Comparison of variable selection methods for clinical predictive modeling.


OBJECTIVE:Modern machine learning-based modeling methods are increasingly applied to clinical problems. One such application is in variable selection methods for predictive modeling. However, there is limited research comparing the performance of classic and modern for variable selection in clinical datasets. MATERIALS AND METHODS:We analyzed the performance of eight different variable selection methods: four regression-based methods (stepwise backward selection using p-value and AIC, Least Absolute Shrinkage and Selection Operator, and Elastic Net) and four tree-based methods (Variable Selection Using Random Forest, Regularized Random Forests, Boruta, and Gradient Boosted Feature Selection). We used two clinical datasets of different sizes, a multicenter adult clinical deterioration cohort and a single center pediatric acute kidney injury cohort. Method evaluation included measures of parsimony, variable importance, and discrimination. RESULTS:In the large, multicenter dataset, the modern tree-based Variable Selection Using Random Forest and the Gradient Boosted Feature Selection methods achieved the best parsimony. In the smaller, single-center dataset, the classic regression-based stepwise backward selection using p-value and AIC methods achieved the best parsimony. In both datasets, variable selection tended to decrease the accuracy of the random forest models and increase the accuracy of logistic regression models. CONCLUSIONS:The performance of classic regression-based and modern tree-based variable selection methods is associated with the size of the clinical dataset used. Classic regression-based variable selection methods seem to achieve better parsimony in clinical prediction problems in smaller datasets while modern tree-based methods perform better in larger datasets.


Int J Med Inform


Sanchez-Pinto LN,Venable LR,Fahrenbach J,Churpek MM




Has Abstract


2018-08-01 00:00:00












  • Speech recognition for the anaesthesia record during crisis scenarios.

    abstract:INTRODUCTION:This article describes the evaluation of a prototype speech-input interface to an anaesthesia patient record, conducted in a full-scale anaesthesia simulator involving six doctor-nurse anaesthetist teams. OBJECTIVE:The aims of the experiment were, first, to assess the potential advantages and disadvantage...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Alapetite A

    更新日期:2008-07-01 00:00:00

  • Healthcare technology management competency and its impacts on IT-healthcare partnerships development.

    abstract:OBJECTIVE:This study presents a conceptual model to investigate the healthcare technology management (HTM) competency required by healthcare IS professionals and the impact of such competency in gaining strategic advantages through information technology (IT) by development of partnerships with people from different di...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Wu JH,Chen YC,Greenes RA

    更新日期:2009-02-01 00:00:00

  • Learning ontological rules to extract multiple relations of genic interactions from text.

    abstract:INTRODUCTION:Information extraction (IE) systems have been proposed in recent years to extract genic interactions from bibliographical resources. They are limited to single interaction relations, and have to face a trade-off between recall and precision, by focusing either on specific interactions (for precision), or g...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Manine AP,Alphonse E,Bessières P

    更新日期:2009-12-01 00:00:00

  • Information quality of a Nursing Information System depends on the nurses: a combined quantitative and qualitative evaluation.

    abstract:PURPOSE:Providing access to patient information is the key factor in nurses' adoption of a Nursing Information System (NIS). In this study the requirements for information quality and the perceived quality of information are investigated. A teaching hospital in the Netherlands has developed a NIS as a module of the Hos...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Michel-Verkerke MB

    更新日期:2012-10-01 00:00:00

  • Training inter-physician communication using the Dynamic Patient Simulator.

    abstract:PURPOSE:Clear and adequate communication between physicians is essential in modern medicine. Nevertheless, the medical curricula in The Netherlands lack an identifiable part in their education concerning inter-physician communication training. To train medical students in inter-physician communication skills using the ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Sijstermans R,Jaspers MW,Bloemendaal PM,Schoonderwaldt EM

    更新日期:2007-05-01 00:00:00

  • Parametric estimation of the continuous non-stationary spectrum and its dynamics in surface EMG studies.

