Search

Step-up therapy is a patient management approach that aims to balance the efficacy, costs and risks posed by different lines of medications. While the initiation of first line medications is a straightforward decision, stepping-up a patient to the next treatment line is often more challenging and difficult to predict. By identifying patients who are likely to move to the next line of therapy, prediction models could be used to help healthcare organizations with resource planning and chronic disease management.To compared supervised learning versus semi-supervised learning to predict which rheumatoid arthritis patients will move from the first line of therapy (i.e., conventional synthetic disease-modifying antirheumatic drugs) to the next line of therapy (i.e., disease-modifying antirheumatic drugs or targeted synthetic disease-modifying antirheumatic drugs) within one year.Five groups of features were extracted from an administrative claims database: demographics, medications, diagnoses, provider characteristics, and procedures. Then, a variety of supervised and semi-supervised learning methods were implemented to identify the most optimal method of each approach and assess the contribution of each feature group. Finally, error analysis was conducted to understand the behavior of misclassified patients.XGBoost yielded the highest F-measure (42%) among the supervised approaches and one-class support vector machine achieved the highest F-measure (65%) among the semi-supervised approaches. The semi-supervised approach had significantly higher F-measure (65% vs. 42%; p 0.01), precision (51% vs. 33%; p 0.01), and recall (89% vs. 59%; p 0.01) than the supervised approach. Excluding demographic, drug, diagnosis, provider, and procedure features reduced theF-measure from 65% to 61%, 57%, 54%, 51% and 49% respectively (p 0.01). The error analysis showed that a substantial portion of false positive patients will change their line of therapy shortly after the prediction period.This study showed that supervised learning approaches are not an optimal option for a difficult clinical decision regarding step-up therapy. More specifically, negative class labels in step-up therapy data are not a robust ground truth, because the costs and risks associated with higher line of therapy impact objective decision making of patients and providers. The proposed semi-supervised learning approach can be applied to other step-up therapy applications.

Objective: Machine learning algorithms are now widely used in predicting acute events for clinical applications. While most of such prediction applications are developed to predict the risk of a particular acute event at one hospital, few efforts have been made in extending the developed solutions to other events or to different hospitals. We provide a scalable solution to extend the process of clinical risk prediction model development of multiple diseases and their deployment in different Electronic Health Records (EHR) systems. Materials and Methods: We defined a generic process for clinical risk prediction model development. A calibration tool has been created to automate the model generation process. We applied the model calibration process at four hospitals, and generated risk prediction models for delirium, sepsis and acute kidney injury (AKI) respectively at each of these hospitals. Results: The delirium risk prediction models achieved area under the receiver-operating characteristic curve (AUROC) ranging from 0.82 to 0.95 over different stages of a hospital stay on the test datasets of the four hospitals. The sepsis models achieved AUROC ranging from 0.88 to 0.95, and the AKI models achieved AUROC ranging from 0.85 to 0.92. Discussion: The scalability discussed in this paper is based on building common data representations (syntactic interoperability) between EHRs stored in different hospitals. Semantic interoperability, a more challenging requirement that different EHRs share the same meaning of data, e.g. a same lab coding system, is not mandated with our approach. Conclusions: Our study describes a method to develop and deploy clinical risk prediction models in a scalable way. We demonstrate its feasibility by developing risk prediction models for three diseases across four hospitals., Comment: Preprint of Journal of Biomedical Informatics

Model calibration, critical to the success and safety of clinical prediction models, deteriorates over time in response to the dynamic nature of clinical environments. To support informed, data-driven model updating strategies, we present and evaluate a calibration drift detection system. Methods are developed for maintaining dynamic calibration curves with optimized online stochastic gradient descent and for detecting increasing miscalibration with adaptive sliding windows. These methods are generalizable to support diverse prediction models developed using a variety of learning algorithms and customizable to address the unique needs of clinical use cases. In both simulation and case studies, our system accurately detected calibration drift. When drift is detected, our system further provides actionable alerts by including information on a window of recent data that may be appropriate for model updating. Simulations showed these windows were primarily composed of data accruing after drift onset, supporting the potential utility of the windows for model updating. By promoting model updating as calibration deteriorates rather than on pre-determined schedules, implementations of our drift detection system may minimize interim periods of insufficient model accuracy and focus analytic resources on those models most in need of attention.

