Comparative analysis of integrative classification methods for multi-omics data.

Recent advances in sequencing, mass spectrometry, and cytometry technologies have enabled researchers to collect multiple 'omics data types from a single sample. These large datasets have led to a growing consensus that a holistic approach is needed to identify new candidate biomarkers and unveil mechanisms underlying disease etiology, a key to precision medicine. While many reviews and benchmarks have been conducted on unsupervised approaches, their supervised counterparts have received less attention in the literature and no gold standard has emerged yet. In this work, we present a thorough comparison of a selection of six methods, representative of the main families of intermediate integrative approaches (matrix factorization, multiple kernel methods, ensemble learning, and graph-based methods). As non-integrative control, random forest was performed on concatenated and separated data types. Methods were evaluated for classification performance on both simulated and real-world datasets, the latter being carefully selected to cover different medical applications (infectious diseases, oncology, and vaccines) and data modalities. A total of 15 simulation scenarios were designed from the real-world datasets to explore a large and realistic parameter space (e.g. sample size, dimensionality, class imbalance, effect size). On real data, the method comparison showed that integrative approaches performed better or equally well than their non-integrative counterpart. By contrast, DIABLO and the four random forest alternatives outperform the others across the majority of simulation scenarios. The strengths and limitations of these methods are discussed in detail as well as guidelines for future applications.

Optimal fusion of genotype and drug embeddings in predicting cancer drug response.

Predicting cancer drug response using both genomics and drug features has shown some success compared to using genomics features alone. However, there has been limited research done on how best to combine or fuse the two types of features. Using a visible neural network with two deep learning branches for genes and drug features as the base architecture, we experimented with different fusion functions and fusion points. Our experiments show that injecting multiplicative relationships between gene and drug latent features into the original concatenation-based architecture DrugCell significantly improved the overall predictive performance and outperformed other baseline models. We also show that different fusion methods respond differently to different fusion points, indicating that the relationship between drug features and different hierarchical biological level of gene features is optimally captured using different methods. Considering both predictive performance and runtime speed, tensor product partial is the best-performing fusion function to combine late-stage representations of drug and gene features to predict cancer drug response.

A review of disease risk prediction methods and applications in the omics era.

Risk prediction and disease prevention are the innovative care challenges of the 21st century. Apart from freeing the individual from the pain of disease, it will lead to low medical costs for society. Until very recently, risk assessments have ushered in a new era with the emergence of omics technologies, including genomics, transcriptomics, epigenomics, proteomics, and so on, which potentially advance the ability of biomarkers to aid prediction models. While risk prediction has achieved great success, there are still some challenges and limitations. We reviewed the general process of omics-based disease risk model construction and the applications in four typical diseases. Meanwhile, we highlighted the problems in current studies and explored the potential opportunities and challenges for future clinical practice.

Multi-omic prediction of incident type 2 diabetes.

AIMS/HYPOTHESIS: The identification of people who are at high risk of developing type 2 diabetes is a key part of population-level prevention strategies. Previous studies have evaluated the predictive utility of omics measurements, such as metabolites, proteins or polygenic scores, but have considered these separately. The improvement that combined omics biomarkers can provide over and above current clinical standard models is unclear. The aim of this study was to test the predictive performance of genome, proteome, metabolome and clinical biomarkers when added to established clinical prediction models for type 2 diabetes. METHODS: We developed sparse interpretable prediction models in a prospective, nested type 2 diabetes case-cohort study (N=1105, incident type 2 diabetes cases=375) with 10,792 person-years of follow-up, selecting from 5759 features across the genome, proteome, metabolome and clinical biomarkers using least absolute shrinkage and selection operator (LASSO) regression. We compared the predictive performance of omics-derived predictors with a clinical model including the variables from the Cambridge Diabetes Risk Score and HbA1c. RESULTS: Among single omics prediction models that did not include clinical risk factors, the top ten proteins alone achieved the highest performance (concordance index [C index]=0.82 [95% CI 0.75, 0.88]), suggesting the proteome as the most informative single omic layer in the absence of clinical information. However, the largest improvement in prediction of type 2 diabetes incidence over and above the clinical model was achieved by the top ten features across several omic layers (C index=0.87 [95% CI 0.82, 0.92], Δ C index=0.05, p=0.045). This improvement by the top ten omic features was also evident in individuals with HbA1c <42 mmol/mol (6.0%), the threshold for prediabetes (C index=0.84 [95% CI 0.77, 0.90], Δ C index=0.07, p=0.03), the group in whom prediction would be most useful since they are not targeted for preventative interventions by current clinical guidelines. In this subgroup, the type 2 diabetes polygenic risk score was the major contributor to the improvement in prediction, and achieved a comparable improvement in performance when added onto the clinical model alone (C index=0.83 [95% CI 0.75, 0.90], Δ C index=0.06, p=0.002). However, compared with those with prediabetes, individuals at high polygenic risk in this group had only around half the absolute risk for type 2 diabetes over a 20 year period. CONCLUSIONS/INTERPRETATION: Omic approaches provided marginal improvements in prediction of incident type 2 diabetes. However, while a polygenic risk score does improve prediction in people with an HbA1c in the normoglycaemic range, the group in whom prediction would be most useful, even individuals with a high polygenic burden in that subgroup had a low absolute type 2 diabetes risk. This suggests a limited feasibility of implementing targeted population-based genetic screening for preventative interventions.

