Measurement and application of patient similarity in personalized predictive modeling based on electronic medical records.
Ontology highlight
ABSTRACT: Background
Conventional risk prediction techniques may not be the most suitable approach for personalized prediction for individual patients. Therefore, individualized predictive modeling based on similar patients has emerged. This study aimed to propose a comprehensive measurement of patient similarity using real-world electronic medical records data, and evaluate the effectiveness of the individualized prediction of a patient's diabetes status based on the patient similarity.Results
When using no more than 30% of the whole training sample, the personalized predictive models outperformed corresponding traditional models built on randomly selected training samples of the same size as the personalized models (P?ConclusionsThe proposed patient similarity measurement was effective when developing personalized predictive models. The successful application of patient similarity in predicting a patient's diabetes status provided useful references for diagnostic decision-making support by investigating the evidence on similar patients.
SUBMITTER: Wang N
PROVIDER: S-EPMC6788002 | biostudies-literature | 2019 Oct
REPOSITORIES: biostudies-literature
Publications
Measurement and application of patient similarity in personalized predictive modeling based on electronic medical records.
Wang Ni N Huang Yanqun Y Liu Honglei H Fei Xiaolu X Wei Lan L Zhao Xiangkun X Chen Hui H
Biomedical engineering online 20191011 1
<h4>Background</h4>Conventional risk prediction techniques may not be the most suitable approach for personalized prediction for individual patients. Therefore, individualized predictive modeling based on similar patients has emerged. This study aimed to propose a comprehensive measurement of patient similarity using real-world electronic medical records data, and evaluate the effectiveness of the individualized prediction of a patient's diabetes status based on the patient similarity.<h4>Result ...[more]
Similar Datasets
Project description:electronic Medical Records & Genomics (eMERGE)
| PRJNA158491 | ENA
Project description:BackgroundA new learning-based patient similarity measurement was proposed to measure patients' similarity for heterogeneous electronic medical records (EMRs) data.MethodsWe first calculated feature-level similarities according to the features' attributes. A domain expert provided patient similarity scores of 30 randomly selected patients. These similarity scores and feature-level similarities for 30 patients comprised the labeled sample set, which was used for the semi-supervised learning algorithm to learn the patient-level similarities for all patients. Then we used the k-nearest neighbor (kNN) classifier to predict four liver conditions. The predictive performances were compared in four different situations. We also compared the performances between personalized kNN models and other machine learning models. We assessed the predictive performances by the area under the receiver operating characteristic curve (AUC), F1-score, and cross-entropy (CE) loss.ResultsAs the size of the random training samples increased, the kNN models using the learned patient similarity to select near neighbors consistently outperformed those using the Euclidean distance to select near neighbors (all P values < 0.001). The kNN models using the learned patient similarity to identify the top k nearest neighbors from the random training samples also had a higher best-performance (AUC: 0.95 vs. 0.89, F1-score: 0.84 vs. 0.67, and CE loss: 1.22 vs. 1.82) than those using the Euclidean distance. As the size of the similar training samples increased, which composed the most similar samples determined by the learned patient similarity, the performance of kNN models using the simple Euclidean distance to select the near neighbors degraded gradually. When exchanging the role of the Euclidean distance, and the learned patient similarity in selecting the near neighbors and similar training samples, the performance of the kNN models gradually increased. These two kinds of kNN models had the same best-performance of AUC 0.95, F1-score 0.84, and CE loss 1.22. Among the four reference models, the highest AUC and F1-score were 0.94 and 0.80, separately, which were both lower than those for the simple and similarity-based kNN models.ConclusionsThis learning-based method opened an opportunity for similarity measurement based on heterogeneous EMR data and supported the secondary use of EMR data.