Graphical abstract, We present EPIsembleVis, a web-based comparative visual analysis tool for evaluating the consistency of multiple COVID-19 prediction models. Our approach analyzes a collection of COVID-19 predictions from different epidemiological models as an ensemble and utilizes two metrics to quantify model performance. These metrics include (a) prediction uncertainty (represented as the dispersion of predictions in each ensemble) and (b) prediction error (calculated by comparing individual model predictions with the recorded data). Through an interactive visual interface, our approach provides a data-driven workflow for (a) selecting and constructing the COVID-19 model prediction ensemble based on the spatiotemporal overlap of available predictions of multiple epidemiological models, (b) quantifying the model performance using both the uncertainty of each model prediction ensemble, and the error of each ensemble member that represents individual model predictions, and (c) visualizing the spatiotemporal variability in the projection performance of individual models using a suite of novel ensemble visualization techniques, such as the data availability map, a spatiotemporal textured-tile calendar, multivariate rose chart, and time-series leaflet glyph. We demonstrate the capability of our ensemble visual interface through a case study that investigates the performance of weekly COVID-19 predictions, which are provided through the COVID-19 Forecast Hub UMass-Amherst Influenza Forecasting Center of Excellence [47] for the United States and United States Territories. The EPIsembleVis tool is implemented using open-source web technologies and adaptive system design, rendering it interoperable with Elasticsearch and Kibana for automatically ingesting COVID-19 predictions from online repositories, and it is generalizable for analyzing worldwide projections from more epidemiological models.

Display Omitted Fold change cut-offs might not extract all vital biological signals from microarrays.Microarray analysis method using Singular Value Decomposition was proposed.A novel model for gene prediction in disease progression using SVD was proposed.The model is based on a simple normal-distribution-based random number generator.Model hypothesis: High disease samples can be extrapolated from low disease. Being a high throughput technique, enormous amounts of microarray data has been generated and there arises a need for more efficient techniques of analysis, in terms of speed and accuracy. Finding the differentially expressed genes based on just fold change and p-value might not extract all the vital biological signals that occur at a lower gene expression level. Besides this, numerous mathematical models have been generated to predict the clinical outcome from microarray data, while very few, if not none, aim at predicting the vital genes that are important in a disease progression. Such models help a basic researcher narrow down and concentrate on a promising set of genes which leads to the discovery of gene-based therapies. In this article, as a first objective, we have used the lesser known and used Singular Value Decomposition (SVD) technique to build a microarray data analysis tool that works with gene expression patterns and intrinsic structure of the data in an unsupervised manner. We have re-analysed a microarray data over the clinical course of Septic shock from Cazalis et al. (2014) and have shown that our proposed analysis provides additional information compared to the conventional method. As a second objective, we developed a novel mathematical model that predicts a set of vital genes in the disease progression that works by generating samples in the continuum between health and disease, using a simple normal-distribution-based random number generator. We also verify that most of the predicted genes are indeed related to septic shock.

Background and purpose Low adherence to medication in chronic disease patients leads to increased morbidity, mortality, and healthcare costs. The widespread adoption of electronic prescription and dispensation records allows a more comprehensive overview of medication utilization. In combination with electronic health records (EHR), such data provides new opportunities for identifying patients at risk of nonadherence and provide more targeted and effective interventions. The purpose of this article is to study the predictability of medication adherence for a cohort of hypertensive patients, focusing on healthcare utilization factors under various predictive scenarios. Furthermore, we discover common proportion of days covered patterns (PDC-patterns) for patients with index prescriptions and simulate medication-taking behaviours that might explain observed patterns. Procedures We predict refill adherence focusing on factors of healthcare utilization, such as visits, prescription information and demographics of patient and prescriber. We train models with machine learning algorithms, using four different data splits: stratified random, patient, temporal forward prediction with and without index patients. We extract frequent, two-year long PDC-patterns using K-means clustering and investigate five simple models of medication-taking that can generate such PDC-patterns. Findings Model performance varies between data splits (AUC test set: 0.77–0.89). Including historical information increases the performance slightly in most cases (approx. 1–2% absolute AUC uplift). Models show low predictive performance (AUC test set: 0.56–0.66) on index-prescriptions and patients with sudden drops in PDC (Recall: 0.58–0.63). We find 21 distinct two-year PDC-patterns, ranging from good adherence to intermittent gaps and early discontinuation in the first or second year. Simulations show that observed PDC-patterns can only be explained by specific medication consumption behaviours. Conclusions Prediction models developed using EHR exhibit bias towards patients with high healthcare utilization. Even though actual medication-taking is not observable, consumption patterns may not be as arbitrary, provided that medication refilling and consumption is linked.

