Comparison of methods for early-readmission prediction in a high-dimensional heterogeneous covariates and time-to-event outcome framework View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2019-12

AUTHORS

Simon Bussy, Raphaël Veil, Vincent Looten, Anita Burgun, Stéphane Gaïffas, Agathe Guilloux, Brigitte Ranque, Anne-Sophie Jannot

ABSTRACT

BACKGROUND: Choosing the most performing method in terms of outcome prediction or variables selection is a recurring problem in prognosis studies, leading to many publications on methods comparison. But some aspects have received little attention. First, most comparison studies treat prediction performance and variable selection aspects separately. Second, methods are either compared within a binary outcome setting (where we want to predict whether the readmission will occur within an arbitrarily chosen delay or not) or within a survival analysis setting (where the outcomes are directly the censored times), but not both. In this paper, we propose a comparison methodology to weight up those different settings both in terms of prediction and variables selection, while incorporating advanced machine learning strategies. METHODS: Using a high-dimensional case study on a sickle-cell disease (SCD) cohort, we compare 8 statistical methods. In the binary outcome setting, we consider logistic regression (LR), support vector machine (SVM), random forest (RF), gradient boosting (GB) and neural network (NN); while on the survival analysis setting, we consider the Cox Proportional Hazards (PH), the CURE and the C-mix models. We also propose a method using Gaussian Processes to extract meaningfull structured covariates from longitudinal data. RESULTS: Among all assessed statistical methods, the survival analysis ones obtain the best results. In particular the C-mix model yields the better performances in both the two considered settings (AUC =0.94 in the binary outcome setting), as well as interesting interpretation aspects. There is some consistency in selected covariates across methods within a setting, but not much across the two settings. CONCLUSIONS: It appears that learning withing the survival analysis setting first (so using all the temporal information), and then going back to a binary prediction using the survival estimates gives significantly better prediction performances than the ones obtained by models trained "directly" within the binary outcome setting. More... »