| S-EPMC8323210 | biostudies-literature
Project description:ObjectivesAcute kidney injury (AKI) in hospitalized patients puts them at much higher risk for developing future health problems such as chronic kidney disease, stroke, and heart disease. Accurate AKI prediction would allow timely prevention and intervention. However, current AKI prediction researches pay less attention to model building strategies that meet complex clinical application scenario. This study aims to build and evaluate AKI prediction models from multiple perspectives that reflect different clinical applications.Materials and methodsA retrospective cohort of 76 957 encounters and relevant clinical variables were extracted from a tertiary care, academic hospital electronic medical record (EMR) system between November 2007 and December 2016. Five machine learning methods were used to build prediction models. Prediction tasks from 4 clinical perspectives with different modeling and evaluation strategies were designed to build and evaluate the models.ResultsExperimental analysis of the AKI prediction models built from 4 different clinical perspectives suggest a realistic prediction performance in cross-validated area under the curve ranging from 0.720 to 0.764.DiscussionResults show that models built at admission is effective for predicting AKI events in the next day; models built using data with a fixed lead time to AKI onset is still effective in the dynamic clinical application scenario in which each patient's lead time to AKI onset is different.ConclusionTo our best knowledge, this is the first systematic study to explore multiple clinical perspectives in building predictive models for AKI in the general inpatient population to reflect real performance in clinical application.
| S-EPMC6447093 | biostudies-literature
Project description:Controlling type 2 diabetes (T2DM) requires a comprehensive approach including patient education, self-monitoring of blood glucose, individualized behavioral strategies, and frequent contact with healthcare professionals (HCPs). We aimed to compare the efficacy of a personalized lifestyle intervention based on a mobile phone application with regular care in participants with T2DM. This is an ongoing randomized controlled open-label parallel-group trial with a target accrual of 282 participants, of which 181 have been enrolled to date. Participants are randomly assigned to one of three groups: (1) regular care; (2) mobile diabetes management; or (3) mobile diabetes management with HCP feedback. The mobile application is enabled to integrate with both electronic medical records (EMR) and a web-based diabetes management system for HCPs. It can send customized messages based on participants' responses to lifestyle questionnaires administered at the baseline. The intervention period is 26 weeks followed by observation for 26 weeks. We evaluate the intervention's features in order to assess its clinical utility and efficacy and compare outcomes with regular care considering relevant clinical factors, such as age, baseline HbA1c, etc. We expect our study to provide new evidence in support of customized mobile application tools for the management of T2DM.
| S-EPMC8155840 | biostudies-literature
Project description:ObjectivesThe development of clinical predictive models hinges upon the availability of comprehensive clinical data. Tapping into such resources requires considerable effort from clinicians, data scientists, and engineers. Specifically, these efforts are focused on data extraction and preprocessing steps required prior to modeling, including complex database queries. A handful of software libraries exist that can reduce this complexity by building upon data standards. However, a gap remains concerning electronic health records (EHRs) stored in star schema clinical data warehouses, an approach often adopted in practice. In this article, we introduce the FlexIBle EHR Retrieval (FIBER) tool: a Python library built on top of a star schema (i2b2) clinical data warehouse that enables flexible generation of modeling-ready cohorts as data frames.Materials and methodsFIBER was developed on top of a large-scale star schema EHR database which contains data from 8 million patients and over 120 million encounters. To illustrate FIBER's capabilities, we present its application by building a heart surgery patient cohort with subsequent prediction of acute kidney injury (AKI) with various machine learning models.ResultsUsing FIBER, we were able to build the heart surgery cohort (n = 12 061), identify the patients that developed AKI (n = 1005), and automatically extract relevant features (n = 774). Finally, we trained machine learning models that achieved area under the curve values of up to 0.77 for this exemplary use case.ConclusionFIBER is an open-source Python library developed for extracting information from star schema clinical data warehouses and reduces time-to-modeling, helping to streamline the clinical modeling process.