A Systematic Review of Predictor Composition, Outcomes, Risk of Bias, and Validation of COVID-19 Prognostic Scores.

BACKGROUND: Numerous prognostic scores have been published to support risk stratification for patients with coronavirus disease 2019 (COVID-19). METHODS: We performed a systematic review to identify the scores for confirmed or clinically assumed COVID-19 cases. An in-depth assessment and risk of bias (ROB) analysis (Prediction model Risk Of Bias ASsessment Tool [PROBAST]) was conducted for scores fulfilling predefined criteria ([I] area under the curve [AUC)] ≥ 0.75; [II] a separate validation cohort present; [III] training data from a multicenter setting [≥2 centers]; [IV] point-scale scoring system). RESULTS: Out of 1522 studies extracted from MEDLINE/Web of Science (20/02/2023), we identified 242 scores for COVID-19 outcome prognosis (mortality 109, severity 116, hospitalization 14, long-term sequelae 3). Most scores were developed using retrospective (75.2%) or single-center (57.1%) cohorts. Predictor analysis revealed the primary use of laboratory data and sociodemographic information in mortality and severity scores. Forty-nine scores were included in the in-depth analysis. The results indicated heterogeneous quality and predictor selection, with only five scores featuring low ROB. Among those, based on the number and heterogeneity of validation studies, only the 4C Mortality Score can be recommended for clinical application so far. CONCLUSIONS: The application and translation of most existing COVID scores appear unreliable. Guided development and predictor selection would have improved the generalizability of the scores and may enhance pandemic preparedness in the future.

Colorectal Cancer Risk Assessment and Precision Approaches to Screening: Brave New World or Worlds Apart?

Current colorectal cancer (CRC) screening recommendations take a "one-size-fits-all" approach using age as the major criterion to initiate screening. Precision screening that incorporates factors beyond age to risk stratify individuals could improve on current approaches and optimally use available resources with benefits for patients, providers, and health care systems. Prediction models could identify high-risk groups who would benefit from more intensive screening, while low-risk groups could be recommended less intensive screening incorporating noninvasive screening modalities. In addition to age, prediction models incorporate well-established risk factors such as genetics (eg, family CRC history, germline, and polygenic risk scores), lifestyle (eg, smoking, alcohol, diet, and physical inactivity), sex, and race and ethnicity among others. Although several risk prediction models have been validated, few have been systematically studied for risk-adapted population CRC screening. In order to envisage clinical implementation of precision screening in the future, it will be critical to develop reliable and accurate prediction models that apply to all individuals in a population; prospectively study risk-adapted CRC screening on the population level; garner acceptance from patients and providers; and assess feasibility, resources, cost, and cost-effectiveness of these new paradigms. This review evaluates the current state of risk prediction modeling and provides a roadmap for future implementation of precision CRC screening.

A meta-inference framework to integrate multiple external models into a current study.

