SEARCH · PubMed Health

Results for “Predictive Learning Models”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 37 records · Page 2Linked to original sources

PubMed2026

Machine learning-based drug susceptibility prediction from Candida genomic data.

OBJECTIVES: Invasive Candida infection is an increasing clinical concern, with antifungal resistance rising across multiple species. However, rapid and accurate antifungal susceptibility testing (AFST) remains limited in routine practice. The study evaluated species distribution and antifungal susceptibility of invasive Candida isolates in China and assessed the feasibility of combining whole-genome sequencing (WGS) with machine learning to predict minimum inhibitory concentrations (MICs). METHODS: Consecutive non-repetitive isolates were collected from 20 hospitals in 13 provinces during 2022-2023. MICs of nine antifungal agents were determined by broth microdilution, and WGS was performed for species accounting for >5% of the total isolates. Genomic 11-mer features were extracted and used to train random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGBoost) models, followed by optimization of the best-performing algorithm. RESULTS: A total of 337 isolates were obtained from blood (n = 232) and sterile body fluids (n = 105), comprising C. albicans (n = 103), C. tropicalis (n = 71), C. parapsilosis (n = 67), and C. glabrata (n = 63). Non-albicans Candida showed higher azole and echinocandin resistance, with C. tropicalis notably resistant to azoles and C. glabrata to echinocandins. Among the three models, RF demonstrated the best performance on 304 sequenced isolates. The optimized RF model was evaluated by the receiver operating characteristic (ROC) curve analysis and achieved an average area under the ROC curve (AUC) of 0.979 (95% CI: 0.974-0.984), essential agreement over 90.1%, and categorical agreement over 93.2% across species. CONCLUSIONS: These findings underscore the clinical challenge posed by non-albicans Candida resistance, and indicate that WGS-based MIC prediction may offer a highly accurate reference for earlier antifungal therapy.

Antifungal Agents↗

PubMed2026

Proteomics-driven discovery of intervention windows and risk subtypes in osteoporosis: A prospective cohort study.

Given the limited feasibility of population-wide bone mineral density screening and the infrequency of long-term monitoring in healthy individuals, identifying the window for early intervention and the populations to be prioritized for screening is critical. This study aimed to identify intervention windows for osteoporosis and to determine potential high-risk subtypes within the healthy population. Based on proteomic data from 41,408 healthy adults, we conducted the DE-SWAN method to identify change peaks in plasma protein during the pre-diagnostic osteoporosis phase, and employed finite Gaussian mixture model-based clustering to delineate high-risk subtypes of osteoporosis. We identified 122 protein biomarkers significantly associated with osteoporosis risk throughout the follow-up period. Importantly, we identified two critical peaks occurring approximately 10 and 6 years before diagnosis, with the former enriched in immune-related pathways and the latter prominently involving responses to retinoic acid and glucocorticoids. Furthermore, one high-risk subtype for osteoporosis was identified in both males and females, termed the Frailty and Obesity Subtype. This subtype is characterized by a high degree of frailty and obesity, accompanied by a significantly elevated risk of both osteoporosis and fractures. Finally, we developed a predictive model comprising 10 proteins for identifying high-risk subtypes of osteoporosis, which demonstrated better performance than the traditional risk factor model (AUC: 0.743 vs. 0.680). Our findings demonstrate that proteomic profiling can reveal early molecular changes and identify high-risk subtypes years before clinical onset, providing a foundation for screening and precision prevention of osteoporosis.

Proteomics↗

PubMed2006

Dopamine, prediction error and associative learning: a model-based account.

