A machine learning model to predict the risk of 30-day readmissions in patients with heart failure: a retrospective analysis of ... View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2018-12

AUTHORS

Sara Bersche Golas, Takuma Shibahara, Stephen Agboola, Hiroko Otaki, Jumpei Sato, Tatsuya Nakae, Toru Hisamitsu, Go Kojima, Jennifer Felsted, Sujay Kakarmath, Joseph Kvedar, Kamal Jethwani

ABSTRACT

BACKGROUND: Heart failure is one of the leading causes of hospitalization in the United States. Advances in big data solutions allow for storage, management, and mining of large volumes of structured and semi-structured data, such as complex healthcare data. Applying these advances to complex healthcare data has led to the development of risk prediction models to help identify patients who would benefit most from disease management programs in an effort to reduce readmissions and healthcare cost, but the results of these efforts have been varied. The primary aim of this study was to develop a 30-day readmission risk prediction model for heart failure patients discharged from a hospital admission. METHODS: We used longitudinal electronic medical record data of heart failure patients admitted within a large healthcare system. Feature vectors included structured demographic, utilization, and clinical data, as well as selected extracts of un-structured data from clinician-authored notes. The risk prediction model was developed using deep unified networks (DUNs), a new mesh-like network structure of deep learning designed to avoid over-fitting. The model was validated with 10-fold cross-validation and results compared to models based on logistic regression, gradient boosting, and maxout networks. Overall model performance was assessed using concordance statistic. We also selected a discrimination threshold based on maximum projected cost saving to the Partners Healthcare system. RESULTS: Data from 11,510 patients with 27,334 admissions and 6369 30-day readmissions were used to train the model. After data processing, the final model included 3512 variables. The DUNs model had the best performance after 10-fold cross-validation. AUCs for prediction models were 0.664 ± 0.015, 0.650 ± 0.011, 0.695 ± 0.016 and 0.705 ± 0.015 for logistic regression, gradient boosting, maxout networks, and DUNs respectively. The DUNs model had an accuracy of 76.4% at the classification threshold that corresponded with maximum cost saving to the hospital. CONCLUSIONS: Deep learning techniques performed better than other traditional techniques in developing this EMR-based prediction model for 30-day readmissions in heart failure patients. Such models can be used to identify heart failure patients with impending hospitalization, enabling care teams to target interventions at their most high-risk patients and improving overall clinical outcomes. More... »