It is becoming increasingly common for researchers to consider incorporating external information from large studies to improve the accuracy of statistical inference instead of relying on a modestly sized data set collected internally. With some new predictors only available internally, we aim to build improved regression models based on individual-level data from an "internal" study while incorporating summary-level information from "external" models. We propose a meta-analysis framework along with two weighted estimators as the composite of empirical Bayes estimators, which combines the estimates from different external models. The proposed framework is flexible and robust in the ways that (i) it is capable of incorporating external models that use a slightly different set of covariates; (ii) it is able to identify the most relevant external information and diminish the influence of information that is less compatible with the internal data; and (iii) it nicely balances the bias-variance trade-off while preserving the most efficiency gain. The proposed estimators are more efficient than the naïve analysis of the internal data and other naïve combinations of external estimators.

Artificial intelligence in the risk prediction models of cardiovascular disease and development of an independent validation screening tool: a systematic review.

BACKGROUND: A comprehensive overview of artificial intelligence (AI) for cardiovascular disease (CVD) prediction and a screening tool of AI models (AI-Ms) for independent external validation are lacking. This systematic review aims to identify, describe, and appraise AI-Ms of CVD prediction in the general and special populations and develop a new independent validation score (IVS) for AI-Ms replicability evaluation. METHODS: PubMed, Web of Science, Embase, and IEEE library were searched up to July 2021. Data extraction and analysis were performed for the populations, distribution, predictors, algorithms, etc. The risk of bias was evaluated with the prediction risk of bias assessment tool (PROBAST). Subsequently, we designed IVS for model replicability evaluation with five steps in five items, including transparency of algorithms, performance of models, feasibility of reproduction, risk of reproduction, and clinical implication, respectively. The review is registered in PROSPERO (No. CRD42021271789). RESULTS: In 20,887 screened references, 79 articles (82.5% in 2017-2021) were included, which contained 114 datasets (67 in Europe and North America, but 0 in Africa). We identified 486 AI-Ms, of which the majority were in development (n = 380), but none of them had undergone independent external validation. A total of 66 idiographic algorithms were found; however, 36.4% were used only once and only 39.4% over three times. A large number of different predictors (range 5-52,000, median 21) and large-span sample size (range 80-3,660,000, median 4466) were observed. All models were at high risk of bias according to PROBAST, primarily due to the incorrect use of statistical methods. IVS analysis confirmed only 10 models as "recommended"; however, 281 and 187 were "not recommended" and "warning," respectively. CONCLUSION: AI has led the digital revolution in the field of CVD prediction, but is still in the early stage of development as the defects of research design, report, and evaluation systems. The IVS we developed may contribute to independent external validation and the development of this field.

Rheumatoid arthritis patients' perspective on the use of prediction models in clinical decision-making.

OBJECTIVES: A rapidly expanding number of prediction models is being developed aiming to improve rheumatoid arthritis (RA) diagnosis and treatment. However, few are actually implemented in clinical practice. This study explores factors influencing the acceptance of prediction models in clinical decision-making by RA patients. METHODS: A qualitative study design was used with thematic analysis of semi-structured interviews. Purposive sampling was applied to capture a complete overview of influencing factors. The interview topic list was based on pilot data. RESULTS: Data saturation was reached after 12 interviews. Patients were generally positive about the use of prediction models in clinical decision-making. Six key themes were identified from the interviews. First, patients have the need for information on prediction models. Second, factors influencing trust in model-supported treatment are described. Third, patients envision the model to have a supportive role in clinical decision-making. Fourth, patients hope to personally benefit from model-supported treatment in various ways. Fifth, patients are willing to contribute time and effort to contribute to model input. And lastly, we discuss the theme on effects of the relationship with the caregiver in model-supported treatment. CONCLUSION: Within this study RA patients were generally positive about the use of prediction models in their treatment given some conditions were met and concerns addressed. The results of this study can be used during the development and implementation in RA care of prediction models in order to enhance patient acceptability.

Prediction of Morbidity and Mortality After Esophagectomy: A Systematic Review.