The notion of prediction error has established itself at the heart of formal models of animal learning and current hypotheses of dopamine function. Several interpretations of prediction error have been offered, including the model-free reinforcement learning method known as temporal difference learning (TD), and the important Rescorla-Wagner (RW) learning rule. Here, we present a model-based adaptation of these ideas that provides a good account of empirical data pertaining to dopamine neuron firing patterns and associative learning paradigms such as latent inhibition, Kamin blocking and overshadowing. Our departure from model-free reinforcement learning also offers: 1) a parsimonious distinction between tonic and phasic dopamine functions; 2) a potential generalization of the role of phasic dopamine from valence-dependent "reward" processing to valence-independent "salience" processing; 3) an explanation for the selectivity of certain dopamine manipulations on motivation for distal rewards; and 4) a plausible link between formal notions of prediction error and accounts of disturbances of thought in schizophrenia (in which dopamine dysfunction is strongly implicated). The model distinguishes itself from existing accounts by offering novel predictions pertaining to the firing of dopamine neurons in various untested behavioral scenarios.

Action Potentials↗

PubMed2026

Predicting 5-Year Mortality in Non-Small-Cell Lung Cancer Using the Korean Central Cancer Registry: Model Development and Validation Study.

BACKGROUND: Non-small-cell lung cancer (NSCLC) is one of the most common cancers and a leading cause of cancer-related mortality, making prognostic prediction clinically essential. Machine learning models are increasingly used to assess prognosis; however, developing systems that combine high discrimination with clear, clinically interpretable reasoning remains challenging. OBJECTIVE: This study aimed to develop deep learning models that predict 5-year mortality in NSCLC using data from the Korea Central Cancer Registry and quantify feature importance through permutation testing. METHODS: We identified 3144 patients diagnosed between 2014 and 2017 who had complete clinical data, pulmonary function test results, histological information, genomic data, and staging details. After preprocessing, the cohort was divided into stratified training, validation, and test sets in a 70%-15%-15% ratio. Five models were tuned using Hyperband across 10 predefined feature groups. The primary evaluation metric was the area under the receiver operating characteristic curve (AUC); additional metrics included accuracy, F1-score, precision, and recall. Groupwise permutation importance was calculated for each model, and the concordance of importance rankings was assessed using the Friedman test. RESULTS: All 5 models yielded comparable discrimination values on the test set (AUC=0.875-0.879). Model A was selected as the primary model and achieved an AUC of 0.879, an accuracy of 0.806, an F1-score of 0.824, and a Brier score of 0.142. Permuting the stage resulted in the largest decrease in AUC (0.217), followed by the pulmonary function test (0.016). Gene mutation had a modest overall impact but became more influential within the adenocarcinoma subset. The Friedman test showed no statistically significant differences in importance rankings across the models (P=.93). CONCLUSIONS: A grouped-input deep learning framework achieved discrimination comparable to a conventional Cox proportional hazards model using the same routine clinical variables for 5-year mortality prediction in NSCLC. Group-level permutation importance provided stable and reproducible insights into the clinical factors influencing risk, which may guide future model refinement and clinical decision-making.

Humans↗

PubMed2026

The Progress of Gout Prediction Models Based on Multi-source Data.

INTRODUCTION: Gout, a highly serious inflammatory disease that is caused by monosodium urate crystals, is becoming an increasingly significant health concern. Artificial Intelligence and multi-omics-based research have made significant gains for the early detection and prevention of gout based on diverse approaches. This review intends to summarize current advances in forecasting gout susceptibility and gout-related symptoms, evaluate the predictive efficacy of different features, and ascertain which clinical and omics characteristics are most effective in these prediction models. METHODS: We explored the PubMed database after 2010 using keywords such as "gout", "predictive model", "risk prediction", and "machine learning", and confined our search to Englishlanguage articles. The original peer-reviewed research articles that developed gout models were selected. Research that was not original or lacked internal validation was excluded. RESULTS: Clinical features, genomics, microbiomics, radiomics, and metabolomics have been utilized to construct models related to gout and have demonstrated excellent predictive performance. Multisource data prediction models usually exhibit better effectiveness. DISCUSSION: Gout-oriented models performed excellently in predictive performance but present limitations in certain clinical and omics domains. However, if they are to affect actual patient care, they must overcome some external confirmation roadblocks and the fiscal and practical implications they will face ahead of time. CONCLUSION: This review indicates that clinical and multi-omics models of gout are significant instruments for clinical decision-making. The models constructed in these studies may be crucial for the treatment of gout and its practical benefits.