Developing a prognostic model for biomedical applications typically requires mapping an individual's set of covariates to a measure of the risk that he or she may experience the event to be predicted. Many scenarios, however, especially those involving adverse pathological outcomes, are better described by explicitly accounting for the timing of these events, as well as their probability. As a result, in these cases, traditional classification or ranking metrics may be inadequate to inform model evaluation or selection. To address this limitation, it is common practice to reframe the problem in the context of survival analysis, and resort, instead, to the concordance index (C-index), which summarises how well a predicted risk score describes an observed sequence of events. A practically meaningful interpretation of the C-index, however, may present several difficulties and pitfalls. Specifically, we identify two main issues: i) the C-index remains implicitly, and subtly, dependent on time, and ii) its relationship with the number of subjects whose risk was incorrectly predicted is not straightforward. Failure to consider these two aspects may introduce undesirable and unwanted biases in the evaluation process, and even result in the selection of a suboptimal model. Hence, here, we discuss ways to obtain a meaningful interpretation in spite of these difficulties. Aiming to assist experimenters regardless of their familiarity with the C-index, we start from an introductory-level presentation of its most popular estimator, highlighting the latter's temporal dependency, and suggesting how it might be correctly used to inform model selection. We also address the nonlinearity of the C-index with respect to the number of correct risk predictions, elaborating a simplified framework that may enable an easier interpretation and quantification of C-index improvements or deteriorations.

Outcomes' prediction in Electronic Health Records (EHR) and specifically in Critical Care is increasingly attracting more exploration and research. In this study, we used clinical data from the Intensive Care Unit (ICU), focusing on ICU acquired sepsis. Looking at the current literature, several evaluation approaches are reported, inspired by epidemiological designs, in which some do not always reflect real-life application's conditions. This problem seems relevant generally to outcomes' prediction in longitudinal EHR data, or generally longitudinal data, while in this study we focused on ICU data. Unlike in most previous studies that investigated all sepsis admissions, we focused specifically on ICU-Acquired Sepsis. Due to the sparse nature of the longitudinal data, we employed the use of Temporal Abstraction and Time Interval-Related Patterns discovery, which are further used as classification features. Two experiments were designed using three different outcomes prediction study designs from the literature, implementing various levels of real-life conditions to evaluate the prediction models. The first experiment focused on predicting whether a patient would suffer from ICU-acquired sepsis and when during her admission, given a sliding observation time window, and the comparison of the three study designs behavior. The second experiment focused only on predicting whether the patient will suffer from ICU-acquired sepsis, based on data taken relatively to his admission start time. Our results show that using Temporal Discretization for Classification (TD4C) led to better performance than using the Equal-Width Discretization, Knowledge-Based, or SAX. Also, using two states abstraction was better than three or four. Using the default Binary TIRP representation method performed better than Mean Duration, Horizontal Support, and horizontally normalized horizontal support. Using XGBoost as a classifier performed better than Logistic Regression, Neural Net, or Random Forest. Additionally, it is demonstrated why the use of case-crossover-control is most appropriate for real life application conditions evaluation, unlike other incomplete designs that may even result in "better performance".

Object Our purpose was to develop a new machine-learning approach (a virtual health check-up) toward identification of those at high risk of hyperuricemia. Applying the system to general health check-ups is expected to reduce medical costs compared with administering an additional test. Methods Data were collected during annual health check-ups performed in Japan between 2011 and 2013 (inclusive). We prepared training and test datasets from the health check-up data to build prediction models; these were composed of 43,524 and 17,789 persons, respectively. Gradient-boosting decision tree (GBDT), random forest (RF), and logistic regression (LR) approaches were trained using the training dataset and were then used to predict hyperuricemia in the test dataset. Undersampling was applied to build the prediction models to deal with the imbalanced class dataset. Results The results showed that the RF and GBDT approaches afforded the best performances in terms of sensitivity and specificity, respectively. The area under the curve (AUC) values of the models, which reflected the total discriminative ability of the classification, were 0.796 [95% confidence interval (CI): 0.766–0.825] for the GBDT, 0.784 [95% CI: 0.752–0.815] for the RF, and 0.785 [95% CI: 0.752–0.819] for the LR approaches. No significant differences were observed between pairs of each approach. Small changes occurred in the AUCs after applying undersampling to build the models. Conclusions We developed a virtual health check-up that predicted the development of hyperuricemia using machine-learning methods. The GBDT, RF, and LR methods had similar predictive capability. Undersampling did not remarkably improve predictive power.