BACKGROUND: Esophagectomy for esophageal cancer has a complication rate of up to 60%. Prediction models could be helpful to preoperatively estimate which patients are at increased risk of morbidity and mortality. The objective of this study was to determine the best prediction models for morbidity and mortality after esophagectomy and to identify commonalities among the models. PATIENTS AND METHODS: A systematic review was performed in accordance to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement and was prospectively registered in PROSPERO ( https://www.crd.york.ac.uk/prospero/ , study ID CRD42022350846). Pubmed, Embase, and Clarivate Analytics/Web of Science Core Collection were searched for studies published between 2010 and August 2022. The Prediction model Risk of Bias Assessment Tool was used to assess the risk of bias. Extracted data were tabulated and a narrative synthesis was performed. RESULTS: Of the 15,011 articles identified, 22 studies were included using data from tens of thousands of patients. This systematic review included 33 different models, of which 18 models were newly developed. Many studies showed a high risk of bias. The prognostic accuracy of models differed between 0.51 and 0.85. For most models, variables are readily available. Two models for mortality and one model for pulmonary complications have the potential to be developed further. CONCLUSIONS: The availability of rigorous prediction models is limited. Several models are promising but need to be further developed. Some models provide information about risk factors for the development of complications. Performance status is a potential modifiable risk factor. None are ready for clinical implementation.

Plasma metabolomic and lipidomic profiles accurately classify mothers of children with congenital heart disease: an observational study.

INTRODUCTION: Congenital heart disease (CHD) is the most common congenital anomaly, representing a significant global disease burden. Limitations exist in our understanding of aetiology, diagnostic methodology and screening, with metabolomics offering promise in addressing these. OBJECTIVE: To evaluate maternal metabolomics and lipidomics in prediction and risk factor identification for childhood CHD. METHODS: We performed an observational study in mothers of children with CHD following pregnancy, using untargeted plasma metabolomics and lipidomics by ultrahigh performance liquid chromatography-high resolution mass spectrometry (UHPLC-HRMS). 190 cases (157 mothers of children with structural CHD (sCHD); 33 mothers of children with genetic CHD (gCHD)) from the children OMACp cohort and 162 controls from the ALSPAC cohort were analysed. CHD diagnoses were stratified by severity and clinical classifications. Univariate, exploratory and supervised chemometric methods were used to identify metabolites and lipids distinguishing cases and controls, alongside predictive modelling. RESULTS: 499 metabolites and lipids were annotated and used to build PLS-DA and SO-CovSel-LDA predictive models to accurately distinguish sCHD and control groups. The best performing model had an sCHD test set mean accuracy of 94.74% (sCHD test group sensitivity 93.33%; specificity 96.00%) utilising only 11 analytes. Similar test performances were seen for gCHD. Across best performing models, 37 analytes contributed to performance including amino acids, lipids, and nucleotides. CONCLUSIONS: Here, maternal metabolomic and lipidomic analysis has facilitated the development of sensitive risk prediction models classifying mothers of children with CHD. Metabolites and lipids identified offer promise for maternal risk factor profiling, and understanding of CHD pathogenesis in the future.

IPF-related new macrophage subpopulations and diagnostic biomarker identification - combine machine learning with single-cell analysis.

Idiopathic pulmonary fibrosis (IPF) is a chronic disease of unknown etiology that lacks a specific treatment. In IPF, macrophages play a key regulatory role as a major component of the lung immune system, especially during inflammation and fibrosis. However, our understanding of the cellular heterogeneity and molecular characterization of macrophages in IPF, as well as their relevance in the clinical setting, is relatively limited. In this study, we analyzed in-depth single-cell transcriptome sequencing (scRNA-seq) data from lung tissues of IPF patients, identified macrophage subpopulations in IPF, and probed their molecular characteristics and biological functions. hdWGCNA identified co-expressed gene modules of a subpopulation of IPF-associated macrophages (IPF-MΦ), and probed the IPF-MΦ by a machine-learning approach. hdWGCNA identified a subpopulation of IPF-associated macrophage subpopulations and probed the IPF-MΦ signature gene (IRMG) for its prognostic value, and a prediction model was developed on this basis. In addition, IPF-MΦ was obtained after recluster analysis of macrophages in IPF lung tissues. Coexpressed gene modules of IPF-MΦ were identified by hdWGCNA. Then, a machine learning approach was utilized to reveal the characteristic genes of IPF-MΦ, and a prediction model was built on this basis. In addition, we discovered a type of macrophage unique to IPF lung tissue named ATP5-MΦ. Its characteristic gene encodes a subunit of the mitochondrial ATP synthase complex, which is closely related to oxidative phosphorylation and proton transmembrane transport, suggesting that ATP5-MΦ may have higher ATP synthesis capacity in IPF lung tissue. This study provides new insights into the pathogenesis of IPF and provides a basis for evaluating disease prognosis and predictive medicine in IPF patients.