Gout↗

PubMed2025

ESMpHLA: Evolutionary Scale Model-Based Deep Learning Prediction of HLA Class I Binding Peptides.

The recognition of endogenous peptides by HLA class I plays a crucial role in CD8+ T cell immune responses and human adaptive cell immune. Thus, the prediction of HLA class I-peptide binding affinities is always the core issue for the research of immune recognition and vaccine development. In this study, an evolutionary scale model (ESM) combined with parallel CNN blocks and a cross attention mechanism was used to construct a novel ESMpHLA model for predicting HLA class I binding peptides. Based on the 91,560 binding peptides of 41 HLA-A alleles, 56,731 of 50 HLA-B alleles and 2444 of 10 HLA-C alleles, the ESMpHLA model was successfully established and achieved satisfying prediction performances with the overall accuracy and AUC values of 0.874 and 0.938 for the test dataset. The results indicate that the ESMpHLA model performs well in dealing with different HLA class I 2-field alleles as well as the peptides with different lengths. Then, the generalisation ability of the ESMpHLA model was validated by an independent test dataset compiled from recent IEDB weekly benchmark datasets. The results showed that the ESMpHLA model achieved the highest ROC-AUC and PR-AUC values when compared with the latest BVMHC, CapsNet-MHC, STMHCpan and BVLSTM models. In addition, two ensemble models were also established by integrating the above 5 deep learning models using soft-voting and hard-voting strategies.

Humans↗

PubMed2026

CLASPP: A unified model for predicting post-translational modifications.

Post-Translational Modifications (PTMs) are a fundamental mechanism for regulating cellular pathways and increasing the functional diversity of the proteome. Accurately predicting the PTM types that are likely to occur at a given site in the primary sequence is a key challenge in functional proteomics. Existing PTM prediction models predominantly focus on either single PTM types or employ ensemble methods that combine multiple models to predict different PTM types. This fragmentation is largely driven by the vast imbalance in data availability across PTM types, making it difficult to predict multiple PTM types with a single model. To address this limitation, we present the Contrastively Learned Attention-based Stratified PTM Predictor (CLASPP), a unified PTM prediction model. CLASPP addresses imbalance challenges by leveraging unsupervised clustering-based undersampling and a novel contrastive learning framework tailored to PTM data. Additionally, our hierarchical data organization and curation are shown to improve CLASPP's performance by balancing the representation of individual PTM types and provides a standardized dataset to train and validate future model designs. Drawing inspiration from advancements in image and natural language processing, the CLASPP model employs a multi-stage training strategy and a high-quality, curated training dataset to improve PTM prediction performance. To uncover what is learned during the contrastive learning stage, the CLASPP model is shown to distinguish known protein kinase substrate specificity profiles as a form of explainability. Finally, we evaluate the application of CLASPP in predicting PTMs in different model organisms and experimentally validated ubiquitination sites in the understudied DCLK3 kinase. Overall, CLASPP represents a unified model for PTM prediction that addresses key bottlenecks in data imbalance and offers new strategies for biological data curation, thereby improving PTM-type prediction performance across diverse organisms.

Protein Processing, Post-Translational↗

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

Systematic review of machine learning approaches for predicting sickle cell crisis and mortality risk at the climate-health nexus.