PAGES

44

Identifiers

URI

http://scigraph.springernature.com/pub.10.1186/s12911-018-0620-z

DOI

http://dx.doi.org/10.1186/s12911-018-0620-z

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1105038824

PUBMED

https://www.ncbi.nlm.nih.gov/pubmed/29929496


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0104", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Statistics", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/01", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Mathematical Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Aged", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Aged, 80 and over", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Deep Learning", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Electronic Health Records", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Female", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Heart Failure", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Humans", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Male", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Middle Aged", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Models, Theoretical", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Patient Readmission", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Prognosis", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Retrospective Studies", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Partners HealthCare", 
          "id": "https://www.grid.ac/institutes/grid.452687.a", 
          "name": [
            "Partners Connected Health Innovation, Partners HealthCare, 25 New Chardon St., Suite 300, 02114, Boston, MA, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Golas", 
        "givenName": "Sara Bersche", 
        "id": "sg:person.014060211643.34", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014060211643.34"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "name": [
            "Research and Development Group, Hitachi, Ltd, Tokyo, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Shibahara", 
        "givenName": "Takuma", 
        "id": "sg:person.013271567674.21", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013271567674.21"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Harvard University", 
          "id": "https://www.grid.ac/institutes/grid.38142.3c", 
          "name": [
            "Partners Connected Health Innovation, Partners HealthCare, 25 New Chardon St., Suite 300, 02114, Boston, MA, USA", 
            "Massachusetts General Hospital, Boston, MA, USA", 
            "Harvard Medical School, Boston, MA, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Agboola", 
        "givenName": "Stephen", 
        "id": "sg:person.01213741420.82", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01213741420.82"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "name": [
            "Research and Development Group, Hitachi, Ltd, Tokyo, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Otaki", 
        "givenName": "Hiroko", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "name": [
            "Research and Development Group, Hitachi, Ltd, Tokyo, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Sato", 
        "givenName": "Jumpei", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "name": [
            "Research and Development Group, Hitachi, Ltd, Tokyo, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Nakae", 
        "givenName": "Tatsuya", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "name": [
            "Research and Development Group, Hitachi, Ltd, Tokyo, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Hisamitsu", 
        "givenName": "Toru", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "name": [
            "Research and Development Group, Hitachi, Ltd, Tokyo, Japan"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Kojima", 
        "givenName": "Go", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Partners HealthCare", 
          "id": "https://www.grid.ac/institutes/grid.452687.a", 
          "name": [
            "Partners Connected Health Innovation, Partners HealthCare, 25 New Chardon St., Suite 300, 02114, Boston, MA, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Felsted", 
        "givenName": "Jennifer", 
        "id": "sg:person.010417361171.27", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010417361171.27"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Harvard University", 
          "id": "https://www.grid.ac/institutes/grid.38142.3c", 
          "name": [
            "Partners Connected Health Innovation, Partners HealthCare, 25 New Chardon St., Suite 300, 02114, Boston, MA, USA", 
            "Massachusetts General Hospital, Boston, MA, USA", 
            "Harvard Medical School, Boston, MA, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Kakarmath", 
        "givenName": "Sujay", 
        "id": "sg:person.01315043075.40", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01315043075.40"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Harvard University", 
          "id": "https://www.grid.ac/institutes/grid.38142.3c", 
          "name": [
            "Partners Connected Health Innovation, Partners HealthCare, 25 New Chardon St., Suite 300, 02114, Boston, MA, USA", 
            "Massachusetts General Hospital, Boston, MA, USA", 
            "Harvard Medical School, Boston, MA, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Kvedar", 
        "givenName": "Joseph", 
        "id": "sg:person.0634547144.29", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0634547144.29"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Harvard University", 
          "id": "https://www.grid.ac/institutes/grid.38142.3c", 
          "name": [
            "Partners Connected Health Innovation, Partners HealthCare, 25 New Chardon St., Suite 300, 02114, Boston, MA, USA", 
            "Massachusetts General Hospital, Boston, MA, USA", 
            "Harvard Medical School, Boston, MA, USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Jethwani", 
        "givenName": "Kamal", 
        "id": "sg:person.01243717261.76", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01243717261.76"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "https://doi.org/10.1016/j.ijcard.2015.05.180", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1003550549"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1371/journal.pone.0109264", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1009761479"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.jchf.2014.04.006", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1011117462"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.jacc.2011.11.040", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1014871075"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.ahj.2004.08.005", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1015025502"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1145/2975167.2985648", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1018105009"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1136/hrt.2005.068296", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1021319717"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1001/jamacardio.2016.3956", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1021734188"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1145/2939672.2939785", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1021899069"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1161/01.cir.0000437741.48606.98", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1027254145"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1161/01.cir.0000437741.48606.98", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1027254145"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1186/1471-2261-14-73", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1030341963", 
