and provide evidence for future research.MethodsWe searched PubMed, EMBASE, Cochrane Library and Web of Science databases for studies on risk models for AECOPD mortality from their inception to 10 April 2022. The risk of bias was assessed using the prediction model risk of bias assessment tool (PROBAST). Stata software (version 16) was used to synthesize the C‐statistics for each model.ResultsA total of 37 studies were included. The development of risk prediction models for mortality in patients with AECOPD was described in 26 articles, in which the most common predictors were age (n = 17), dyspnea grade (n = 11), altered mental status (n = 8), pneumonia (n = 6) and blood urea nitrogen (BUN, n = 6). The remaining 11 articles only externally validated existing models. All 37 studies were evaluated at a high risk of bias using PROBAST. We performed a meta‐analysis of five models included in 15 studies. DECAF (dyspnoea, eosinopenia, consolidation, acidemia and atrial fibrillation) performed well in predicting in‐hospital death [C‐statistic = 0.91, 95% confidence interval (CI): 0.83, 0.98] and 90‐day death [C‐statistic = 0.76, 95% CI: 0.69, 0.82] and CURB‐65 (confusion, urea, respiratory rate, blood pressure and age) performed well in predicting 30‐day death [C‐statistic = 0.74, 95% CI: 0.70, 0.77].ConclusionsThis study provides information on the characteristics, performance and risk of bias of a risk model for AECOPD mortality. This pooled analysis of the present study suggests that the DECAF performs well in predicting in‐hospital and 90‐day deaths. Yet, external validation in different populations is still needed to prove this performance.

The Prediction model Risk Of Bias ASsessment Tool (PROBAST) is used to assess the quality, risk of bias, and applicability of prediction models or algorithms and of prediction model/algorithm studies....

BackgroundHIV treatment interruption remains a significant barrier to achieving global HIV/AIDS control goals. Machine learning (ML) models offer potential for predicting treatment interruption by leveraging large clinical data. Understanding how these models were developed, validated, and applied remains essential for advancing research.MethodsWe searched databases including the PubMed, BMC, Cochrane Library, Scopus, ScienceDirect, Lancet, and Google Scholar, for studies published in English from 1990 to September 2024. Search terms covered HIV, machine learning, treatment interruption, and loss to follow-up. Articles were screened and reviewed independently, and data were extracted using the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies (CHARMS) tool. Risk of bias was assessed with Prediction model Risk Of Bias Assessment Tool (PROBAST). The Preferred Reporting Items for Systematic reviews and Meta-analysis (PRISMA) guidelines were followed throughout.ResultsOut of 116,672 records, 9 studies met the inclusion criteria and reported 12 ML models. Random Forest, XGBoost, and AdaBoost were predominant models (91.7%). Internal validation was performed in all models, but only two models included external validation. Performance varied, with a mean area under the receiver operating characteristic curve (AUC-ROC) of 0.668 (standard deviation (SD) = 0.066), indicating moderate discrimination. About 75% of models showed a high risk of bias due to inadequate handling of missing data, lack of calibration, and the absence of decision curve analysis (DCA).ConclusionsML models show promise for predicting HIV treatment interruption, particularly in resource-limited settings. Future research should prioritize external validation, robust missing data handling, and decision curve analysis and include sociocultural predictors to improve model robustness.Systematic review registrationPROSPERO CRD42024578109.Supplementary InformationThe online version contains supplementary material available at 10.1186/s44263-025-00184-4.