INTEGRATE - Closing the loop between clinical research and care in oncology through innovative applications and by improving the information and knowledge flow. The four applications described in this paper leverage our sematic interoperability solution that integrates clinical research and care infrastructures. While CPR and DECIMA impact the care domain and the execution phase of clinical research, Nona and AP address the needs of the clinical researcher, and the definition and design phases of clinical research. The loop is closed by delivering the new data from clinical care and from trials back into the INTEGRATE environment. This solution also helps closing the knowledge loop: new hypotheses lead to new clinical trials and validated trial results become new treatment options.Display Omitted Integration of multi-scale biomedical data with post-genomic clinical trials.Ethical and Security Framework for post-genomic clinical trials.Advanced Patient Screening for clinical trials.Collaborative remote environment for pathology review.Cohort selection, qualitative analysis and prediction models for large multi-level patient datasets. The objective of the INTEGRATE project (http://www.fp7-integrate.eu/) that has recently concluded successfully was the development of innovative biomedical applications focused on streamlining the execution of clinical research, on enabling multidisciplinary collaboration, on management and large-scale sharing of multi-level heterogeneous datasets, and on the development of new methodologies and of predictive multi-scale models in cancer.In this paper, we present the way the INTEGRATE consortium has approached important challenges such as the integration of multi-scale biomedical data in the context of post-genomic clinical trials, the development of predictive models and the implementation of tools to facilitate the efficient execution of postgenomic multi-centric clinical trials in breast cancer.Furthermore, we provide a number of key "lessons learned" during the process and give directions for further future research and development.

Display Omitted Claims are windowed on four time scales to construct regularly spaced time series.The four temporal models were utilized for forecasting future days in hospital.Non-yearly models outperformed the yearly model on the entire population.Superiority of the half-yearly model was pronounced in the senior sub-population. Health insurers maintain large databases containing information on medical services utilized by claimants, often spanning several healthcare services and providers. Proper use of these databases could facilitate better clinical and administrative decisions. In these data sets, there exists many unequally spaced events, such as hospital visits. However, data mining of temporal data and point processes is still a developing research area and extracting useful information from such data series is a challenging task. In this paper, we developed a time series data mining approach to predict the number of days in hospital in the coming year for individuals from a general insured population based on their insurance claim data. In the proposed method, the data were windowed at four different timescales (bi-monthly, quarterly, half-yearly and yearly) to construct regularly spaced time series features extracted from such events, resulting in four associated prediction models. A comparison of these models indicates models using a half-yearly windowing scheme delivers the best performance on all three populations (the whole population, a senior sub-population and a non-senior sub-population). The superiority of the half-yearly model was found to be particularly pronounced in the senior sub-population. A bagged decision tree approach was able to predict 'no hospitalization' versus 'at least one day in hospital' with a Matthews correlation coefficient (MCC) of 0.426. This was significantly better than the corresponding yearly model, which achieved 0.375 for this group of customers. Further reducing the length of the analysis windows to three or two months did not produce further improvements.

Widespread adoption of electronic health records (EHRs) has fueled the development of using machine learning to build prediction models for various clinical outcomes. However, this process is often constrained by having a relatively small number of patient records for training the model. We demonstrate that using patient representation schemes inspired from techniques in natural language processing can increase the accuracy of clinical prediction models by transferring information learned from the entire patient population to the task of training a specific model, where only a subset of the population is relevant. Such patient representation schemes enable a 3.5% mean improvement in AUROC on five prediction tasks compared to standard baselines, with the average improvement rising to 19% when only a small number of patient records are available for training the clinical prediction model.

Acute kidney injury (AKI) is a common clinical condition with high mortality and resource consumption. Early identification of high-risk patients to achieve an appropriate allocation of limited clinical resources and timely interventions is of significant importance, which has attracted substantial research to develop prediction models for AKI risk stratification. However, most available AKI prediction models have moderate performance and lack of interpretability, which limits their applicability in supporting care intervention. In this paper, a machine learning-based framework for AKI prediction and interpretation in critical care is presented. First, an ensemble model is developed to predict a patient’s risk of AKI within 72 h of admission to the intensive care units. Next, the model is interpreted both globally and locally. For the global interpretation, the important predictors are pinpointed and the detailed relationships between AKI risk and these predictors are illustrated. For the local interpretation, patient-specific analysis is presented to provide a visualized explanation for each individual prediction. Experimental results show that such a prediction and interpretation framework can lead to good prediction and interpretation performance, which has the potential to provide effective clinical decision support.