          "https://doi.org/10.1186/1471-2261-14-73"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.jacc.2014.08.036", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1032652311"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1001/jama.2012.216476", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1032723779"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1161/cir.0000000000000350", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1037084752"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1161/cir.0000000000000350", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1037084752"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1111/jgs.12062", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1037968505"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1056/nejmsa0803563", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1042476196"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.jbi.2015.05.016", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1048455712"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1038/nrcardio.2016.42", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1048608264", 
          "https://doi.org/10.1038/nrcardio.2016.42"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1142/9789813207813_0027", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1049736053"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1136/bmjopen-2016-011060", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1050996757"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1024/0040-5930/a000127", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1056316792"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/jbhi.2016.2636665", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061277348"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/jbhi.2016.2636665", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061277348"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1161/circheartfailure.115.002594", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1063343165"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1161/circheartfailure.115.002594", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1063343165"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.2196/jmir.4417", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1069286306"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.7326/m16-0185", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1073742993"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://app.dimensions.ai/details/publication/pub.1078672329", 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/bigdata.2013.6691760", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1094462783"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/healthcom.2016.7749452", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1094564374"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.3115/979617.979625", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1099142424"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2018-12", 
    "datePublishedReg": "2018-12-01", 
    "description": "BACKGROUND: Heart failure is one of the leading causes of hospitalization in the United States. Advances in big data solutions allow for storage, management, and mining of large volumes of structured and semi-structured data, such as complex healthcare data. Applying these advances to complex healthcare data has led to the development of risk prediction models to help identify patients who would benefit\u00a0most from disease management programs in an\u00a0effort to reduce readmissions and healthcare cost, but the results of these efforts have been varied. The primary aim of this study was to develop a 30-day readmission risk prediction model for heart failure patients discharged from a hospital admission.\nMETHODS: We used longitudinal electronic medical record data of heart failure patients admitted within a large healthcare system. Feature vectors included structured demographic, utilization, and clinical data, as well as selected extracts of un-structured data from clinician-authored notes. The risk prediction model was developed using deep unified networks (DUNs), a new mesh-like network structure of deep learning designed to avoid over-fitting. The model was validated with 10-fold cross-validation and results compared to models based on logistic regression, gradient boosting, and maxout networks. Overall model performance was assessed using concordance statistic. We also selected a discrimination threshold based on maximum projected cost saving to the Partners Healthcare system.\nRESULTS: Data from 11,510 patients with 27,334 admissions and 6369 30-day readmissions were used to train the model. After data processing, the final model included 3512 variables. The DUNs model had the best performance after\u00a010-fold cross-validation. AUCs for prediction models were 0.664\u2009\u00b1\u20090.015, 0.650\u2009\u00b1\u20090.011, 0.695\u2009\u00b1\u20090.016 and 0.705\u2009\u00b1\u20090.015 for logistic regression, gradient boosting, maxout networks, and DUNs respectively. The DUNs model had an accuracy of 76.4% at the classification threshold that corresponded with maximum cost saving to the hospital.\nCONCLUSIONS: Deep learning techniques performed better than other traditional techniques in developing this EMR-based prediction model for 30-day readmissions in heart failure patients. Such models can be used to identify heart failure patients with impending hospitalization, enabling care teams to target interventions at their most high-risk patients and improving overall clinical outcomes.", 
    "genre": "research_article", 
    "id": "sg:pub.10.1186/s12911-018-0620-z", 
    "inLanguage": [
      "en"
    ], 
    "isAccessibleForFree": true, 
    "isPartOf": [
      {
        "id": "sg:journal.1028432", 
        "issn": [
          "1472-6947"
        ], 
        "name": "BMC Medical Informatics and Decision Making", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "1", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "18"
      }
    ], 
    "name": "A machine learning model to predict the risk of 30-day readmissions in patients with heart failure: a retrospective analysis of electronic medical records data", 
    "pagination": "44", 
    "productId": [
      {
        "name": "readcube_id", 
        "type": "PropertyValue", 
        "value": [
          "54dabca9d050461357c1c7320c92271356ba1429a6b362480be7ff9d0bdc0068"
        ]
      }, 
      {
        "name": "pubmed_id", 
        "type": "PropertyValue", 
        "value": [
          "29929496"
        ]
      }, 
      {
        "name": "nlm_unique_id", 
        "type": "PropertyValue", 
        "value": [
          "101088682"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1186/s12911-018-0620-z"
        ]
      }, 
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1105038824"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1186/s12911-018-0620-z", 
      "https://app.dimensions.ai/details/publication/pub.1105038824"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2019-04-11T12:23", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-uberresearch-data-dimensions-target-20181106-alternative/cleanup/v134/2549eaecd7973599484d7c17b260dba0a4ecb94b/merge/v9/a6c9fde33151104705d4d7ff012ea9563521a3ce/jats-lookup/v90/0000000362_0000000362/records_87091_00000001.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "https://link.springer.com/10.1186%2Fs12911-018-0620-z"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1186/s12911-018-0620-z'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1186/s12911-018-0620-z'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1186/s12911-018-0620-z'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1186/s12911-018-0620-z'