PAGES

50

Identifiers

URI

http://scigraph.springernature.com/pub.10.1186/s12874-019-0673-4

DOI

http://dx.doi.org/10.1186/s12874-019-0673-4

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1112585404

PUBMED

https://www.ncbi.nlm.nih.gov/pubmed/30841867


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"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Sorbonne University", 
          "id": "https://www.grid.ac/institutes/grid.462844.8", 
          "name": [
            "Laboratoire de Probabilit\u00e9s Statistique et Mod\u00e9lisation (LPSM), UMR 8001, Sorbonne University, 4 Place Jussieu, 75005, Paris, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Bussy", 
        "givenName": "Simon", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Centre de Recherche des Cordeliers", 
          "id": "https://www.grid.ac/institutes/grid.417925.c", 
          "name": [
            "Assistance Publique-H\u00f4pitaux de Paris, Biomedical Informatics and Public Health Department, European Georges Pompidou Hospital, 20 Rue Leblanc, 75015, Paris, France", 
            "INSERM UMRS 1138, Eq22, Centre de Recherche des Cordeliers, Universit\u00e9 Paris Descartes, 15 Rue de l\u2019\u00c9cole de M\u00e9decine, 75006, Paris, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Veil", 
        "givenName": "Rapha\u00ebl", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Centre de Recherche des Cordeliers", 
          "id": "https://www.grid.ac/institutes/grid.417925.c", 
          "name": [
            "Assistance Publique-H\u00f4pitaux de Paris, Biomedical Informatics and Public Health Department, European Georges Pompidou Hospital, 20 Rue Leblanc, 75015, Paris, France", 
            "INSERM UMRS 1138, Eq22, Centre de Recherche des Cordeliers, Universit\u00e9 Paris Descartes, 15 Rue de l\u2019\u00c9cole de M\u00e9decine, 75006, Paris, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Looten", 
        "givenName": "Vincent", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Centre de Recherche des Cordeliers", 
          "id": "https://www.grid.ac/institutes/grid.417925.c", 
          "name": [
            "Assistance Publique-H\u00f4pitaux de Paris, Biomedical Informatics and Public Health Department, European Georges Pompidou Hospital, 20 Rue Leblanc, 75015, Paris, France", 
            "INSERM UMRS 1138, Eq22, Centre de Recherche des Cordeliers, Universit\u00e9 Paris Descartes, 15 Rue de l\u2019\u00c9cole de M\u00e9decine, 75006, Paris, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Burgun", 
        "givenName": "Anita", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Centre de Math\u00e9matiques Appliqu\u00e9es", 
          "id": "https://www.grid.ac/institutes/grid.462265.1", 
          "name": [
            "Laboratoire de Probabilit\u00e9s Statistique et Mod\u00e9lisation (LPSM), UMR 8001, Sorbonne University, 4 Place Jussieu, 75005, Paris, France", 
            "CMAP, UMR 7641 \u00c9cole Polytechnique CNRS, Route de Saclay, 91128, Palaiseau, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Ga\u00efffas", 
        "givenName": "St\u00e9phane", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "French National Centre for Scientific Research", 
          "id": "https://www.grid.ac/institutes/grid.4444.0", 
          "name": [
            "LAMME, Univ Evry, CNRS, Universit\u00e9 Paris-Saclay, 23 boulevard de France, 91025, Evry, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Guilloux", 
        "givenName": "Agathe", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "H\u00f4pital Europ\u00e9en Georges-Pompidou", 
          "id": "https://www.grid.ac/institutes/grid.414093.b", 
          "name": [
            "INSERM UMRS 970, Universit\u00e9 Paris Descartes, 56 rue Leblanc, 75015, Paris, France", 
            "Assistance Publique-H\u00f4pitaux de Paris, Internal Medicine Department, Georges Pompidou European Hospital, 20 Rue Leblanc, 75015, Paris, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Ranque", 
        "givenName": "Brigitte", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Centre de Recherche des Cordeliers", 
          "id": "https://www.grid.ac/institutes/grid.417925.c", 
          "name": [
            "Assistance Publique-H\u00f4pitaux de Paris, Biomedical Informatics and Public Health Department, European Georges Pompidou Hospital, 20 Rue Leblanc, 75015, Paris, France", 
            "INSERM UMRS 1138, Eq22, Centre de Recherche des Cordeliers, Universit\u00e9 Paris Descartes, 15 Rue de l\u2019\u00c9cole de M\u00e9decine, 75006, Paris, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Jannot", 
        "givenName": "Anne-Sophie", 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1186/1471-2288-12-100", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1000235299", 
          "https://doi.org/10.1186/1471-2288-12-100"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1111/j.1532-5415.1990.tb03450.x", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1001041978"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1186/1471-2105-10-213", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1002419282", 
          "https://doi.org/10.1186/1471-2105-10-213"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s00439-008-0592-7", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1003137479", 
          "https://doi.org/10.1007/s00439-008-0592-7"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s00439-008-0592-7", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1003137479", 
          "https://doi.org/10.1007/s00439-008-0592-7"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1186/1471-2288-12-102", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1003624527", 
          "https://doi.org/10.1186/1471-2288-12-102"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1073/pnas.0401736101", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1006123154"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1001/jama.2011.1561", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1007755515"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1001/jama.2010.378", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1008462250"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1186/s12874-016-0128-0", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1008888493", 
          "https://doi.org/10.1186/s12874-016-0128-0"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1002/pbc.21854", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1009687810"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1056/nejm199511023331806", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1010799486"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/s0167-9473(01)00065-2", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1012543995"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1046/j.1365-2141.2003.04193.x", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1012960136"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/aog.0b013e318160f38e", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1015539591"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/aog.0b013e318160f38e", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1015539591"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s10115-006-0040-8", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1018091708", 
          "https://doi.org/10.1007/s10115-006-0040-8"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s10115-006-0040-8", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1018091708", 
          "https://doi.org/10.1007/s10115-006-0040-8"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tnn.2005.848998", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1018563512"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1023/a:1010933404324", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1024739340", 
          "https://doi.org/10.1023/a:1010933404324"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1056/nejm199709113371107", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1028319673"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://app.dimensions.ai/details/publication/pub.1028948927", 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-1-4419-6646-9", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1028948927", 