    abstract::Frequency spectrum of surface electromyographic signals (SEMGs) exhibit a non-stationary nature even in the case of constant level isometric muscle contractions due to changes related to muscle fatigue processes. These changes can be evaluated by methods for estimation of time-varying (TV) spectrum. The most widely ad...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Korosec D

    更新日期:2000-09-01 00:00:00

  • Feasibility and acceptability of virtual academic detailing on opioid prescribing.

    abstract:INTRODUCTION:Social distancing requirements during COVID-19 pose a challenge to conducting traditional academic detailing, which typically involves in-person peer education visits to improve patient outcomes. The main alternative is to conduct virtual academic detailing delivered through web-based technology, but this ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Smart MH,Mandava MR,Lee TA,Pickard AS

    更新日期:2021-03-01 00:00:00

  • GEMS: a system for automated cancer diagnosis and biomarker discovery from microarray gene expression data.

    abstract::The success of treatment of patients with cancer depends on establishing an accurate diagnosis. To this end, we have built a system called GEMS (gene expression model selector) for the automated development and evaluation of high-quality cancer diagnostic models and biomarker discovery from microarray gene expression ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Statnikov A,Tsamardinos I,Dosbayev Y,Aliferis CF

    更新日期:2005-08-01 00:00:00

  • Multimedia system based on programmed instruction in medical genetics: construction and evaluation.

    abstract::A multimedia system used as an auxiliary didactic tool for teaching medical genetics, HGEN, is based on non-linear programmed instruction and multimedia. HGEN was implemented in layers for PC compatible using MULTIMEDIA TOOLBOOK and DELPHI. It includes basic medical genetics concepts (inheritance patterns and cytogene...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Volpe RM,Aquino MT,Norato DY

    更新日期:1998-06-01 00:00:00

  • Characterizing vaping posts on Instagram by using unsupervised machine learning.

    abstract::Electronic cigarettes (e-cigarettes) usage has surged substantially across the globe, particularly among adolescents and young adults. The ever-increasing prevalence of social media makes it highly convenient to access and engage with content on numerous substances, including e-cigarettes. A comprehensive dataset of 5...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Ketonen V,Malik A

    更新日期:2020-09-01 00:00:00

  • Motivating and assisting physical exercise in independently living older adults: a pilot study.

    abstract:BACKGROUND:With age reaction time, coordination and cognition tend to deteriorate, which may lead to gait impairments, falls and injuries. To reduce this problem in elderly and to improve health, well-being and independence, regular balance and strength exercises are recommended. However, elderly face strong barriers t...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Silveira P,van het Reve E,Daniel F,Casati F,de Bruin ED

    更新日期:2013-05-01 00:00:00

  • Measuring mobile patient safety information system success: an empirical study.

    abstract:OBJECTIVE:The Health Risk Reminders and Surveillance (HRRS) system was designed to deliver critical abnormal test results of severely ill patients from Laboratory, Radiology, and Pathology departments to physicians within 5 min using cell phone text messages. This paper explores the success of the HRRS system. METHOD:...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Jen WY,Chao CC

    更新日期:2008-10-01 00:00:00

  • Towards reinforcing telemedicine adoption amongst clinicians in Nigeria.

    abstract::Telemedicine systems have been considered as a necessary measure to alleviate the shortfall in skilled medical specialists in developing countries. However, the obvious challenge is whether clinicians are willing to use this technological innovation, which has aided medical practice globally. One factor which has rece...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Adenuga KI,Iahad NA,Miskon S

    更新日期:2017-08-01 00:00:00

  • Theorizing the health service usage behavior of family caregivers: a qualitative study of an internet-based intervention.

    abstract:PURPOSE:The purpose of this qualitative study was to improve understanding of family caregivers' use of Web-based intervention support by integrating three theoretical models. The study applied the Anderson's model of health service utilization, Venkatesh's theory of technology acceptance, and Chatman's and Wilson's in...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Chiu TM,Eysenbach G

