Comparison of variable selection methods for clinical predictive modeling.

Abstract:

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.

journal_name

Int J Med Inform

authors

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

doi

10.1016/j.ijmedinf.2018.05.006

subject

Has Abstract

pub_date

2018-08-01 00:00:00

pages

10-17

eissn

1386-5056

issn

1872-8243

pii

S1386-5056(18)30581-1

journal_volume

116

pub_type

杂志文章
  • The emerging role of cognitive computing in healthcare: A systematic literature review.

    abstract::To assist medical professional in better treatment of diseases, and improve patient outcomes, healthcare has brought about a cognitive computing revolution. The cognitive computing system processes enormous amounts of data instantly to answer specific queries and makes customized intelligent recommendations. Cognitive...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2019.04.024

    authors: Behera RK,Bala PK,Dhir A

    更新日期:2019-09-01 00:00:00

  • LinkEHR-Ed: a multi-reference model archetype editor based on formal semantics.

    abstract:PURPOSE:To develop a powerful archetype editing framework capable of handling multiple reference models and oriented towards the semantic description and standardization of legacy data. METHODS:The main prerequisite for implementing tools providing enhanced support for archetypes is the clear specification of archetyp...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2009.03.006

    authors: Maldonado JA,Moner D,Boscá D,Fernández-Breis JT,Angulo C,Robles M

    更新日期:2009-08-01 00:00:00

  • Scaling-up health information systems to improve HIV treatment: An assessment of initial patient monitoring systems in Mozambique.

    abstract:INTRODUCTION:The rapid scale-up of HIV care and treatment in resource-limited countries requires concurrent, rapid development of health information systems to support quality service delivery. Mozambique, a country with an 11.5% prevalence of HIV, has developed nation-wide patient monitoring systems (PMS) with standar...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2016.11.002

    authors: Hochgesang M,Zamudio-Haas S,Moran L,Nhampossa L,Packel L,Leslie H,Richards J,Shade SB

    更新日期:2017-01-01 00:00:00

  • Fractal analysis of surface EMG signals from the biceps.

    abstract::Nonlinear analysis techniques are necessary to understand the complexity of the EMG. The purpose of the present study was to determine the fractal dimension of surface EMG obtained from the biceps brachii of normal subjects during isokinetic flexion-extension of the arm. The measurements were obtained with different l...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/s1386-5056(97)00029-4

    authors: Gupta V,Suryanarayanan S,Reddy NP

    更新日期:1997-07-01 00:00:00

  • Healthcare professional's demand for knowledge in informatics.

    abstract:OBJECTIVE:To develop an economic model of health care professional demand for knowledge capital in health informatics. DESIGN:Case study with application of the contingent valuation method to develop a small-scale model. SETTING:Specialized clinic at a university Hospital in Sweden. RESULTS:The model displays the ec...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/s1386-5056(98)00151-8

    authors: Vimarlund V,Timpka T,Hallberg N

    更新日期:1999-02-01 00:00:00

  • Feasibility of mobile health game "Fume" in supporting tobacco-related health literacy among early adolescents: A three-armed cluster randomized design.

    abstract:OBJECTIVE:New interventions supporting health literacy and a tobacco-free lifestyle in adolescence are needed to narrow the widening gap in existing health inequalities. Health games offer potential and could be utilized for example in school healthcare, but more research is needed to increase the understanding of the ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章,随机对照试验

    doi:10.1016/j.ijmedinf.2018.02.013

    authors: Parisod H,Pakarinen A,Axelin A,Löyttyniemi E,Smed J,Salanterä S

    更新日期:2018-05-01 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

    doi:10.1016/j.ijmedinf.2020.104162

    authors: Cho I,Lee M,Kim Y

    更新日期:2020-08-01 00:00:00

  • Self-assessed health among older people in Europe and internet use.

    abstract:BACKGROUND:About ten years ago, an age-related digital divide was identified, where 'the elderly' denoted a group of people at risk of losing the benefits of a digital society. The aims of this work are to find a relationship between self-assessed health and internet use by older people in European countries and to asc...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2020.104240

    authors: Tavares AI

    更新日期:2020-09-01 00:00:00

  • Evaluating the agreement of users with usability problems identified by heuristic evaluation.

    abstract:OBJECTIVE:Heuristic evaluation (HE) is an inspection method that can be used for identifying usability problems of health information systems. There is insufficient evidence to what extent these problems are experienced by users in their working environment. The objective of this study was to investigate the agreement ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2018.05.012

    authors: Khajouei R,Ameri A,Jahani Y

    更新日期:2018-09-01 00:00:00

  • Predictive factors of telemedicine service acceptance and behavioral intention of physicians.