          "https://doi.org/10.1007/978-1-4419-6646-9"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-1-4419-6646-9", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1028948927", 
          "https://doi.org/10.1007/978-1-4419-6646-9"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1056/nejm199107043250103", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1030040630"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1186/1471-2288-9-85", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1030193087", 
          "https://doi.org/10.1186/1471-2288-9-85"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.2202/1544-6115.1147", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1041046514"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1111/j.1467-9868.2005.00503.x", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1043971564"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1111/j.0006-341x.2005.030814.x", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1044144954"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/s0031-3203(96)00142-2", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1044522995"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1177/1077558704263799", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1053805691"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1177/1077558704263799", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1053805691"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1021/ci0342472", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1055401738"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1021/ci0342472", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1055401738"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1093/biomet/69.3.553", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1059419215"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1093/biomet/79.3.531", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1059420291"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1136/bmj.313.7057.628", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1062777078"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1136/bmj.313.7057.628", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1062777078"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1177/183335830303100103", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064077667"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1177/183335830303100103", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1064077667"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.18637/jss.v039.i05", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1068672583"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.2307/2529885", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1069975475"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.2307/2982890", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1070160732"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.2307/2982890", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1070160732"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://app.dimensions.ai/details/publication/pub.1078219311", 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://app.dimensions.ai/details/publication/pub.1078354293", 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.2307/3001968", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1102728208"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1177/0962280218766389", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1103284945"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://app.dimensions.ai/details/publication/pub.1106875674", 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1002/9781118548387", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1106875674"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1002/9781118150672", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1106892893"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1111/j.2517-6161.1972.tb00899.x", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1110457843"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1111/j.2517-6161.1972.tb00899.x", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1110457843"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2019-12", 
    "datePublishedReg": "2019-12-01", 
    "description": "BACKGROUND: Choosing the most performing method in terms of outcome prediction or variables selection is a recurring problem in prognosis studies, leading to many publications on methods comparison. But some aspects have received little attention. First, most comparison studies treat prediction performance and variable selection aspects separately. Second, methods are either compared within a binary outcome setting (where we want to predict whether the readmission will occur within an arbitrarily chosen delay or not) or within a survival analysis setting (where the outcomes are directly the censored times), but not both. In this paper, we propose a comparison methodology to weight up those different settings both in terms of prediction and variables selection, while incorporating advanced machine learning strategies.\nMETHODS: Using a high-dimensional case study on a sickle-cell disease (SCD) cohort, we compare 8 statistical methods. In the binary outcome setting, we consider logistic regression (LR), support vector machine (SVM), random forest (RF), gradient boosting (GB) and neural network (NN); while on the survival analysis setting, we consider the Cox Proportional Hazards (PH), the CURE and the C-mix models. We also propose a method using Gaussian Processes to extract meaningfull structured covariates from longitudinal data.\nRESULTS: Among all assessed statistical methods, the survival analysis ones obtain the best results. In particular the C-mix model yields the better performances in both the two considered settings (AUC =0.94 in the binary outcome setting), as well as interesting interpretation aspects. There is some consistency in selected covariates across methods within a setting, but not much across the two settings.\nCONCLUSIONS: It appears that learning withing the survival analysis setting first (so using all the temporal information), and then going back to a binary prediction using the survival estimates gives significantly better prediction performances than the ones obtained by models trained \"directly\" within the binary outcome setting.", 
    "genre": "research_article", 
    "id": "sg:pub.10.1186/s12874-019-0673-4", 
    "inLanguage": [
      "en"
    ], 
    "isAccessibleForFree": true, 
    "isPartOf": [
      {
        "id": "sg:journal.1024940", 
        "issn": [
          "1471-2288"
        ], 
        "name": "BMC Medical Research Methodology", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "1", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "19"
      }
    ], 
    "name": "Comparison of methods for early-readmission prediction in a high-dimensional heterogeneous covariates and time-to-event outcome framework", 
    "pagination": "50", 
    "productId": [
      {
        "name": "readcube_id", 
        "type": "PropertyValue", 
        "value": [
          "aedaa9aa4d69af21e7b83a8d7d5038c7fcffa24282fbed0b62442ad52759ed12"
        ]
      }, 
      {
        "name": "pubmed_id", 
        "type": "PropertyValue", 
        "value": [
          "30841867"
        ]
      }, 
      {
        "name": "nlm_unique_id", 
        "type": "PropertyValue", 
        "value": [
          "100968545"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1186/s12874-019-0673-4"
        ]
      }, 
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1112585404"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1186/s12874-019-0673-4", 
      "https://app.dimensions.ai/details/publication/pub.1112585404"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2019-04-11T13:20", 
    "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/v90/0000000368_0000000368/records_78968_00000001.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "https://link.springer.com/10.1186%2Fs12874-019-0673-4"
  }
]
 