 

This table displays all metadata directly associated to this object as RDF triples.

297 TRIPLES      21 PREDICATES      71 URIs      34 LITERALS      22 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1186/s12911-018-0620-z schema:about N1602541122df4add8fad2582b3489095
2 N160e4fd837b54ac286806c5c39120c13
3 N27e85d8914d24424b83e5a1ceaebd9a6
4 N29aea5bd8d7f4ab793d10e530bea822b
5 N5e12417f11e8434081485d61aa5bc4f2
6 N664d0b3a5a4b4b29aa1e630570cc4214
7 N6db7e2ed11ee4b0ab941caf120012aff
8 N709eb81337564a21af5774eb4477f0dd
9 N70eb6e54ea5749feb7e456a18532929d
10 Nbe85de6d95f2435eb820be35dca5ea06
11 Ncbc902c4761f48219c20fbaedde312ab
12 Nef67ef49129e492aab4fa76a8b0443c5
13 Nf76aa3a825d84531a035129d69f071ff
14 anzsrc-for:01
15 anzsrc-for:0104
16 schema:author N13f946398269435794709b3cecd19900
17 schema:citation sg:pub.10.1038/nrcardio.2016.42
18 sg:pub.10.1186/1471-2261-14-73
19 https://app.dimensions.ai/details/publication/pub.1078672329
20 https://doi.org/10.1001/jama.2012.216476
21 https://doi.org/10.1001/jamacardio.2016.3956
22 https://doi.org/10.1016/j.ahj.2004.08.005
23 https://doi.org/10.1016/j.ijcard.2015.05.180
24 https://doi.org/10.1016/j.jacc.2011.11.040
25 https://doi.org/10.1016/j.jacc.2014.08.036
26 https://doi.org/10.1016/j.jbi.2015.05.016
27 https://doi.org/10.1016/j.jchf.2014.04.006
28 https://doi.org/10.1024/0040-5930/a000127
29 https://doi.org/10.1056/nejmsa0803563
30 https://doi.org/10.1109/bigdata.2013.6691760
31 https://doi.org/10.1109/healthcom.2016.7749452
32 https://doi.org/10.1109/jbhi.2016.2636665
33 https://doi.org/10.1111/jgs.12062
34 https://doi.org/10.1136/bmjopen-2016-011060
35 https://doi.org/10.1136/hrt.2005.068296
36 https://doi.org/10.1142/9789813207813_0027
37 https://doi.org/10.1145/2939672.2939785
38 https://doi.org/10.1145/2975167.2985648
39 https://doi.org/10.1161/01.cir.0000437741.48606.98
40 https://doi.org/10.1161/cir.0000000000000350
41 https://doi.org/10.1161/circheartfailure.115.002594
42 https://doi.org/10.1371/journal.pone.0109264
43 https://doi.org/10.2196/jmir.4417
44 https://doi.org/10.3115/979617.979625
45 https://doi.org/10.7326/m16-0185
46 schema:datePublished 2018-12
47 schema:datePublishedReg 2018-12-01
48 schema:description BACKGROUND: Heart failure is one of the leading causes of hospitalization in the United States. Advances in big data solutions allow for storage, management, and mining of large volumes of structured and semi-structured data, such as complex healthcare data. Applying these advances to complex healthcare data has led to the development of risk prediction models to help identify patients who would benefit most from disease management programs in an effort to reduce readmissions and healthcare cost, but the results of these efforts have been varied. The primary aim of this study was to develop a 30-day readmission risk prediction model for heart failure patients discharged from a hospital admission. METHODS: We used longitudinal electronic medical record data of heart failure patients admitted within a large healthcare system. Feature vectors included structured demographic, utilization, and clinical data, as well as selected extracts of un-structured data from clinician-authored notes. The risk prediction model was developed using deep unified networks (DUNs), a new mesh-like network structure of deep learning designed to avoid over-fitting. The model was validated with 10-fold cross-validation and results compared to models based on logistic regression, gradient boosting, and maxout networks. Overall model performance was assessed using concordance statistic. We also selected a discrimination threshold based on maximum projected cost saving to the Partners Healthcare system. RESULTS: Data from 11,510 patients with 27,334 admissions and 6369 30-day readmissions were used to train the