BACKGROUND: Sickle cell anemia (SCA) is a severe genetic blood disorder characterized by recurrent vaso-occlusive crises and increased mortality, with the greatest burden occurring in low- and middle-income countries. Climatic and environmental conditions, including temperature variability, humidity, rainfall, air pollution, and seasonal changes, have been associated with disease exacerbation. However, the extent to which these factors have been incorporated into predictive models remains unclear. This study systematically reviews the application of machine learning (ML) models for predicting SCA crises and mortality in relation to climate and environmental factors. METHODOLOGY: The PRISMA guidelines were used, and 34 peer-reviewed studies published between 2005 and 2026 were analyzed to identify the climate variables, ML approaches employed, and predictive performance. The reviewed studies applied a range of ML techniques, including artificial neural networks, random forests, support vector machines, decision trees, logistic regression, and deep learning models. Temperature, humidity, rainfall, wind speed, air quality indicators, and seasonal patterns were the most frequently examined environmental variables. RESULTS: The findings indicate that most existing models rely predominantly on clinical and demographic data, with limited integration of climate information and inadequate representation of high-burden regions, especially Sub-Saharan Africa. Studies incorporating environmental variables reported improved predictive performance and highlighted the potential of climate-informed early warning systems for SCA management. CONCLUSION: The review recommends development of interdisciplinary, climate-aware ML frameworks, expansion of longitudinal environmental datasets, and increased research in underrepresented regions to support climate-resilient and patient-centered SCA care.

Humans↗

PubMed2006

Support vector machine learning model for the prediction of sentinel node status in patients with cutaneous melanoma.

BACKGROUND: Currently, approximately 80% of melanoma patients undergoing sentinel node biopsy (SNB) have negative sentinel lymph nodes (SLNs), and no prediction system is reliable enough to be implemented in the clinical setting to reduce the number of SNB procedures. In this study, the predictive power of support vector machine (SVM)-based statistical analysis was tested. METHODS: The clinical records of 246 patients who underwent SNB at our institution were used for this analysis. The following clinicopathologic variables were considered: the patient's age and sex and the tumor's histological subtype, Breslow thickness, Clark level, ulceration, mitotic index, lymphocyte infiltration, regression, angiolymphatic invasion, microsatellitosis, and growth phase. The results of SVM-based prediction of SLN status were compared with those achieved with logistic regression. RESULTS: The SLN positivity rate was 22% (52 of 234). When the accuracy was > or = 80%, the negative predictive value, positive predictive value, specificity, and sensitivity were 98%, 54%, 94%, and 77% and 82%, 41%, 69%, and 93% by using SVM and logistic regression, respectively. Moreover, SVM and logistic regression were associated with a diagnostic error and an SNB percentage reduction of (1) 1% and 60% and (2) 15% and 73%, respectively. CONCLUSIONS: The results from this pilot study suggest that SVM-based prediction of SLN status might be evaluated as a prognostic method to avoid the SNB procedure in 60% of patients currently eligible, with a very low error rate. If validated in larger series, this strategy would lead to obvious advantages in terms of both patient quality of life and costs for the health care system.

Artificial Intelligence↗

PubMed2025

Development and evaluation of a machine learning model for osteoporosis risk prediction in Korean women.

BACKGROUND: The aim of this study was to develop a machine learning (ML) model for classifying osteoporosis in Korean women based on a large-scale population cohort study. This study also aimed to assess ML model performance compared with traditional osteoporosis screening tools. Furthermore, this study aimed to examine the factors influencing the risk of osteoporosis through variable importance. METHODS: Data was collected from 4199 women aged 40-69 years in the baseline survey of the Ansan and Ansung cohort of the Korean Genome and Epidemiology Study. Osteoporosis was set as the dependent variable to develop ML classification models. Independent variables included 122 factors related to osteoporosis risk, such as socio-demographic characteristics, anthropometric parameters, lifestyle factors, reproductive factors, nutrient intakes, diet quality indices, medical history, medication history, family history, biochemical parameters, and genetic factors. The six classification models were developed using ML techniques, including decision tree, random forest, multilayer perceptron, support vector machine, light gradient boosting machine, and extreme gradient boosting (XGBoost). The six ML classification models were compared with two traditional osteoporosis screening tools, including the osteoporosis risk assessment instrument (ORAI) and the osteoporosis self-assessment tool (OST). The ML model performances were evaluated and compared using the confusion matrix and area under the curve (AUC) metrics. Variable importance was assessed using the XGBoost technique to investigate osteoporosis risk factors. RESULTS: The XGBoost model showed the highest performance out of the six ML classification models, with an accuracy of 0.705, precision of 0.664, recall of 0.830, and F1 score of 0.738. Moreover, the XGBoost model showed a higher performance on AUC than ORAI and OST. Variable importance scores were identified for 69 out of the 122 variables associated with osteoporosis risk factors. Age at menopause ranked first in variable importance. Variables of arthritis, physical activities, hypertension, education level, income level; alcohol intake, potassium intake, homeostatic model assessment for insulin resistance; energy intake, vitamin C intake, gout; and dietary inflammatory index ranked in the top 20 out of the 69 variables, using the XGBoost technique. CONCLUSIONS: This study found that an XGBoost model can be utilized to classify osteoporosis in Korean women. Age at menopause is a significant factor in osteoporosis risk, followed by arthritis, physical activities, hypertension, and education level.