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/s12874-019-0673-4'

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/s12874-019-0673-4'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1186/s12874-019-0673-4'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1186/s12874-019-0673-4'


 

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

258 TRIPLES      21 PREDICATES      72 URIs      21 LITERALS      9 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1186/s12874-019-0673-4 schema:about anzsrc-for:01
2 anzsrc-for:0104
3 schema:author N72c5daf5a20d4b7ca593031b19459b58
4 schema:citation sg:pub.10.1007/978-1-4419-6646-9
5 sg:pub.10.1007/s00439-008-0592-7
6 sg:pub.10.1007/s10115-006-0040-8
7 sg:pub.10.1023/a:1010933404324
8 sg:pub.10.1186/1471-2105-10-213
9 sg:pub.10.1186/1471-2288-12-100
10 sg:pub.10.1186/1471-2288-12-102
11 sg:pub.10.1186/1471-2288-9-85
12 sg:pub.10.1186/s12874-016-0128-0
13 https://app.dimensions.ai/details/publication/pub.1028948927
14 https://app.dimensions.ai/details/publication/pub.1078219311
15 https://app.dimensions.ai/details/publication/pub.1078354293
16 https://app.dimensions.ai/details/publication/pub.1106875674
17 https://doi.org/10.1001/jama.2010.378
18 https://doi.org/10.1001/jama.2011.1561
19 https://doi.org/10.1002/9781118150672
20 https://doi.org/10.1002/9781118548387
21 https://doi.org/10.1002/pbc.21854
22 https://doi.org/10.1016/s0031-3203(96)00142-2
23 https://doi.org/10.1016/s0167-9473(01)00065-2
24 https://doi.org/10.1021/ci0342472
25 https://doi.org/10.1046/j.1365-2141.2003.04193.x
26 https://doi.org/10.1056/nejm199107043250103
27 https://doi.org/10.1056/nejm199511023331806
28 https://doi.org/10.1056/nejm199709113371107
29 https://doi.org/10.1073/pnas.0401736101
30 https://doi.org/10.1093/biomet/69.3.553
31 https://doi.org/10.1093/biomet/79.3.531
32 https://doi.org/10.1097/aog.0b013e318160f38e
33 https://doi.org/10.1109/tnn.2005.848998
34 https://doi.org/10.1111/j.0006-341x.2005.030814.x
35 https://doi.org/10.1111/j.1467-9868.2005.00503.x
36 https://doi.org/10.1111/j.1532-5415.1990.tb03450.x
37 https://doi.org/10.1111/j.2517-6161.1972.tb00899.x
38 https://doi.org/10.1136/bmj.313.7057.628
39 https://doi.org/10.1177/0962280218766389
40 https://doi.org/10.1177/1077558704263799
41 https://doi.org/10.1177/183335830303100103
42 https://doi.org/10.18637/jss.v039.i05
43 https://doi.org/10.2202/1544-6115.1147
44 https://doi.org/10.2307/2529885
45 https://doi.org/10.2307/2982890
46 https://doi.org/10.2307/3001968
47 schema:datePublished 2019-12
48 schema:datePublishedReg 2019-12-01
49 schema:description BACKGROUND: Choosing the most performing method in terms of outcome prediction or variables selection is a recurring problem in prognosis studies, leading to many publications on methods comparison. But some aspects have received little attention. First, most comparison studies treat prediction performance and variable selection aspects separately. Second, methods are either compared within a binary outcome setting (where we want to predict whether the readmission will occur within an arbitrarily chosen delay or not) or within a survival analysis setting (where the outcomes are directly the censored times), but not both. In this paper, we propose a comparison methodology to weight up those different settings both in terms of prediction and variables selection, while incorporating advanced machine learning strategies. METHODS: Using a high-dimensional case study on a sickle-cell disease (SCD) cohort, we compare 8 statistical