model. After data processing, the final model included 3512 variables. The DUNs model had the best performance after 10-fold cross-validation. AUCs for prediction models were 0.664 ± 0.015, 0.650 ± 0.011, 0.695 ± 0.016 and 0.705 ± 0.015 for logistic regression, gradient boosting, maxout networks, and DUNs respectively. The DUNs model had an accuracy of 76.4% at the classification threshold that corresponded with maximum cost saving to the hospital. CONCLUSIONS: Deep learning techniques performed better than other traditional techniques in developing this EMR-based prediction model for 30-day readmissions in heart failure patients. Such models can be used to identify heart failure patients with impending hospitalization, enabling care teams to target interventions at their most high-risk patients and improving overall clinical outcomes.
49 schema:genre research_article
50 schema:inLanguage en
51 schema:isAccessibleForFree true
52 schema:isPartOf N60f24bbd0ada4f8cbad4a1763d02c9fd
53 Nfbb6806d36654e19afe1a76598a44ea9
54 sg:journal.1028432
55 schema:name A machine learning model to predict the risk of 30-day readmissions in patients with heart failure: a retrospective analysis of electronic medical records data
56 schema:pagination 44
57 schema:productId N4a3402cab8e54986abef26fd01161208
58 N4d0d0e7753994880bc414140a8903d89
59 Na8051e77feee499e92aee74e5e89de21
60 Nbedee72d930e4558a1c14eac7687eff8
61 Ne161af5a92844f8b92f378ff468ddb6e
62 schema:sameAs https://app.dimensions.ai/details/publication/pub.1105038824
63 https://doi.org/10.1186/s12911-018-0620-z
64 schema:sdDatePublished 2019-04-11T12:23
65 schema:sdLicense https://scigraph.springernature.com/explorer/license/
66 schema:sdPublisher N8a0b099e238b47d88f7403cf09a5fc73
67 schema:url https://link.springer.com/10.1186%2Fs12911-018-0620-z
68 sgo:license sg:explorer/license/
69 sgo:sdDataset articles
70 rdf:type schema:ScholarlyArticle
71 N13f946398269435794709b3cecd19900 rdf:first sg:person.014060211643.34
72 rdf:rest Nf8cd563c825743c6b8262fedb2a517b5
73 N1602541122df4add8fad2582b3489095 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
74 schema:name Male
75 rdf:type schema:DefinedTerm
76 N160e4fd837b54ac286806c5c39120c13 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
77 schema:name Deep Learning
78 rdf:type schema:DefinedTerm
79 N1929b589b9fe4790bf17a4aff35a9fe8 schema:affiliation Nffd8627322d048ee97f674b7b61087c6
80 schema:familyName Nakae
81 schema:givenName Tatsuya
82 rdf:type schema:Person
83 N27e85d8914d24424b83e5a1ceaebd9a6 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
84 schema:name Humans
85 rdf:type schema:DefinedTerm
86 N29aea5bd8d7f4ab793d10e530bea822b schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
87 schema:name Prognosis
88 rdf:type schema:DefinedTerm
89 N31e3302063db40fba24f9de3f913bd93 rdf:first sg:person.01213741420.82
90 rdf:rest N5c057cb18e0a4cb2b89d82718e7b97f8
91 N36a651f8e4ae4a758ac025386cdd1685 schema:name Research and Development Group, Hitachi, Ltd, Tokyo, Japan
92 rdf:type schema:Organization
93 N3a80a30e06984d7fbc6853dbdd6544f2 schema:name Research and Development Group, Hitachi, Ltd, Tokyo, Japan
94 rdf:type schema:Organization
95 N420d9f68b4f64cf384e8ba1f61f72897 rdf:first sg:person.01243717261.76
96 rdf:rest rdf:nil
97 N4a3402cab8e54986abef26fd01161208 schema:name nlm_unique_id
98 schema:value 101088682
99 rdf:type schema:PropertyValue
100 N4d0d0e7753994880bc414140a8903d89 schema:name pubmed_id
101 schema:value 29929496
102 rdf:type schema:PropertyValue
103 N4d268c7b9c864aa1bd3bdea85c4c1199 schema:affiliation N36a651f8e4ae4a758ac025386cdd1685
104 schema:familyName Sato
105 schema:givenName Jumpei
106 rdf:type schema:Person
107 N5c057cb18e0a4cb2b89d82718e7b97f8 rdf:first Nfb516841fa6b4355a04e4b2063433c05
108 rdf:rest N8bfeb51c8e4240848061473d38938692