Objective We aim to investigate the hypothesis that using information about which variables are missing along with appropriate imputation improves the performance of severity of illness scoring systems used to predict critical patient outcomes. Study design and setting We quantify the impact of missing and imputed variables on the performance of prediction models used in the development of a sepsis-related severity of illness scoring system. Electronic health records (EHR) data were compiled from Christiana Care Health System (CCHS) on 119,968 adult patients hospitalized between July 2013 and December 2015. Two outcomes of interest were considered for prediction: (1) first transfer to intensive care unit (ICU) and (2) in-hospital mortality. Five different prediction models were employed. Indicators were utilized in these prediction models to identify when variables were missing and imputed. Results We observed statistically significant gains in prediction performance when moving from models that did not indicate missing information to those that did. Moreover, this increase was higher in models that use summary variables as predictors compared to those that use all variables. Conclusion When developing prediction models using longitudinal EHR data, researchers should explore the incorporation of indicators for missing variables along with appropriate imputation.

Background and objective Clinical prognosis prediction plays an important role in clinical research and practice. The construction of prediction models based on electronic health record data has recently become a research focus. Due to the lack of external validation, prediction models based on single-center, hospital-specific datasets may not perform well with datasets from other medical institutions. Therefore, research investigating prognosis prediction model construction based on a collaborative analysis of multi-center electronic health record data could increase the number and coverage of patients used for model training, enrich patient prognostic features and ultimately improve the accuracy and generalization of prognosis prediction. Materials and methods A web service for individual prognosis prediction based on multi-center clinical data collaboration without patient-level data sharing (POPCORN) was proposed. POPCORN focuses on solving key issues in multi-center collaborative research based on electronic health record systems; these issues include the standardization of clinical data expression, the preservation of patient privacy during model training and the effect of case mix variance on the prediction model construction and application. POPCORN is based on a multivariable meta-analysis and a Bayesian framework and can construct suitable prediction models for multiple clinical scenarios that can effectively adapt to complex clinical application environments. Results POPCORN was validated using a joint, multi-center collaborative research network between China and the United States with patients diagnosed with colorectal cancer. The performance of the models based on POPCORN was comparable to that of the standard prognosis prediction model; however, POPCORN did not expose raw patient data. The prediction models had similar AUC, but the BMA model had the lowest ECI across all prediction models, indicating that this model had better calibration performance than the other models, especially for patients in Chinese hospitals. Conclusions The POPCORN system can build prediction models that perform well in complex clinical application scenarios and can provide effective decision support for individual patient prognostic predictions.

Display Omitted A model that can learn supervised feature grouping and perform feature selection.Our method finds more stable features compared to other feature selection methods.The method results in better prediction performance compared to other methods. Emerging Electronic Medical Records (EMRs) have reformed the modern healthcare. These records have great potential to be used for building clinical prediction models. However, a problem in using them is their high dimensionality. Since a lot of information may not be relevant for prediction, the underlying complexity of the prediction models may not be high. A popular way to deal with this problem is to employ feature selection. Lasso and l 1 -norm based feature selection methods have shown promising results. But, in presence of correlated features, these methods select features that change considerably with small changes in data. This prevents clinicians to obtain a stable feature set, which is crucial for clinical decision making. Grouping correlated variables together can improve the stability of feature selection, however, such grouping is usually not known and needs to be estimated for optimal performance. Addressing this problem, we propose a new model that can simultaneously learn the grouping of correlated features and perform stable feature selection. We formulate the model as a constrained optimization problem and provide an efficient solution with guaranteed convergence. Our experiments with both synthetic and real-world datasets show that the proposed model is significantly more stable than Lasso and many existing state-of-the-art shrinkage and classification methods. We further show that in terms of prediction performance, the proposed method consistently outperforms Lasso and other baselines. Our model can be used for selecting stable risk factors for a variety of healthcare problems, so it can assist clinicians toward accurate decision making.