| S-EPMC8327378 | biostudies-literature
Project description:BackgroundThe main goal of this study is to explore the use of features representing patient-level electronic health record (EHR) data, generated by the unsupervised deep learning algorithm autoencoder, in predictive modeling. Since autoencoder features are unsupervised, this paper focuses on their general lower-dimensional representation of EHR information in a wide variety of predictive tasks.MethodsWe compare the model with autoencoder features to traditional models: logistic model with least absolute shrinkage and selection operator (LASSO) and Random Forest algorithm. In addition, we include a predictive model using a small subset of response-specific variables (Simple Reg) and a model combining these variables with features from autoencoder (Enhanced Reg). We performed the study first on simulated data that mimics real world EHR data and then on actual EHR data from eight Advocate hospitals.ResultsOn simulated data with incorrect categories and missing data, the precision for autoencoder is 24.16% when fixing recall at 0.7, which is higher than Random Forest (23.61%) and lower than LASSO (25.32%). The precision is 20.92% in Simple Reg and improves to 24.89% in Enhanced Reg. When using real EHR data to predict the 30-day readmission rate, the precision of autoencoder is 19.04%, which again is higher than Random Forest (18.48%) and lower than LASSO (19.70%). The precisions for Simple Reg and Enhanced Reg are 18.70 and 19.69% respectively. That is, Enhanced Reg can have competitive prediction performance compared to LASSO. In addition, results show that Enhanced Reg usually relies on fewer features under the setting of simulations of this paper.ConclusionsWe conclude that autoencoder can create useful features representing the entire space of EHR data and which are applicable to a wide array of predictive tasks. Together with important response-specific predictors, we can derive efficient and robust predictive models with less labor in data extraction and model training.
| S-EPMC7043035 | biostudies-literature
Project description:PurposeWe aimed to evaluate the current situation of electronic health records (EHRs) and patient registries in the oncology departments of hospitals in Spain.MethodsThis was a cross-sectional study conducted from December 2018 to September 2019. The survey was designed ad hoc by the Outcomes Evaluation and Clinical Practice Section of the Spanish Society of Medical Oncology (SEOM) and was distributed to all head of medical oncology department members of SEOM.ResultsWe invited 148 heads of oncology departments, and 81 (54.7%) questionnaires were completed, with representation from all 17 Spanish autonomous communities. Seventy-seven (95%) of the respondents had EHRs implemented at their hospitals; of them, over 80% considered EHRs to have a positive impact on work organization and clinical practice, and 73% considered that EHRs improve the quality of patient care. In contrast, 27 (35.1%) of these respondents felt that EHRs worsened the physician-patient relationship and conveyed an additional workload (n = 29; 37.6%). Several drawbacks in the implementation of EHRs were identified, including the limited inclusion of information on both outpatients and inpatients, information recorded in free text data fields, and the availability of specific informed consent. Forty-six (56.7%) respondents had patient registries where they recorded information from all patients seen in the department.ConclusionOur study indicates that EHRs are almost universally implemented in the hospitals surveyed and are considered to have a positive impact on work organization and clinical practice. However, EHRs currently have several drawbacks that limit their use for investigational purposes.Clinical trial registrationNot applicable.
| S-EPMC8390424 | biostudies-literature
Project description:BackgroundEngaging patients in the delivery of health care has the potential to improve health outcomes and patient satisfaction. Patient portals may enhance patient engagement by enabling patients to access their electronic medical records (EMRs) and facilitating secure patient-provider communication.ObjectiveThe aim of this study was to review literature describing patient portals tethered to an EMR in inpatient settings, their role in patient engagement, and their impact on health care delivery in order to identify factors and best practices for successful implementation of this technology and areas that require further research.MethodsA systematic search for articles in the PubMed, CINAHL, and Embase databases was conducted using keywords associated with patient engagement, electronic health records, and patient portals and their respective subject headings in each database. Articles for inclusion were evaluated for quality using A Measurement Tool to Assess Systematic Reviews (AMSTAR) for systematic review articles and the Quality Assessment Tool for Studies with Diverse Designs for empirical studies. Included studies were categorized by their focus on input factors (eg, portal design), process factors (eg, portal use), and output factors (eg, benefits) and by the valence of their findings regarding patient portals (ie, positive, negative, or mixed).ResultsThe systematic search identified 58 articles for inclusion. The inputs category was addressed by 40 articles, while the processes and outputs categories were addressed by 36 and 46 articles, respectively: 47 articles addressed multiple themes across the three categories, and 11 addressed only a single theme. Nineteen articles had high- to very high-quality, 21 had medium quality, and 18 had low- to very low-quality. Findings in the inputs category showed wide-ranging portal designs; patients' privacy concerns and lack of encouragement from providers were among portal adoption barriers while information access and patient-provider communication were among facilitators. Several methods were used to train portal users with varying success. In the processes category, sociodemographic characteristics and medical conditions of patients were predictors of portal use; some patients wanted unlimited access to their EMRs, personalized health education, and nonclinical information; and patients were keen to use portals for communicating with their health care teams. In the outputs category, some but not all studies found patient portals improved patient engagement; patients perceived some portal functions as inadequate but others as useful; patients and staff thought portals may improve patient care but could cause anxiety in some patients; and portals improved patient safety, adherence to medications, and patient-provider communication but had no impact on objective health outcomes.ConclusionsWhile the evidence is currently immature, patient portals have demonstrated benefit by enabling the discovery of medical errors, improving adherence to medications, and providing patient-provider communication, etc. High-quality studies are needed to fully understand, improve, and evaluate their impact.