A comparison of hyperparameter tuning procedures for clinical prediction models: A simulation study.

Tuning hyperparameters, such as the regularization parameter in Ridge or Lasso regression, is often aimed at improving the predictive performance of risk prediction models. In this study, various hyperparameter tuning procedures for clinical prediction models were systematically compared and evaluated in low-dimensional data. The focus was on out-of-sample predictive performance (discrimination, calibration, and overall prediction error) of risk prediction models developed using Ridge, Lasso, Elastic Net, or Random Forest. The influence of sample size, number of predictors and events fraction on performance of the hyperparameter tuning procedures was studied using extensive simulations. The results indicate important differences between tuning procedures in calibration performance, while generally showing similar discriminative performance. The one-standard-error rule for tuning applied to cross-validation (1SE CV) often resulted in severe miscalibration. Standard non-repeated and repeated cross-validation (both 5-fold and 10-fold) performed similarly well and outperformed the other tuning procedures. Bootstrap showed a slight tendency to more severe miscalibration than standard cross-validation-based tuning procedures. Differences between tuning procedures were larger for smaller sample sizes, lower events fractions and fewer predictors. These results imply that the choice of tuning procedure can have a profound influence on the predictive performance of prediction models. The results support the application of standard 5-fold or 10-fold cross-validation that minimizes out-of-sample prediction error. Despite an increased computational burden, we found no clear benefit of repeated over non-repeated cross-validation for hyperparameter tuning. We warn against the potentially detrimental effects on model calibration of the popular 1SE CV rule for tuning prediction models in low-dimensional settings.

Machine Learning Algorithms Versus Classical Regression Models in Pre-Eclampsia Prediction: A Systematic Review.

PURPOSE OF REVIEW: Machine learning (ML) approaches are an emerging alternative for healthcare risk prediction. We aimed to synthesise the literature on ML and classical regression studies exploring potential prognostic factors and to compare prediction performance for pre-eclampsia. RECENT FINDINGS: From 9382 studies retrieved, 82 were included. Sixty-six publications exclusively reported eighty-four classical regression models to predict variable timing of onset of pre-eclampsia. Another six publications reported purely ML algorithms, whilst another 10 publications reported ML algorithms and classical regression models in the same sample with 8 of 10 findings that ML algorithms outperformed classical regression models. The most frequent prognostic factors were age, pre-pregnancy body mass index, chronic medical conditions, parity, prior history of pre-eclampsia, mean arterial pressure, uterine artery pulsatility index, placental growth factor, and pregnancy-associated plasma protein A. Top performing ML algorithms were random forest (area under the curve (AUC) = 0.94, 95% confidence interval (CI) 0.91-0.96) and extreme gradient boosting (AUC = 0.92, 95% CI 0.90-0.94). The competing risk model had similar performance (AUC = 0.92, 95% CI 0.91-0.92) compared with a neural network. Calibration performance was not reported in the majority of publications. ML algorithms had better performance compared to classical regression models in pre-eclampsia prediction. Random forest and boosting-type algorithms had the best prediction performance. Further research should focus on comparing ML algorithms to classical regression models using the same samples and evaluation metrics to gain insight into their performance. External validation of ML algorithms is warranted to gain insights into their generalisability.

Prediction models for children/adolescents with obesity/overweight: A systematic review and meta-analysis.