    更新日期:2011-11-01 00:00:00

  • What are the main patient safety concerns of healthcare stakeholders: a mixed-method study of Web-based text.

    abstract:OBJECTIVES:Various healthcare stakeholders define quality of care in different ways. Public policy could advocate all these concerns. This study was conducted to identify the main themes on patient safety of stakeholders expressed before and after the Patient Safety Act was enacted in Korea in 2015. DESIGN:Longitudina...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Cho I,Lee M,Kim Y

    更新日期:2020-08-01 00:00:00

  • Medical informatics at Heidelberg/Heilbronn: status-evaluation-new challenges in a specialised curriculum for medical informatics after thirty years of evolution.

    abstract::After reporting on characteristics, structure and contents of the specialised informatics-based curriculum for medical informatics (MI) at the University of Heidelberg/University of Applied Sciences Heilbronn, the paper describes the development during the last 5 years, and in particular a complementary health care or...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Leven FJ,Knaup P,Schmidt D,Wetter T

    更新日期:2004-03-18 00:00:00

  • Feasibility and acceptability of an iris biometric system for unique patient identification in routine HIV services in Kenya.

    abstract:BACKGROUND:Use of routine HIV programme data for surveillance is often limited due to inaccuracies associated with patient misclassification which can be addressed by unique patient identification.We assessed the feasibility and acceptability of integrating an iris recognition biometric identification system into routi...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Anne N,Dunbar MD,Abuna F,Simpson P,Macharia P,Betz B,Cherutich P,Bukusi D,Carey F

    更新日期:2020-01-01 00:00:00

  • Morphosemantic parsing of medical compound words: transferring a French analyzer to English.

    abstract:PURPOSE:Medical language, as many technical languages, is rich with morphologically complex words, many of which take their roots in Greek and Latin--in which case they are called neoclassical compounds. Morphosemantic analysis can help generate definitions of such words. The similarity of structure of those compounds ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Deléger L,Namer F,Zweigenbaum P

    更新日期:2009-04-01 00:00:00

  • Supporting medical communication for older patients with a shared touch-screen computer.

    abstract:OBJECTIVE:Increasingly health care facilities are adopting electronic medical record systems and installing computer workstations in patient exam rooms. The introduction of computer workstations into the medical interview process makes it important to consider the impact of such technology on older patients as well as ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Piper AM,Hollan JD

    更新日期:2013-11-01 00:00:00

  • SCP-ECG and Vital Signs Information Representation--two examples of successful transcontinental cooperation in medical informatics standardization.

    abstract::During the past 2-3 decades, development work on Medical Devices focused on improving their functionality (device control, signal analysis, pattern recognition and classification). At present, the dominant user requirement is information integration. This requires interconnectivity and interoperability of devices and ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章,评审


    authors: Zywietz C

    更新日期:1998-02-01 00:00:00

  • Classifying disease outbreak reports using n-grams and semantic features.

    abstract:INTRODUCTION:This paper explores the benefits of using n-grams and semantic features for the classification of disease outbreak reports, in the context of the BioCaster disease outbreak report text mining system. A novel feature of this work is the use of a general purpose semantic tagger - the USAS tagger - to generat...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Conway M,Doan S,Kawazoe A,Collier N

    更新日期:2009-12-01 00:00:00

  • Using SNOMED CT-encoded problems to improve ICD-10-CM coding-A randomized controlled experiment.

    abstract:OBJECTIVE:Clinical problems in the Electronic Health Record that are encoded in SNOMED CT can be translated into ICD-10-CM codes through the NLM's SNOMED CT to ICD-10-CM map (NLM Map). This study evaluates the potential benefits of using the map-generated codes to assist manual ICD-10-CM coding. METHODS:De-identified ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章,随机对照试验


    authors: Fung KW,Xu J,Rosenbloom ST,Campbell JR

    更新日期:2019-06-01 00:00:00

  • The role of standardized data and terminological systems in computerized clinical decision support systems: literature review and survey.