Display Omitted Survival prediction of HCC patients is modeled in a real and high complex scenario.Its complexity is due to a small dataset with heterogeneous and missing input data.We propose a new cluster-based oversampling method to overcome these drawbacks.Neural networks and logistic regression are applied in the classification stage.Experimental results show that our method outperforms previous existing approaches. Liver cancer is the sixth most frequently diagnosed cancer and, particularly, Hepatocellular Carcinoma (HCC) represents more than 90% of primary liver cancers. Clinicians assess each patients treatment on the basis of evidence-based medicine, which may not always apply to a specific patient, given the biological variability among individuals. Over the years, and for the particular case of Hepatocellular Carcinoma, some research studies have been developing strategies for assisting clinicians in decision making, using computational methods (e.g. machine learning techniques) to extract knowledge from the clinical data. However, these studies have some limitations that have not yet been addressed: some do not focus entirely on Hepatocellular Carcinoma patients, others have strict application boundaries, and none considers the heterogeneity between patients nor the presence of missing data, a common drawback in healthcare contexts. In this work, a real complex Hepatocellular Carcinoma database composed of heterogeneous clinical features is studied. We propose a new cluster-based oversampling approach robust to small and imbalanced datasets, which accounts for the heterogeneity of patients with Hepatocellular Carcinoma. The preprocessing procedures of this work are based on data imputation considering appropriate distance metrics for both heterogeneous and missing data (HEOM) and clustering studies to assess the underlying patient groups in the studied dataset (K-means). The final approach is applied in order to diminish the impact of underlying patient profiles with reduced sizes on survival prediction. It is based on K-means clustering and the SMOTE algorithm to build a representative dataset and use it as training example for different machine learning procedures (logistic regression and neural networks). The results are evaluated in terms of survival prediction and compared across baseline approaches that do not consider clustering and/or oversampling using the Friedman rank test. Our proposed methodology coupled with neural networks outperformed all others, suggesting an improvement over the classical approaches currently used in Hepatocellular Carcinoma prediction models.

Display Omitted A novel neural-inspired algorithm called Artificial Neural Cell System for classification (ANCSc) was developed.ANCSc algorithm is inspired by neural mechanisms like neurogenesis, neuroplasticity and apoptosis.Performance of ANCSc, when applied to a clinical risk prediction task, was juxtaposed with SVM and EDC-AIRS algorithms.ANCSc algorithm (statistically) outperforms both SVM and EDC-AIRS algorithms on this clinical prediction task.Key biomarkers identified by ANCSc algorithm are also found to be clinically significant. Clinical risk prediction - the estimation of the likelihood an individual is at risk of a disease - is a coveted and exigent clinical task, and a cornerstone to the recommendation of life saving management strategies. This is especially important for individuals at risk of cardiovascular disease (CVD) given the fact that it is the leading causes of death in many developed counties. To this end, we introduce a novel learning algorithm - a key factor that influences the performance of machine learning-based prediction models - and utilities it to develop CVD risk prediction tool. This novel neural-inspired algorithm, called the Artificial Neural Cell System for classification (ANCSc), is inspired by mechanisms that develop the brain and empowering it with capabilities such as information processing/storage and recall, decision making and initiating actions on external environment. Specifically, we exploit on 3 natural neural mechanisms responsible for developing and enriching the brain - namely neurogenesis, neuroplasticity via nurturing and apoptosis - when implementing ANCSc algorithm. Benchmark testing was conducted using the Honolulu Heart Program (HHP) dataset and results are juxtaposed with 2 other algorithms - i.e. Support Vector Machine (SVM) and Evolutionary Data-Conscious Artificial Immune Recognition System (EDC-AIRS). Empirical experiments indicate that ANCSc algorithm (statistically) outperforms both SVM and EDC-AIRS algorithms. Key clinical markers identified by ANCSc algorithm include risk factors related to diet/lifestyle, pulmonary function, personal/family/medical history, blood data, blood pressure, and electrocardiography. These clinical markers, in general, are also found to be clinically significant - providing a promising avenue for identifying potential cardiovascular risk factors to be evaluated in clinical trials.

Graphical abstract, Highlights • A LASSO based forecast model for endemic infectious diseases is proposed. • Predictions at 4 weeks achieve desirable accuracy. • Models predict outbreaks but may struggle to predict outbreak size., Introduction Accurate and timely prediction for endemic infectious diseases is vital for public health agencies to plan and carry out any control methods at an early stage of disease outbreaks. Climatic variables has been identified as important predictors in models for infectious disease forecasts. Various approaches have been proposed in the literature to produce accurate and timely predictions and potentially improve public health response. Methods We assessed how the machine learning LASSO method may be useful in providing useful forecasts for different pathogens in countries with different climates. Separate LASSO models were constructed for different disease/country/forecast window with different model complexity by including different sets of predictors to assess the importance of different predictors under various conditions. Results There was a more apparent cyclicity for both climatic variables and incidence in regions further away from the equator. For most diseases, predictions made beyond 4 weeks ahead were increasingly discrepant from the actual scenario. Prediction models were more accurate in capturing the outbreak but less sensitive to predict the outbreak size. In different situations, climatic variables have different levels of importance in prediction accuracy. Conclusions For LASSO models used for prediction, including different sets of predictors has varying effect in different situations. Short term predictions generally perform better than longer term predictions, suggesting public health agencies may need the capacity to respond at short-notice to early warnings.