Humans↗

PubMed2003

Comparing statistical and machine learning classifiers: alternatives for predictive modeling in human factors research.

Multivariate classification models play an increasingly important role in human factors research. In the past, these models have been based primarily on discriminant analysis and logistic regression. Models developed from machine learning research offer the human factors professional a viable alternative to these traditional statistical classification methods. To illustrate this point, two machine learning approaches--genetic programming and decision tree induction--were used to construct classification models designed to predict whether or not a student truck driver would pass his or her commercial driver license (CDL) examination. The models were developed and validated using the curriculum scores and CDL exam performances of 37 student truck drivers who had completed a 320-hr driver training course. Results indicated that the machine learning classification models were superior to discriminant analysis and logistic regression in terms of predictive accuracy. Actual or potential applications of this research include the creation of models that more accurately predict human performance outcomes.

Adolescent↗

PubMed2026

An encyclopedia of human enhancer-gene regulatory interactions.

Identifying transcriptional enhancers and their target genes is essential for understanding gene regulation and the effect of human genetic variation on disease1-6. Here we create and evaluate a resource of more than 92 million enhancer-gene regulatory interactions across 1,458 biosamples covering 369 cell types and tissues, by integrating predictive models, chromatin states, three-dimensional contacts and large-scale genetic perturbations generated by the ENCODE Consortium7. We first create a systematic benchmarking pipeline to compare predictive models, assembling a dataset of 10,356 element-gene pairs measured in CRISPR perturbation experiments, more than 30,000 fine-mapped expression quantitative trait loci and 569 fine-mapped genome-wide association study (GWAS) variants linked to a probable causal gene. Using this framework, we develop ENCODE-rE2G, a predictive model achieving state-of-the-art performance across several prediction tasks, demonstrating that iterative perturbations and supervised machine learning can build increasingly accurate predictive models of enhancer regulation. Using ENCODE-rE2G, we build an encyclopedia of enhancer-gene regulatory interactions in the human genome, revealing global properties of enhancer networks, identifying differences in regulatory complexity across genes and improving analyses linking noncoding variants to target genes and cell types for common complex diseases. By interpreting the model, we find that beyond enhancer activity and three-dimensional enhancer-promoter contacts, additional features that guide enhancer-promoter communication include promoter class and enhancer-enhancer synergy. These genome-wide maps of enhancer-gene regulatory interactions, benchmarking software, predictive models and insights about enhancer function provide a valuable resource for future studies of gene regulation and human genetics.

Humans↗

PubMed2026

Clinical Variable-Based Machine Learning for Predicting Early mCRPC Using Exclusively Clinical Variables: Development and Multicenter External Validation.