109 N5e12417f11e8434081485d61aa5bc4f2 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
110 schema:name Electronic Health Records
111 rdf:type schema:DefinedTerm
112 N60f24bbd0ada4f8cbad4a1763d02c9fd schema:volumeNumber 18
113 rdf:type schema:PublicationVolume
114 N625c2fdb43b544ef8d7ea52d26866f60 rdf:first sg:person.0634547144.29
115 rdf:rest N420d9f68b4f64cf384e8ba1f61f72897
116 N664d0b3a5a4b4b29aa1e630570cc4214 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
117 schema:name Aged
118 rdf:type schema:DefinedTerm
119 N6db7e2ed11ee4b0ab941caf120012aff schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
120 schema:name Retrospective Studies
121 rdf:type schema:DefinedTerm
122 N709eb81337564a21af5774eb4477f0dd schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
123 schema:name Heart Failure
124 rdf:type schema:DefinedTerm
125 N70eb6e54ea5749feb7e456a18532929d schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
126 schema:name Female
127 rdf:type schema:DefinedTerm
128 N7ff0ada4886a4859b1744cf130504f69 rdf:first sg:person.01315043075.40
129 rdf:rest N625c2fdb43b544ef8d7ea52d26866f60
130 N8a0b099e238b47d88f7403cf09a5fc73 schema:name Springer Nature - SN SciGraph project
131 rdf:type schema:Organization
132 N8bfeb51c8e4240848061473d38938692 rdf:first N4d268c7b9c864aa1bd3bdea85c4c1199
133 rdf:rest Ndfb3617e13fc4ec98774484ac39a6ea3
134 N9d6b27b9a5bb48c8ab9dcf179aaaee95 rdf:first sg:person.010417361171.27
135 rdf:rest N7ff0ada4886a4859b1744cf130504f69
136 Na8051e77feee499e92aee74e5e89de21 schema:name readcube_id
137 schema:value 54dabca9d050461357c1c7320c92271356ba1429a6b362480be7ff9d0bdc0068
138 rdf:type schema:PropertyValue
139 Nbda12e85815b43c88073e6920cb6b4cb schema:affiliation Nfec02dabeb694ba196f74af5cdafcf6f
140 schema:familyName Kojima
141 schema:givenName Go
142 rdf:type schema:Person
143 Nbe85de6d95f2435eb820be35dca5ea06 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
144 schema:name Patient Readmission
145 rdf:type schema:DefinedTerm
146 Nbedee72d930e4558a1c14eac7687eff8 schema:name dimensions_id
147 schema:value pub.1105038824
148 rdf:type schema:PropertyValue
149 Ncbc902c4761f48219c20fbaedde312ab schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
150 schema:name Aged, 80 and over
151 rdf:type schema:DefinedTerm
152 Nd598327ca03947b083c6c90a62cc7fc8 schema:affiliation Nd90c097d8de94d47bbc04985e21b43af
153 schema:familyName Hisamitsu
154 schema:givenName Toru
155 rdf:type schema:Person
156 Nd90c097d8de94d47bbc04985e21b43af schema:name Research and Development Group, Hitachi, Ltd, Tokyo, Japan
157 rdf:type schema:Organization
158 Ndfb3617e13fc4ec98774484ac39a6ea3 rdf:first N1929b589b9fe4790bf17a4aff35a9fe8
159 rdf:rest Nf5937a30857a4751a6321826a91e8b1e
160 Ne161af5a92844f8b92f378ff468ddb6e schema:name doi
161 schema:value 10.1186/s12911-018-0620-z
162 rdf:type schema:PropertyValue
163 Ne4406b1253d24f8cb2ad762dd7e9b7aa rdf:first Nbda12e85815b43c88073e6920cb6b4cb
164 rdf:rest N9d6b27b9a5bb48c8ab9dcf179aaaee95
165 Nef67ef49129e492aab4fa76a8b0443c5 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
166 schema:name Models, Theoretical
167 rdf:type schema:DefinedTerm
168 Nf5937a30857a4751a6321826a91e8b1e rdf:first Nd598327ca03947b083c6c90a62cc7fc8
169 rdf:rest Ne4406b1253d24f8cb2ad762dd7e9b7aa
170 Nf76aa3a825d84531a035129d69f071ff schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
171 schema:name Middle Aged
172 rdf:type schema:DefinedTerm
173 Nf8cd563c825743c6b8262fedb2a517b5 rdf:first sg:person.013271567674.21
174 rdf:rest N31e3302063db40fba24f9de3f913bd93
175 Nf90bb187b0a446b5893d424bf447655a schema:name Research and Development Group, Hitachi, Ltd, Tokyo, Japan
176 rdf:type schema:Organization
177 Nfb516841fa6b4355a04e4b2063433c05 schema:affiliation N3a80a30e06984d7fbc6853dbdd6544f2
178 schema:familyName Otaki