    abstract:PURPOSE:Despite the proliferation of telemedicine technology, telemedicine service acceptance has been slow in actual healthcare settings. The purpose of this research is to develop a theoretical model for explaining the predictive factors influencing physicians' willingness to use telemedicine technology to provide he...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2014.05.005

    authors: Rho MJ,Choi IY,Lee J

    更新日期:2014-08-01 00:00:00

  • PCLiON: An Ontology for Data Standardization and Sharing of Prostate Cancer Associated Lifestyles.

    abstract:BACKGROUND:Researches on Lifestyle medicine (LM) have emerged in recent years to garner wide attention. Prostate cancer (PCa) could be prevented and treated by positive lifestyles, but the association between lifestyles and PCa is always personalized. OBJECTIVES:In order to solve the heterogeneity and diversity of dif...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2020.104332

    authors: Chen Y,Yu C,Liu X,Xi T,Xu G,Sun Y,Zhu F,Shen B

    更新日期:2021-01-01 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

    doi:10.1016/j.ijmedinf.2018.03.017

    authors: Kaveh-Yazdy F,Zareh-Bidoki AM

    更新日期:2018-07-01 00:00:00

  • Applying density-based outlier identifications using multiple datasets for validation of stroke clinical outcomes.

    abstract:INTRODUCTION:Clinicians commonly use the modified Rankin Scale (mRS) and the Barthel Index (BI) to measure clinical outcome after stroke. These are potential targets in machine learning models for stroke outcome prediction. Therefore, the quality of the measurements is crucial for training and validation of these model...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2019.103988

    authors: Lin CH,Hsu KC,Johnson KR,Luby M,Fann YC

    更新日期:2019-12-01 00:00:00

  • Decaying relevance of clinical data towards future decisions in data-driven inpatient clinical order sets.

    abstract:OBJECTIVE:Determine how varying longitudinal historical training data can impact prediction of future clinical decisions. Estimate the "decay rate" of clinical data source relevance. MATERIALS AND METHODS:We trained a clinical order recommender system, analogous to Netflix or Amazon's "Customers who bought A also boug...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2017.03.006

    authors: Chen JH,Alagappan M,Goldstein MK,Asch SM,Altman RB

    更新日期:2017-06-01 00:00:00

  • Inferred joint multigram models for medical term normalization according to ICD.

    abstract:BACKGROUND:Electronic Health Records (EHRs) are written using spontaneous natural language. Often, terms do not match standard terminology like the one available through the International Classification of Diseases (ICD). OBJECTIVE:Information retrieval and exchange can be improved using standard terminology. Our aim ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2017.12.007

    authors: Pérez A,Atutxa A,Casillas A,Gojenola K,Sellart Á

    更新日期:2018-02-01 00:00:00

  • Perspectives and expectations for telemedicine opportunities from families of nursing home residents and caregivers in nursing homes.

    abstract:BACKGROUND:This study assessed current perspectives and expectations for telemedicine by nursing home caregivers and families of nursing home patients in Taipei, Taiwan. METHODS:A total of 116 interviews were conducted with family members (n=37) and caregivers (n=79) using an original, four-part questionnaire devised ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2009.02.009

    authors: Chang JY,Chen LK,Chang CC

    更新日期:2009-07-01 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

    doi:10.1016/j.ijmedinf.2020.104365

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

    更新日期:2021-03-01 00:00:00

  • Organizational routines, innovation, and flexibility: the application of narrative networks to dynamic workflow.

    abstract:OBJECTIVE:The purpose of this paper is to demonstrate how current visual representations of organizational and technological processes do not fully account for the variability present in everyday practices. We further demonstrate how narrative networks can augment these representations to indicate potential areas for s...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2011.01.005

    authors: Hayes GR,Lee CP,Dourish P

    更新日期:2011-08-01 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

    doi:10.1016/j.ijmedinf.2017.05.008

    authors: Adenuga KI,Iahad NA,Miskon S

    更新日期:2017-08-01 00:00:00

  • Trustworthy reuse of health data: a transnational perspective.

    abstract:BACKGROUND:The widespread adoption of electronic health records (EHRs) is accelerating the collection of sensitive clinical data. The availability of these data raises privacy concerns, yet sharing the data is essential for public health, longitudinal patient care, and clinical research. METHOD:Following previous work...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2012.11.003

    authors: Geissbuhler A,Safran C,Buchan I,Bellazzi R,Labkoff S,Eilenberg K,Leese A,Richardson C,Mantas J,Murray P,De Moor G

    更新日期:2013-01-01 00:00:00

  • Education and training in health informatics: the IT-EDUCTRA project.

    abstract::In this contribution both the EDUCTRA project of the European Advanced Informatics in Medicine Programme and the IT-EDUCTRA project of the Telematics Applications Programme (Health Sector) are described. EDUCTRA had as aim to investigate which gaps in knowledge health professionals have with respect to health informat...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/s1386-5056(98)00066-5

    authors: Hasman A

    更新日期:1998-06-01 00:00:00

  • The quality of evidence in health informatics: how did the quality of healthcare IT evaluation publications develop from 1982 to 2005?