BACKGROUND AND OBJECTIVE: Metastatic hormone-sensitive prostate cancer (mHSPC) exhibits heterogeneous progression patterns, with early progression to metastatic castration-resistant prostate cancer (mCRPC) within 12 months indicating aggressive tumor biology and poor prognosis. Current risk stratification tools (CHAARTED, LATITUDE) offer limited individualized prediction. Machine learning approaches are increasingly applied to predict prostate cancer progression, but most models show modest performance (AUC 0.68-0.72), limited external validation, or require genomic variables unavailable in routine practice. This study aimed to develop and externally validate a novel RINH algorithm for predicting early mCRPC progression (≤ 12 months) using exclusively clinical variables, positioning it as a superior alternative to conventional ML classifiers. METHODS: This multicenter study enrolled 412 patients with de novo mHSPC from seven Spanish academic centers using mixed retrospective-prospective data collection. Twenty clinical variables were recorded, including demographics, PSA, ISUP grade, metastatic localization, CHAARTED/LATITUDE classifications, and treatment modalities. Following RINH-based outlier exclusion (55 patients), 357 patients (29 with early progression, 8.1%) were used to train six ML algorithms: RINH, Logistic Regression, Linear Discriminant, Support Vector Machine, Random Forest, and Subspace Discriminant. A two-tiered validation strategy integrated stratified fivefold cross-validation across all centers and formal external validation using center 1 (n = 121, 19 events) for training and centers 2-7 (n = 207, 10 events) for independent testing. Performance metrics included AUC, sensitivity, specificity, accuracy, and F1-score. KEY FINDINGS AND LIMITATIONS: Artificial intelligence and machine learning (ML) are transforming oncology, promising personalized risk stratification beyond traditional clinical criteria. In metastatic hormone-sensitive prostate cancer (mHSPC), early progression to castration resistance (mCRPC) within 12 months signals aggressive biology and poor prognosis, yet current tools (CHAARTED, LATITUDE) offer limited individualized prediction. Multiple ML models have been proposed with variable success: most achieve modest performance (AUC 0.68-0.72), lack robust external validation, or rely on genomic variables inaccessible in routine practice. We propose a novel approach using the Rivality Index Neighborhood (RINH) algorithm, demonstrating superior predictive capacity in an initial multicenter validation with exclusively clinical variables. This study provides rigorous multicenter external validation, advancing toward implementable precision oncology tools. CONCLUSIONS AND CLINICAL IMPLICATIONS: The RINH algorithm achieves superior predictive performance for early mCRPC progression using exclusively clinical variables, representing a significant advance toward implementable risk stratification. However, low reliability scores in external validation underscore that excellent performance metrics alone do not guarantee stability. Before clinical deployment, validation in substantially larger cohorts with higher progression events is essential. If validated, this model could enable personalized, risk-adapted therapeutic strategies, refining patient selection for treatment intensification or de-escalation.

Humans↗

PubMed2026

Machine learning for population-level risk prediction of future cholangiocarcinoma.

BACKGROUND: The poor prognosis of cholangiocarcinoma (CCA) is largely driven by rapid, asymptomatic disease progression, which usually results in a late diagnosis in the absence of established screening strategies. An early, cost-effective, and universally applicable risk assessment strategy would therefore be valuable. METHODS: We developed machine learning (ML) models on prospective, multimodal data from 487,495 UK Biobank (UKB) participants, of whom 649 developed CCA during follow-up. Data from England (80%) were utilised for ML development via five-fold cross-validation, and then all models were tested on withheld data from Scotland, Wales, and Newcastle (20%). Iterative ablation studies reduced inputs from >150 features across demographic data, lifestyle, health records, blood parameters, genomics, and metabolomics to models built on five and ten routinely available clinical parameters. These were externally validated in the Penn Medicine Biobank (PMBB; n = 2638; 28 CCA), All of Us Research Program (AOU; n = 330,433; 362 CCA), Japan Medical Data Centre Claims Database (JMDC; n = 8,425,522; 723 CCA) and TriNetX (n = 728,886; 1592 CCA). FINDINGS: We show that ML models integrating biliary-disease associated health records and Gamma glutamyltransferase can stratify risk of future CCA. Evaluation on the UKB test set as well as three independent cohorts revealed robust performance and generalisability across ethnicities. We achieved AUROCs of 0.71 [95% CI: 0.703-0.711], 0.77 [95% CI: 0.764-0.778 ], 0.796 [95% CI: 0.795-0.798] and 0.8 [95% CI: 0.794-0.805] for UKB, PMBB, AOU, and JMDC respectively, with respective AUPRCs of 0.014 [95% CI: 0.009-0.018], 0.042 [95% CI: 0.037-0.048], 0.038 [95% CI: 0.033-0.042] and 0.001 [95% CI: 0.001-0.001]. In AOU, application of the Youden J-optimised threshold yielded a number needed to screen of 79. Separate models for intra- and extrahepatic CCA did not improve performance. In line with the pathophysiology, performance declined for longer intervals between assessment and event. A group-level analysis in the TriNetX cohort revealed hazard ratios of up to 82.5 [95% CI: 26.4-257.96]. We provide extensive interpretability results and release all source codes used to develop the presented models. INTERPRETATION: We provide a comprehensive framework for early CCA risk stratification in the general population, identifying key predictors, and demonstrating the potential of data-driven models in personalised screening for hepatobiliary cancer. FUNDING: German Cancer Aid (grant #70115730), Junior Principal Investigator Fellowship programme of RWTH Aachen Excellence strategy.