methods. In the binary outcome setting, we consider logistic regression (LR), support vector machine (SVM), random forest (RF), gradient boosting (GB) and neural network (NN); while on the survival analysis setting, we consider the Cox Proportional Hazards (PH), the CURE and the C-mix models. We also propose a method using Gaussian Processes to extract meaningfull structured covariates from longitudinal data. RESULTS: Among all assessed statistical methods, the survival analysis ones obtain the best results. In particular the C-mix model yields the better performances in both the two considered settings (AUC =0.94 in the binary outcome setting), as well as interesting interpretation aspects. There is some consistency in selected covariates across methods within a setting, but not much across the two settings. CONCLUSIONS: It appears that learning withing the survival analysis setting first (so using all the temporal information), and then going back to a binary prediction using the survival estimates gives significantly better prediction performances than the ones obtained by models trained "directly" within the binary outcome setting.
50 schema:genre research_article
51 schema:inLanguage en
52 schema:isAccessibleForFree true
53 schema:isPartOf N016a753e5d0b4a0a9c61b9e92ebb673f
54 Nffc081f93cce47e39fa1b266025ef8ef
55 sg:journal.1024940
56 schema:name Comparison of methods for early-readmission prediction in a high-dimensional heterogeneous covariates and time-to-event outcome framework
57 schema:pagination 50
58 schema:productId N4b5d5e12e80a455187f308e3166c74f1
59 N6286e3a917824a9ea3e9e5b69946c9f8
60 N6ca014ef4419469590a24d8b0e8844a0
61 Nd9ee7d4f3f314fa38c09a740b84ad6cb
62 Nff9d329ef73441778dda90dbc9c9b170
63 schema:sameAs https://app.dimensions.ai/details/publication/pub.1112585404
64 https://doi.org/10.1186/s12874-019-0673-4
65 schema:sdDatePublished 2019-04-11T13:20
66 schema:sdLicense https://scigraph.springernature.com/explorer/license/
67 schema:sdPublisher N0f18a3dec8134f6fbfab2954ee7a4081
68 schema:url https://link.springer.com/10.1186%2Fs12874-019-0673-4
69 sgo:license sg:explorer/license/
70 sgo:sdDataset articles
71 rdf:type schema:ScholarlyArticle
72 N016a753e5d0b4a0a9c61b9e92ebb673f schema:volumeNumber 19
73 rdf:type schema:PublicationVolume
74 N0f18a3dec8134f6fbfab2954ee7a4081 schema:name Springer Nature - SN SciGraph project
75 rdf:type schema:Organization
76 N22852cb5d39f49f19ac51e7aa6a30344 rdf:first N934ff58d63b7408baa6e8985e7730aff
77 rdf:rest Nc1c8cb8dab3740f0b6eb4482cc2eb177
78 N319e62d2ceb7473fbb469546cae18d5b rdf:first N7bad25665c2f407a9e1e8a74484526a4
79 rdf:rest N630ee955cd8c48d88c7c728318d8cf33
80 N3a6c20aec8f8401697307161f1c774c3 rdf:first Nf74718baf7af4cb0bcc14821a62cf6a7
81 rdf:rest N319e62d2ceb7473fbb469546cae18d5b
82 N4b5d5e12e80a455187f308e3166c74f1 schema:name doi
83 schema:value 10.1186/s12874-019-0673-4
84 rdf:type schema:PropertyValue
85 N61eb9782b5a44c65a424ee3acf9037c8 schema:affiliation https://www.grid.ac/institutes/grid.462844.8
86 schema:familyName Bussy
87 schema:givenName Simon
88 rdf:type schema:Person
89 N6286e3a917824a9ea3e9e5b69946c9f8 schema:name nlm_unique_id
90 schema:value 100968545
91 rdf:type schema:PropertyValue
92 N630ee955cd8c48d88c7c728318d8cf33 rdf:first N7ac4613c861c42c5a11a50283fe1e974
93 rdf:rest rdf:nil
94 N68b08eade4f049cd965019ecdc34565c rdf:first Nd82b1f2e786143acaf8665a2445d823c
95 rdf:rest N3a6c20aec8f8401697307161f1c774c3
96 N6ca014ef4419469590a24d8b0e8844a0 schema:name readcube_id