179 schema:givenName Hiroko
180 rdf:type schema:Person
181 Nfbb6806d36654e19afe1a76598a44ea9 schema:issueNumber 1
182 rdf:type schema:PublicationIssue
183 Nfec02dabeb694ba196f74af5cdafcf6f schema:name Research and Development Group, Hitachi, Ltd, Tokyo, Japan
184 rdf:type schema:Organization
185 Nffd8627322d048ee97f674b7b61087c6 schema:name Research and Development Group, Hitachi, Ltd, Tokyo, Japan
186 rdf:type schema:Organization
187 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
188 schema:name Mathematical Sciences
189 rdf:type schema:DefinedTerm
190 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
191 schema:name Statistics
192 rdf:type schema:DefinedTerm
193 sg:journal.1028432 schema:issn 1472-6947
194 schema:name BMC Medical Informatics and Decision Making
195 rdf:type schema:Periodical
196 sg:person.010417361171.27 schema:affiliation https://www.grid.ac/institutes/grid.452687.a
197 schema:familyName Felsted
198 schema:givenName Jennifer
199 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010417361171.27
200 rdf:type schema:Person
201 sg:person.01213741420.82 schema:affiliation https://www.grid.ac/institutes/grid.38142.3c
202 schema:familyName Agboola
203 schema:givenName Stephen
204 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01213741420.82
205 rdf:type schema:Person
206 sg:person.01243717261.76 schema:affiliation https://www.grid.ac/institutes/grid.38142.3c
207 schema:familyName Jethwani
208 schema:givenName Kamal
209 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01243717261.76
210 rdf:type schema:Person
211 sg:person.01315043075.40 schema:affiliation https://www.grid.ac/institutes/grid.38142.3c
212 schema:familyName Kakarmath
213 schema:givenName Sujay
214 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01315043075.40
215 rdf:type schema:Person
216 sg:person.013271567674.21 schema:affiliation Nf90bb187b0a446b5893d424bf447655a
217 schema:familyName Shibahara
218 schema:givenName Takuma
219 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013271567674.21
220 rdf:type schema:Person
221 sg:person.014060211643.34 schema:affiliation https://www.grid.ac/institutes/grid.452687.a
222 schema:familyName Golas
223 schema:givenName Sara Bersche
224 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014060211643.34
225 rdf:type schema:Person
226 sg:person.0634547144.29 schema:affiliation https://www.grid.ac/institutes/grid.38142.3c
227 schema:familyName Kvedar
228 schema:givenName Joseph
229 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0634547144.29
230 rdf:type schema:Person
231 sg:pub.10.1038/nrcardio.2016.42 schema:sameAs https://app.dimensions.ai/details/publication/pub.1048608264
232 https://doi.org/10.1038/nrcardio.2016.42
233 rdf:type schema:CreativeWork
234 sg:pub.10.1186/1471-2261-14-73 schema:sameAs https://app.dimensions.ai/details/publication/pub.1030341963
235 https://doi.org/10.1186/1471-2261-14-73
236 rdf:type schema:CreativeWork
237 https://app.dimensions.ai/details/publication/pub.1078672329 schema:CreativeWork
238 https://doi.org/10.1001/jama.2012.216476 schema:sameAs https://app.dimensions.ai/details/publication/pub.1032723779
239 rdf:type schema:CreativeWork
240 https://doi.org/10.1001/jamacardio.2016.3956 schema:sameAs https://app.dimensions.ai/details/publication/pub.1021734188
241 rdf:type schema:CreativeWork
242 https://doi.org/10.1016/j.ahj.2004.08.005 schema:sameAs https://app.dimensions.ai/details/publication/pub.1015025502
243 rdf:type schema:CreativeWork
244 https://doi.org/10.1016/j.ijcard.2015.05.180 schema:sameAs https://app.dimensions.ai/details/publication/pub.1003550549
245 rdf:type schema:CreativeWork
246 https://doi.org/10.1016/j.jacc.2011.11.040 schema:sameAs https://app.dimensions.ai/details/publication/pub.1014871075
247 rdf:type schema:CreativeWork