Background The use of machine learning techniques is especially pertinent to the composite and challenging conditions of emergency departments (EDs). Repeat ED visits (i.e. revisits) are an example of potentially inappropriate utilization of resources that can be forecasted by these techniques. Objective To track the ED revisit risk over time using the hidden Markov model (HMM) as a major latent class model. Given the HMM states, we carried out forecasting of future ED revisits with various data mining models. Methods Information integrated from four distributed sources (e.g. electronic health records and health information exchange) was integrated into four HMMs which capture the relationships between an observed and a hidden progression that shift over time through a series of hidden states in an adult patient population. Results Assimilating a pre-analysis of the various patients by applying latent class models and directing them to well-known classifiers functioned well. The performance was significantly better than without utilizing pre-analysis of HMM for all prediction models (classifiers(. Conclusions These findings suggest that one prospective approach to advanced risk prediction is to leverage the longitudinal nature of health care data by exploiting patients’ between state variation.

The utility of a prediction model depends on its generalizability to patients drawn from different but related populations. We explored whether a semi-supervised learning model could improve the generalizability of colorectal cancer (CRC) risk prediction relative to supervised learning methods. Data on 113,141 patients diagnosed with nonmetastatic CRC from 2004 to 2012 were obtained from the Surveillance Epidemiology End Results registry for model development, and data on 1149 patients from the Second Affiliated Hospital, Zhejiang University School of Medicine, who were diagnosed between 2004 and 2011, were collected for generalizability testing. A clinical prediction model for CRC survival risk using a semi-supervised logistic regression method was developed and validated to investigate the model discrimination, calibration, generalizability, interpretability and clinical usefulness. Rigorous model performance comparisons with other supervised learning models were performed. The area under the curve of the logistic membership model revealed a large heterogeneity between the development cohort and validation cohort, which is typical of generalizability studies of prediction models. The discrimination was good for all models. Calibration was poor for supervised learning models, while the semi-supervised logistic regression model exhibited a good calibration on the validation cohort, which indicated good generalizability. Clinical usefulness analysis showed that semi-supervised logistic regression can lead to better clinical outcomes than supervised learning methods. These results increase our current understanding of the generalizability of different models and provide a reference for predictive model development for clinical decision-making.

Display Omitted We present a measurement to calculate the semantic similarity between diseases.We construct a heterogeneous network and analyze the CBG patterns on the network.A subgraph-based method is proposed for uncovering the miRNA-disease associations.The method achieved good performance and can effectively discover disease miRNAs. MicroRNAs (miRNAs) play a critical role by regulating their targets in post-transcriptional level. Identification of potential miRNA-disease associations will aid in deciphering the pathogenesis of human polygenic diseases. Several computational models have been developed to uncover novel miRNA-disease associations based on the predicted target genes. However, due to the insufficient number of experimentally validated miRNA-target interactions as well as the relatively high false-positive and false-negative rates of predicted target genes, it is still challenging for these prediction models to obtain remarkable performances. The purpose of this study is to prioritize miRNA candidates for diseases. We first construct a heterogeneous network, which consists of a disease similarity network, a miRNA functional similarity network and a known miRNA-disease association network. Then, an unbalanced bi-random walk-based algorithm on the heterogeneous network (BRWH) is adopted to discover potential associations by exploiting bipartite subgraphs. Based on 5-fold cross validation, the proposed network-based method achieves AUC values ranging from 0.782 to 0.907 for the 22 human diseases and an average AUC of almost 0.846. The experiments indicated that BRWH can achieve better performances compared with several popular methods. In addition, case studies of some common diseases further demonstrated the superior performance of our proposed method on prioritizing disease-related miRNA candidates.