Humans↗

PubMed1992

Predictive and diagnostic learning within causal models: asymmetries in cue competition.

Several researchers have recently claimed that higher order types of learning, such as categorization and causal induction, can be reduced to lower order associative learning. These claims are based in part on reports of cue competition in higher order learning, apparently analogous to blocking in classical conditioning. Three experiments are reported in which subjects had to learn to respond on the basis of cues that were defined either as possible causes of a common effect (predictive learning) or as possible effects of a common cause (diagnostic learning). The results indicate that diagnostic and predictive reasoning, far from being identical as predicted by associationistic models, are not even symmetrical. Although cue competition occurs among multiple possible causes during predictive learning, multiple possible effects need not compete during diagnostic learning. The results favor a causal-model theory.

Adult↗

PubMed2005

The infant as a prelinguistic model for language learning impairments: predicting from event-related potentials to behavior.

Associations between efficient processing of brief, rapidly presented, successive stimuli and language learning impairments (LLI) in older children and adults have been well documented. In this paper we examine the role that impaired rapid auditory processing (RAP) might play during early language acquisition. Using behavioral measures we have demonstrated that RAP abilities in infancy are critically linked to later language abilities for both non-speech and speech stimuli. Variance in infant RAP thresholds reliably predict language outcome at 3 years-of-age for infants at risk for LLI and control infants. We present data here describing patterns of electrocortical (EEG/ERP) activation at 6 month-of-age to the same non-verbal stimuli used in our behavioral studies. Well-defined differences were seen between infants from families with a history of LLI (FH+) and FH- controls in the amplitude of the mismatch response (MMR) as well as the latency of the N250 component in the 70 ms ISI condition only. Smaller mismatch responses and delayed onsets of the N250 component were seen in the FH+ group. The latency differences in the N250 component, but not the MMR amplitude variation, were significantly related to 24-month language outcome. Such converging tasks provide the opportunity to examine early precursors of LLI and allow the opportunity for earlier identification and intervention.

Brain Mapping↗

PubMed2004

Comparison of machine learning techniques with classical statistical models in predicting health outcomes.

Several machine learning techniques (multilayer and single layer perceptron, logistic regression, least square linear separation and support vector machines) are applied to calculate the risk of death from two biomedical data sets, one from patient care records, and another from a population survey. Each dataset contained multiple sources of information: history of related symptoms and other illnesses, physical examination findings, laboratory tests, medications (patient records dataset), health attitudes, and disabilities in activities of daily living (survey dataset). Each technique showed very good mortality prediction in the acute patients data sample (AUC up to 0.89) and fair prediction accuracy for six year mortality (AUC from 0.70 to 0.76) in individuals from epidemiological database surveys. The results suggest that the nature of data is of primary importance rather than the learning technique. However, the consistently superior performance of the artificial neural network (multi-layer perceptron) indicates that nonlinear relationships (which cannot be discerned by linear separation techniques) can provide additional improvement in correctly predicting health outcomes.

Aged↗