    abstract:OBJECTIVE:To obtain an overview of study designs and study methods used in research evaluating IT in health care, to present a list of quality criteria by which all kinds of reported evaluation studies on IT systems in health care can be assessed, and to assess the quality of reported evaluation studies on IT in health...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章,评审

    doi:10.1016/j.ijmedinf.2006.11.009

    authors: de Keizer NF,Ammenwerth E

    更新日期:2008-01-01 00:00:00

  • Complexity and the science of implementation in health IT--knowledge gaps and future visions.

    abstract:OBJECTIVES:The intent of this paper is in the examination of health IT implementation processes - the barriers to and facilitators of successful implementation, identification of a beginning set of implementation best practices, the identification of gaps in the health IT implementation body of knowledge, and recommend...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2013.10.009

    authors: Abbott PA,Foster J,Marin Hde F,Dykes PC

    更新日期:2014-07-01 00:00:00

  • A computer-based outpatient clinical referral system.

    abstract::The process of generating a clinical referral for a patient, and the resulting transfer of information from the primary care physician to the specialist and back again, are key components in the struggle to deliver less costly and more effective clinical care. We have created a computer-based, outpatient clinical refe...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/s1386-5056(99)00027-1

    authors: Sittig DF,Gandhi TK,Franklin M,Turetsky M,Sussman AJ,Fairchild DG,Bates DW,Komaroff AL,Teich JM

    更新日期:1999-08-01 00:00:00

  • Validating a decision support system for anti-epileptic drug treatment. Part II: adjusting anti-epileptic drug treatment.

    abstract::A model of expertise for monitoring antiepileptic drug treatment was implemented in a decision support system. We validated the advice of the system regarding treatment decisions at first follow-up with 265 paper cases based on patient records. The reference for comparison is based on the opinions of neurologists. We ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/s1386-5056(99)00052-0

    authors: Smeets R,Talmon J,Meinardi H,Hasman A

    更新日期:1999-11-01 00:00:00

  • Usability of quality measures for online health information: Can commonly used technical quality criteria be reliably assessed?

    abstract:PURPOSE:Many criteria have been developed to rate the quality of online health information. To effectively evaluate quality, consumers must use quality criteria that can be reliably assessed. However, few instruments have been validated for inter-rater agreement. Therefore, we assessed the degree to which two raters co...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2005.02.002

    authors: Bernstam EV,Sagaram S,Walji M,Johnson CW,Meric-Bernstam F

    更新日期:2005-08-01 00:00:00

  • Using secure messaging to update medications list in ambulatory care setting.

    abstract:UNLABELLED:This study analyzed patient adoption of secure messaging to update medication list in an ambulatory care setting. The objective was to establish demographic differences between users and non-users of secure messaging for medications list update. Efficiency of secure messaging for the updates was compared to ...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2015.06.003

    authors: Raghu TS,Frey K,Chang YH,Cheng MR,Freimund S,Patel A

    更新日期:2015-10-01 00:00:00

  • SHARE road map for HealthGrids: methodology.

    abstract::The SHARE(1) project (http://www.eu-share.org) was asked to identify the key developments needed to achieve wide adoption and deployment of HealthGrids throughout Europe. The project was asked to organise these as milestones on a road map, so that all technical advances, social actions, economic investments and ethica...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/j.ijmedinf.2008.10.003

    authors: Olive M,Rahmouni H,Solomonides T,Breton V,Legré Y,Blanquer I,Hernandez V

    更新日期:2009-04-01 00:00:00

  • Exploiting the terminological approach from CEN/TC251 and GALEN to support semantic interoperability of healthcare record systems.

    abstract::We apply the principles included in two CEN standards (ENV 12265, ENV 12264) to the analysis of the semantic structure of health record systems, to support their semantic interoperability. This result was made possible by dramatic methodological progress in the field of terminological systems--due to a worldwide evolu...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/s1386-5056(97)00116-0

    authors: Rossi Mori A,Consorti F

    更新日期:1998-02-01 00:00:00

  • The anaesthesia critical incident reporting system: an experience based database.

    abstract::To date there have been fewer than a dozen studies on the nature of, and contributory factors in, critical incidents (CI) in anaesthesia. The first of these, by Cooper and colleagues, showed that the vast majority of their CI involved human error [1]. Most recently, the on-going Australian Incident Monitoring Study (A...

    journal_title:International journal of medical informatics

    pub_type: 杂志文章

    doi:10.1016/s1386-5056(97)00087-7

    authors: Staender S,Davies J,Helmreich B,Sexton B,Kaufmann M

    更新日期:1997-11-01 00:00:00