| S-EPMC6482406 | biostudies-literature
Project description:ObjectiveAttention networks learn an intelligent weighted averaging mechanism over a series of entities, providing increases to both performance and interpretability. In this article, we propose a novel time-aware transformer-based network and compare it to another leading model with similar characteristics. We also decompose model performance along several critical axes and examine which features contribute most to our model's performance.Materials and methodsUsing data sets representing patient records obtained between 2017 and 2019 by the Kaiser Permanente Mid-Atlantic States medical system, we construct four attentional models with varying levels of complexity on two targets (patient mortality and hospitalization). We examine how incorporating transfer learning and demographic features contribute to model success. We also test the performance of a model proposed in recent medical modeling literature. We compare these models with out-of-sample data using the area under the receiver-operator characteristic (AUROC) curve and average precision as measures of performance. We also analyze the attentional weights assigned by these models to patient diagnoses.ResultsWe found that our model significantly outperformed the alternative on a mortality prediction task (91.96% AUROC against 73.82% AUROC). Our model also outperformed on the hospitalization task, although the models were significantly more competitive in that space (82.41% AUROC against 80.33% AUROC). Furthermore, we found that demographic features and transfer learning features which are frequently omitted from new models proposed in the EMR modeling space contributed significantly to the success of our model.DiscussionWe proposed an original construction of deep learning electronic medical record models which achieved very strong performance. We found that our unique model construction outperformed on several tasks in comparison to a leading literature alternative, even when input data was held constant between them. We obtained further improvements by incorporating several methods that are frequently overlooked in new model proposals, suggesting that it will be useful to explore these options further in the future.
| S-EPMC8358476 | biostudies-literature
Project description:BackgroundAdverse drug events (ADEs) detection and assessment is at the center of pharmacovigilance. Data mining of systems, such as FDA's Adverse Event Reporting System (AERS) and more recently, Electronic Health Records (EHRs), can aid in the automatic detection and analysis of ADEs. Although different data mining approaches have been shown to be valuable, it is still crucial to improve the quality of the generated signals.ObjectiveTo leverage structural similarity by developing molecular fingerprint-based models (MFBMs) to strengthen ADE signals generated from EHR data.MethodsA reference standard of drugs known to be causally associated with the adverse event pancreatitis was used to create a MFBM. Electronic Health Records (EHRs) from the New York Presbyterian Hospital were mined to generate structured data. Disproportionality Analysis (DPA) was applied to the data, and 278 possible signals related to the ADE pancreatitis were detected. Candidate drugs associated with these signals were then assessed using the MFBM to find the most promising candidates based on structural similarity.ResultsThe use of MFBM as a means to strengthen or prioritize signals generated from the EHR significantly improved the detection accuracy of ADEs related to pancreatitis. MFBM also highlights the etiology of the ADE by identifying structurally similar drugs, which could follow a similar mechanism of action.ConclusionThe method proposed in this paper provides evidence of being a promising adjunct to existing automated ADE detection and analysis approaches.
| S-EPMC3404072 | biostudies-literature