248 https://doi.org/10.1016/j.jacc.2014.08.036 schema:sameAs https://app.dimensions.ai/details/publication/pub.1032652311
249 rdf:type schema:CreativeWork
250 https://doi.org/10.1016/j.jbi.2015.05.016 schema:sameAs https://app.dimensions.ai/details/publication/pub.1048455712
251 rdf:type schema:CreativeWork
252 https://doi.org/10.1016/j.jchf.2014.04.006 schema:sameAs https://app.dimensions.ai/details/publication/pub.1011117462
253 rdf:type schema:CreativeWork
254 https://doi.org/10.1024/0040-5930/a000127 schema:sameAs https://app.dimensions.ai/details/publication/pub.1056316792
255 rdf:type schema:CreativeWork
256 https://doi.org/10.1056/nejmsa0803563 schema:sameAs https://app.dimensions.ai/details/publication/pub.1042476196
257 rdf:type schema:CreativeWork
258 https://doi.org/10.1109/bigdata.2013.6691760 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094462783
259 rdf:type schema:CreativeWork
260 https://doi.org/10.1109/healthcom.2016.7749452 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094564374
261 rdf:type schema:CreativeWork
262 https://doi.org/10.1109/jbhi.2016.2636665 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061277348
263 rdf:type schema:CreativeWork
264 https://doi.org/10.1111/jgs.12062 schema:sameAs https://app.dimensions.ai/details/publication/pub.1037968505
265 rdf:type schema:CreativeWork
266 https://doi.org/10.1136/bmjopen-2016-011060 schema:sameAs https://app.dimensions.ai/details/publication/pub.1050996757
267 rdf:type schema:CreativeWork
268 https://doi.org/10.1136/hrt.2005.068296 schema:sameAs https://app.dimensions.ai/details/publication/pub.1021319717
269 rdf:type schema:CreativeWork
270 https://doi.org/10.1142/9789813207813_0027 schema:sameAs https://app.dimensions.ai/details/publication/pub.1049736053
271 rdf:type schema:CreativeWork
272 https://doi.org/10.1145/2939672.2939785 schema:sameAs https://app.dimensions.ai/details/publication/pub.1021899069
273 rdf:type schema:CreativeWork
274 https://doi.org/10.1145/2975167.2985648 schema:sameAs https://app.dimensions.ai/details/publication/pub.1018105009
275 rdf:type schema:CreativeWork
276 https://doi.org/10.1161/01.cir.0000437741.48606.98 schema:sameAs https://app.dimensions.ai/details/publication/pub.1027254145
277 rdf:type schema:CreativeWork
278 https://doi.org/10.1161/cir.0000000000000350 schema:sameAs https://app.dimensions.ai/details/publication/pub.1037084752
279 rdf:type schema:CreativeWork
280 https://doi.org/10.1161/circheartfailure.115.002594 schema:sameAs https://app.dimensions.ai/details/publication/pub.1063343165
281 rdf:type schema:CreativeWork
282 https://doi.org/10.1371/journal.pone.0109264 schema:sameAs https://app.dimensions.ai/details/publication/pub.1009761479
283 rdf:type schema:CreativeWork
284 https://doi.org/10.2196/jmir.4417 schema:sameAs https://app.dimensions.ai/details/publication/pub.1069286306
285 rdf:type schema:CreativeWork
286 https://doi.org/10.3115/979617.979625 schema:sameAs https://app.dimensions.ai/details/publication/pub.1099142424
287 rdf:type schema:CreativeWork
288 https://doi.org/10.7326/m16-0185 schema:sameAs https://app.dimensions.ai/details/publication/pub.1073742993
289 rdf:type schema:CreativeWork
290 https://www.grid.ac/institutes/grid.38142.3c schema:alternateName Harvard University
291 schema:name Harvard Medical School, Boston, MA, USA
292 Massachusetts General Hospital, Boston, MA, USA
293 Partners Connected Health Innovation, Partners HealthCare, 25 New Chardon St., Suite 300, 02114, Boston, MA, USA
294 rdf:type schema:Organization
295 https://www.grid.ac/institutes/grid.452687.a schema:alternateName Partners HealthCare
296 schema:name Partners Connected Health Innovation, Partners HealthCare, 25 New Chardon St., Suite 300, 02114, Boston, MA, USA
297 rdf:type schema:Organization
 




Preview window. Press ESC to close (or click here)


...