BackgroundThe IOM report, Preventing Medication Errors, emphasizes the overall lack of knowledge of the incidence of adverse drug events (ADE). Operating rooms, emergency departments and intensive care units are known to have a higher incidence of ADE. Labor and delivery (L&D) is an emergency care unit that could have an increased risk of ADE, where reported rates remain low and under-reporting is suspected. Risk factor identification with electronic pattern recognition techniques could improve ADE detection rates.ObjectiveThe objective of the present study is to apply Synthetic Minority Over Sampling Technique (SMOTE) as an enhanced sampling method in a sparse dataset to generate prediction models to identify ADE in women admitted for labor and delivery based on patient risk factors and comorbidities.ResultsBy creating synthetic cases with the SMOTE algorithm and using a 10-fold cross-validation technique, we demonstrated improved performance of the Naïve Bayes and the decision tree algorithms. The true positive rate (TPR) of 0.32 in the raw dataset increased to 0.67 in the 800% over-sampled dataset.ConclusionEnhanced performance from classification algorithms can be attained with the use of synthetic minority class oversampling techniques in sparse clinical datasets. Predictive models created in this manner can be used to develop evidence based ADE monitoring systems.

This study predicted graft and recipient survival in kidney transplantation based on the USRDS dataset by regression models and artificial neural networks (ANNs). We examined single time-point models (logistic regression and single-output ANNs) versus multiple time-point models (Cox models and multiple-output ANNs). These models in general achieved good prediction discrimination (AUC up to 0.82) and model calibration. This study found that: (1) Single time-point and multiple time-point models can achieve comparable AUC, except for multiple-output ANNs, which may perform poorly when a large proportion of observations are censored, (2) Logistic regression is able to achieve comparable performance as ANNs if there are no strong interactions or non-linear relationships among the predictors and the outcomes, (3) Time-varying effects must be modeled explicitly in Cox models when predictors have significantly different effects on short-term versus long-term survival, and (4) Appropriate baseline survivor function should be specified for Cox models to achieve good model calibration, especially when clinical decision support is designed to provide exact predicted survival rates.

Display Omitted An approach called Contrast Pattern Aided Regression, CPXR(Log) with a new loss function is proposed.A series of heart failure survival risk models are developed using CPXR(Log) on an EHR dataset.The performance of CPXR(Log) models are compared with some of the state-of-the-art classification methods. Computerized survival prediction in healthcare identifying the risk of disease mortality, helps healthcare providers to effectively manage their patients by providing appropriate treatment options. In this study, we propose to apply a classification algorithm, Contrast Pattern Aided Logistic Regression (CPXR(Log)) with the probabilistic loss function, to develop and validate prognostic risk models to predict 1, 2, and 5year survival in heart failure (HF) using data from electronic health records (EHRs) at Mayo Clinic. The CPXR(Log) constructs a pattern aided logistic regression model defined by several patterns and corresponding local logistic regression models. One of the models generated by CPXR(Log) achieved an AUC and accuracy of 0.94 and 0.91, respectively, and significantly outperformed prognostic models reported in prior studies. Data extracted from EHRs allowed incorporation of patient co-morbidities into our models which helped improve the performance of the CPXR(Log) models (15.9% AUC improvement), although did not improve the accuracy of the models built by other classifiers. We also propose a probabilistic loss function to determine the large error and small error instances. The new loss function used in the algorithm outperforms other functions used in the previous studies by 1% improvement in the AUC. This study revealed that using EHR data to build prediction models can be very challenging using existing classification methods due to the high dimensionality and complexity of EHR data. The risk models developed by CPXR(Log) also reveal that HF is a highly heterogeneous disease, i.e., different subgroups of HF patients require different types of considerations with their diagnosis and treatment. Our risk models provided two valuable insights for application of predictive modeling techniques in biomedicine: Logistic risk models often make systematic prediction errors, and it is prudent to use subgroup based prediction models such as those given by CPXR(Log) when investigating heterogeneous diseases.

Receiver operating characteristic (ROC) curves are frequently used in biomedical informatics research to evaluate classification and prediction models for decision support, diagnosis, and prognosis. ROC analysis investigates the accuracy of a model’s ability to separate positive from negative cases (such as predicting the presence or absence of disease), and the results are independent of the prevalence of positive cases in the study population. It is especially useful in evaluating predictive models or other tests that produce output values over a continuous range, since it captures the trade-off between sensitivity and specificity over that range. There are many ways to conduct an ROC analysis. The best approach depends on the experiment; an inappropriate approach can easily lead to incorrect conclusions. In this article, we review the basic concepts of ROC analysis, illustrate their use with sample calculations, make recommendations drawn from the literature, and list readily available software.