    abstract:INTRODUCTION:Clinical decision support systems (CDSSs) should be seamlessly integrated with existing clinical information systems to enable automatic provision of advice at the time and place where decisions are made. It has been suggested that a lack of agreed data standards frequently hampers this integration. We per...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章,评审


    authors: Ahmadian L,van Engen-Verheul M,Bakhshi-Raiez F,Peek N,Cornet R,de Keizer NF

    更新日期:2011-02-01 00:00:00

  • Prevention of prescription errors by computerized, on-line surveillance of drug order entry.

    abstract:AIMS:The present study was undertaken to quantify the impact of computerized drug order entry system (CDOE) connected to the patients' database, on the incidence and type of prescription errors (PEs) in the medical service, and to delineate the causes for remaining errors. METHODS:Drug orders were reviewed daily by a ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Oliven A,Michalake I,Zalman D,Dorman E,Yeshurun D,Odeh M

    更新日期:2005-06-01 00:00:00

  • Security concept in 'MyAngelWeb' a website for the individual patient at risk of emergency.

    abstract::We describe the Security Plan for the 'MyAngelWeb' service. The different actors involved in the service are subject to different security procedures. The core of the security system is implemented at the host site by means of a DBMS and standard Information Technology tools. Hardware requirements for sustainable secu...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Pinciroli F,Nahaissi D,Boschini M,Ferrari R,Meloni G,Camnasio M,Spaggiari P,Carnerone G

    更新日期:2000-11-01 00:00:00

  • A proposed taxonomy for characterization and assessment of avian influenza outbreaks.

    abstract:PURPOSE:The speed and high potential impact of avian influenza's (AI) on local bird populations, poultry economies and human health make timely and coordinated characterization, assessment and response to possible threats essential. To collaborate effectively, stakeholders (public health, medical, veterinary, and agric...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Mohammed SL,Lehmann HP,Kim GR

    更新日期:2009-03-01 00:00:00

  • Risk-based postprandial hypoglycemia forecasting using supervised learning.

    abstract:BACKGROUND:Predicting insulin-induced postprandial hypoglycemic events is critical for the safety of type 1 diabetes patients because an early warning of hypoglycemia facilitates correction of the insulin bolus before its administration. The postprandial hypoglycemic event counts can be lowered by reducing the size of ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Oviedo S,Contreras I,Quirós C,Giménez M,Conget I,Vehi J

    更新日期:2019-06-01 00:00:00

  • A novel concept for integrating and delivering health information using a comprehensive digital dashboard: An analysis of healthcare professionals' intention to adopt a new system and the trend of its real usage.

    abstract:OBJECTIVE:To introduce a new concept of medical dashboard system called BESTBoard. Such a system was implemented in all wards in a tertiary academic hospital to explore the development process, core designs, functions, usability and feasibility. METHODS:The task-force team made user interface designs for 6 months base...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Lee K,Jung SY,Hwang H,Yoo S,Baek HY,Baek RM,Kim S

    更新日期:2017-01-01 00:00:00

  • Introducing electronic messaging in Norwegian healthcare: unintended consequences for interprofessional collaboration.

    abstract:OBJECTIVE:The introduction of health information technologies (HIT) can lead to unintended consequences. We studied a newly introduced electronic messaging (e-messaging) system for communication between homecare providers and general practitioners (GPs) in Norway. The objective of this paper is to identify and discuss ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Melby L,Hellesø R

    更新日期:2014-05-01 00:00:00

  • Search engines, news wires and digital epidemiology: Presumptions and facts.

    abstract:BACKGROUND:Digital epidemiology tries to identify diseases dynamics and spread behaviors using digital traces collected via search engines logs and social media posts. However, the impacts of news on information-seeking behaviors have been remained unknown. METHODS:Data employed in this research provided from two sour...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章


    authors: Kaveh-Yazdy F,Zareh-Bidoki AM

    更新日期:2018-07-01 00:00:00