The incidence of obesity and overweight in children and adolescents is increasing worldwide and becomes a global health concern. This study aims to evaluate the accuracy of available prediction models in early identification of obesity and overweight in general children or adolescents and identify predictive factors for the models, thus provide a reference for subsequent development of risk prediction tools for obesity and overweight in children or adolescents. Related publications were obtained from several databases such as PubMed, Embase, Cochrane Library, and Web of Science from their inception to September 18th, 2022. The novel Prediction Model Risk of Bias Assessment Tool (PROBAST) was employed to assess the bias risk of the included studies. R4.2.0 and Stata15.1 softwares were used to conduct meta-analysis. This study involved 45 cross-sectional and/or prospective studies with 126 models. Meta-analyses showed that the overall pooled index of concordance (c-index) of prediction models for children/adolescents with obesity and overweight in the training set was 0.769 (95% CI 0.754-0.785) and 0.835(95% CI 0.792-0.879), respectively. Additionally, a large number of predictors were found to be related to children's lifestyles, such as sleep duration, sleep quality, and eating speed. In conclusions, prediction models can be employed to predict obesity/overweight in children and adolescents. Most predictors are controllable factors and are associated with lifestyle. Therefore, the prediction model serves as an excellent tool to formulate effective strategies for combating obesity/overweight in pediatric patients.

The impact of different imputation methods on estimates and model performance: an example using a risk prediction model for premature mortality.

OBJECTIVE: To compare how different imputation methods affect the estimates and performance of a prediction model for premature mortality. STUDY DESIGN AND SETTING: Sex-specific Weibull accelerated failure time survival models were run on four separate datasets using complete case, mode, single and multiple imputation to impute missing values. Six performance measures were compared to access predictive accuracy (Nagelkerke R2, integrated brier score), discrimination (Harrell's c-index, discrimination slope) and calibration (calibration in the large, calibration slope). RESULTS: The highest proportion of missingness for a single variable was 10.86% for the female model and 8.24% for the male model. Comparing the performance measures for complete case, mode, single and multiple imputation: the Nagelkerke R2 values for the female model was 0.1084, 0.1116, 0.1120 and 0.111-0.1120 with the male model exhibited similar variation of 0.1050, 0.1078, 0.1078 and 0.1078-0.1081. Harrell's c-index also demonstrated small variation with values of 0.8666, 0.8719, 0.8719 and 0.8711-0.8719 for the female model and 0.8549, 0.8548, 0.8550 and 0.8550-0.8553 for the male model. CONCLUSION: In the scenarios examined in this study, mode imputation performed well when using a population health survey compared to single and multiple imputation when predictive performance measures is the main model goal. To generate unbiased hazard ratios, multiple imputation methods were superior. This study shows the need to consider the best imputation approach for a predictive model development given the conditions of missing data and the goals of the analysis.

Artificial Intelligence and IBD: Where are We Now and Where Will We Be in the Future?

PURPOSE OF REVIEW: Artificial intelligence (AI) is quickly demonstrating the ability to address problems and challenges in the care of IBD. This review with commentary will highlight today's advancements in AI applications for IBD in image analysis, understanding text, and replicating clinical knowledge and experience. RECENT FINDINGS: Advancements in machine learning methods, availability of high-performance computing, and increasing digitization of medical data are providing opportunities for AI to assist in IBD care. Multiple groups have demonstrated the ability of AI to replicate expert endoscopic scoring in IBD, with expansion into automated capsule endoscopy, enterography, and histologic interpretations. Further, AI image analysis is being used to develop new endoscopic scoring with more granularity and detail than is possible using conventional methods. Advancements in natural language processing are proving to reduce laborious tasks required in the care of IBD, including documentation, information searches, and chart review. Finally, large language models and chatbots that can understand language and generate human-like replies are beginning to exhibit clinical intelligence that will revolutionize how we deliver IBD care. Today, AI is being deployed to replicate expert judgement in specific tasks where disagreement, subjectivity, and bias are common. However, the near future will herald contributions of AI doing what we cannot, including new detailed measures of IBD, enhanced analysis of images, and perhaps even fully automating care. As we speculate on future technologic capabilities that may improve how we care for IBD, this review will also consider how we will implement and fairly use AI in practice.

Evaluating accuracy and fairness of clinical decision support algorithms when health care resources are limited.