97 schema:value aedaa9aa4d69af21e7b83a8d7d5038c7fcffa24282fbed0b62442ad52759ed12
98 rdf:type schema:PropertyValue
99 N72b2bea0d956475a88d5b163f6821a1b schema:affiliation https://www.grid.ac/institutes/grid.417925.c
100 schema:familyName Veil
101 schema:givenName Raphaël
102 rdf:type schema:Person
103 N72c5daf5a20d4b7ca593031b19459b58 rdf:first N61eb9782b5a44c65a424ee3acf9037c8
104 rdf:rest Nce7b44ff2aad41d1961d658b2086704e
105 N7ac4613c861c42c5a11a50283fe1e974 schema:affiliation https://www.grid.ac/institutes/grid.417925.c
106 schema:familyName Jannot
107 schema:givenName Anne-Sophie
108 rdf:type schema:Person
109 N7bad25665c2f407a9e1e8a74484526a4 schema:affiliation https://www.grid.ac/institutes/grid.414093.b
110 schema:familyName Ranque
111 schema:givenName Brigitte
112 rdf:type schema:Person
113 N7cd7a1b4f7da405ba3957a1b824d485d schema:affiliation https://www.grid.ac/institutes/grid.417925.c
114 schema:familyName Burgun
115 schema:givenName Anita
116 rdf:type schema:Person
117 N934ff58d63b7408baa6e8985e7730aff schema:affiliation https://www.grid.ac/institutes/grid.417925.c
118 schema:familyName Looten
119 schema:givenName Vincent
120 rdf:type schema:Person
121 Nc1c8cb8dab3740f0b6eb4482cc2eb177 rdf:first N7cd7a1b4f7da405ba3957a1b824d485d
122 rdf:rest N68b08eade4f049cd965019ecdc34565c
123 Nce7b44ff2aad41d1961d658b2086704e rdf:first N72b2bea0d956475a88d5b163f6821a1b
124 rdf:rest N22852cb5d39f49f19ac51e7aa6a30344
125 Nd82b1f2e786143acaf8665a2445d823c schema:affiliation https://www.grid.ac/institutes/grid.462265.1
126 schema:familyName Gaïffas
127 schema:givenName Stéphane
128 rdf:type schema:Person
129 Nd9ee7d4f3f314fa38c09a740b84ad6cb schema:name pubmed_id
130 schema:value 30841867
131 rdf:type schema:PropertyValue
132 Nf74718baf7af4cb0bcc14821a62cf6a7 schema:affiliation https://www.grid.ac/institutes/grid.4444.0
133 schema:familyName Guilloux
134 schema:givenName Agathe
135 rdf:type schema:Person
136 Nff9d329ef73441778dda90dbc9c9b170 schema:name dimensions_id
137 schema:value pub.1112585404
138 rdf:type schema:PropertyValue
139 Nffc081f93cce47e39fa1b266025ef8ef schema:issueNumber 1
140 rdf:type schema:PublicationIssue
141 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
142 schema:name Mathematical Sciences
143 rdf:type schema:DefinedTerm
144 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
145 schema:name Statistics
146 rdf:type schema:DefinedTerm
147 sg:journal.1024940 schema:issn 1471-2288
148 schema:name BMC Medical Research Methodology
149 rdf:type schema:Periodical
150 sg:pub.10.1007/978-1-4419-6646-9 schema:sameAs https://app.dimensions.ai/details/publication/pub.1028948927
151 https://doi.org/10.1007/978-1-4419-6646-9
152 rdf:type schema:CreativeWork
153 sg:pub.10.1007/s00439-008-0592-7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1003137479
154 https://doi.org/10.1007/s00439-008-0592-7
155 rdf:type schema:CreativeWork
156 sg:pub.10.1007/s10115-006-0040-8 schema:sameAs https://app.dimensions.ai/details/publication/pub.1018091708
157 https://doi.org/10.1007/s10115-006-0040-8
158 rdf:type schema:CreativeWork
159 sg:pub.10.1023/a:1010933404324 schema:sameAs https://app.dimensions.ai/details/publication/pub.1024739340
160 https://doi.org/10.1023/a:1010933404324
161 rdf:type schema:CreativeWork
162 sg:pub.10.1186/1471-2105-10-213 schema:sameAs https://app.dimensions.ai/details/publication/pub.1002419282
163 https://doi.org/10.1186/1471-2105-10-213
164 rdf:type schema:CreativeWork
165 sg:pub.10.1186/1471-2288-12-100 schema:sameAs https://app.dimensions.ai/details/publication/pub.1000235299