OBJECTIVE: Guidance on how to evaluate accuracy and algorithmic fairness across subgroups is missing for clinical models that flag patients for an intervention but when health care resources to administer that intervention are limited. We aimed to propose a framework of metrics that would fit this specific use case. METHODS: We evaluated the following metrics and applied them to a Veterans Health Administration clinical model that flags patients for intervention who are at risk of overdose or a suicidal event among outpatients who were prescribed opioids (N = 405,817): Receiver - Operating Characteristic and area under the curve, precision - recall curve, calibration - reliability curve, false positive rate, false negative rate, and false omission rate. In addition, we developed a new approach to visualize false positives and false negatives that we named 'per true positive bars.' We demonstrate the utility of these metrics to our use case for three cohorts of patients at the highest risk (top 0.5 %, 1.0 %, and 5.0 %) by evaluating algorithmic fairness across the following age groups: <=30, 31-50, 51-65, and >65 years old. RESULTS: Metrics that allowed us to assess group differences more clearly were the false positive rate, false negative rate, false omission rate, and the new 'per true positive bars'. Metrics with limited utility to our use case were the Receiver - Operating Characteristic and area under the curve, the calibration - reliability curve, and the precision - recall curve. CONCLUSION: There is no "one size fits all" approach to model performance monitoring and bias analysis. Our work informs future researchers and clinicians who seek to evaluate accuracy and fairness of predictive models that identify patients to intervene on in the context of limited health care resources. In terms of ease of interpretation and utility for our use case, the new 'per true positive bars' may be the most intuitive to a range of stakeholders and facilitates choosing a threshold that allows weighing false positives against false negatives, which is especially important when predicting severe adverse events.

Ventilatory efficiency as a prognostic factor for postoperative complications in patients undergoing elective major surgery: a systematic review.

BACKGROUND: Major surgery is associated with high complication rates. Several risk scores exist to assess individual patient risk before surgery but have limited precision. Novel prognostic factors can be included as additional building blocks in existing prediction models. A candidate prognostic factor, measured by cardiopulmonary exercise testing, is ventilatory efficiency (VE/VCO2). The aim of this systematic review was to summarise evidence regarding VE/VCO2 as a prognostic factor for postoperative complications in patients undergoing major surgery. METHODS: A medical library specialist developed the search strategy. No database-provided limits, considering study types, languages, publication years, or any other formal criteria were applied to any of the sources. Two reviewers assessed eligibility of each record and rated risk of bias in included studies. RESULTS: From 10,082 screened records, 65 studies were identified as eligible. We extracted adjusted associations from 32 studies and unadjusted from 33 studies. Risk of bias was a concern in the domains 'study confounding' and 'statistical analysis'. VE/VCO2 was reported as a prognostic factor for short-term complications after thoracic and abdominal surgery. VE/VCO2 was also reported as a prognostic factor for mid- to long-term mortality. Data-driven covariable selection was applied in 31 studies. Eighteen studies excluded VE/VCO2 from the final multivariable regression owing to data-driven model-building approaches. CONCLUSIONS: This systematic review identifies VE/VCO2 as a predictor for short-term complications after thoracic and abdominal surgery. However, the available data do not allow conclusions about clinical decision-making. Future studies should select covariables for adjustment a priori based on external knowledge. SYSTEMATIC REVIEW PROTOCOL: PROSPERO (CRD42022369944).

Prediction of misfolded proteins spreading in Alzheimer's disease using machine learning and spreading models.

The pervasive impact of Alzheimer's disease on aging society represents one of the main challenges at this time. Current investigations highlight 2 specific misfolded proteins in its development: Amyloid-$\beta$ and tau. Previous studies focused on spreading for misfolded proteins exploited simulations, which required several parameters to be empirically estimated. Here, we provide an alternative view based on 2 machine learning approaches which we compare with known simulation models. The first approach applies an autoregressive model constrained by structural connectivity, while the second is based on graph convolutional networks. The aim is to predict concentrations of Amyloid-$\beta$ 2 yr after a provided baseline. We also evaluate its real-world effectiveness and suitability by providing a web service for physicians and researchers. In experiments, the autoregressive model generally outperformed state-of-the-art models resulting in lower prediction errors. While it is important to note that a comprehensive prognostic plan cannot solely rely on amyloid beta concentrations, their prediction, achieved by the discussed approaches, can be valuable for planning therapies and other cures, especially when dealing with asymptomatic patients for whom novel therapies could prove effective.