166 https://doi.org/10.1186/1471-2288-12-100
167 rdf:type schema:CreativeWork
168 sg:pub.10.1186/1471-2288-12-102 schema:sameAs https://app.dimensions.ai/details/publication/pub.1003624527
169 https://doi.org/10.1186/1471-2288-12-102
170 rdf:type schema:CreativeWork
171 sg:pub.10.1186/1471-2288-9-85 schema:sameAs https://app.dimensions.ai/details/publication/pub.1030193087
172 https://doi.org/10.1186/1471-2288-9-85
173 rdf:type schema:CreativeWork
174 sg:pub.10.1186/s12874-016-0128-0 schema:sameAs https://app.dimensions.ai/details/publication/pub.1008888493
175 https://doi.org/10.1186/s12874-016-0128-0
176 rdf:type schema:CreativeWork
177 https://app.dimensions.ai/details/publication/pub.1028948927 schema:CreativeWork
178 https://app.dimensions.ai/details/publication/pub.1078219311 schema:CreativeWork
179 https://app.dimensions.ai/details/publication/pub.1078354293 schema:CreativeWork
180 https://app.dimensions.ai/details/publication/pub.1106875674 schema:CreativeWork
181 https://doi.org/10.1001/jama.2010.378 schema:sameAs https://app.dimensions.ai/details/publication/pub.1008462250
182 rdf:type schema:CreativeWork
183 https://doi.org/10.1001/jama.2011.1561 schema:sameAs https://app.dimensions.ai/details/publication/pub.1007755515
184 rdf:type schema:CreativeWork
185 https://doi.org/10.1002/9781118150672 schema:sameAs https://app.dimensions.ai/details/publication/pub.1106892893
186 rdf:type schema:CreativeWork
187 https://doi.org/10.1002/9781118548387 schema:sameAs https://app.dimensions.ai/details/publication/pub.1106875674
188 rdf:type schema:CreativeWork
189 https://doi.org/10.1002/pbc.21854 schema:sameAs https://app.dimensions.ai/details/publication/pub.1009687810
190 rdf:type schema:CreativeWork
191 https://doi.org/10.1016/s0031-3203(96)00142-2 schema:sameAs https://app.dimensions.ai/details/publication/pub.1044522995
192 rdf:type schema:CreativeWork
193 https://doi.org/10.1016/s0167-9473(01)00065-2 schema:sameAs https://app.dimensions.ai/details/publication/pub.1012543995
194 rdf:type schema:CreativeWork
195 https://doi.org/10.1021/ci0342472 schema:sameAs https://app.dimensions.ai/details/publication/pub.1055401738
196 rdf:type schema:CreativeWork
197 https://doi.org/10.1046/j.1365-2141.2003.04193.x schema:sameAs https://app.dimensions.ai/details/publication/pub.1012960136
198 rdf:type schema:CreativeWork
199 https://doi.org/10.1056/nejm199107043250103 schema:sameAs https://app.dimensions.ai/details/publication/pub.1030040630
200 rdf:type schema:CreativeWork
201 https://doi.org/10.1056/nejm199511023331806 schema:sameAs https://app.dimensions.ai/details/publication/pub.1010799486
202 rdf:type schema:CreativeWork
203 https://doi.org/10.1056/nejm199709113371107 schema:sameAs https://app.dimensions.ai/details/publication/pub.1028319673
204 rdf:type schema:CreativeWork
205 https://doi.org/10.1073/pnas.0401736101 schema:sameAs https://app.dimensions.ai/details/publication/pub.1006123154
206 rdf:type schema:CreativeWork
207 https://doi.org/10.1093/biomet/69.3.553 schema:sameAs https://app.dimensions.ai/details/publication/pub.1059419215
208 rdf:type schema:CreativeWork
209 https://doi.org/10.1093/biomet/79.3.531 schema:sameAs https://app.dimensions.ai/details/publication/pub.1059420291
210 rdf:type schema:CreativeWork
211 https://doi.org/10.1097/aog.0b013e318160f38e schema:sameAs https://app.dimensions.ai/details/publication/pub.1015539591
212 rdf:type schema:CreativeWork
213 https://doi.org/10.1109/tnn.2005.848998 schema:sameAs https://app.dimensions.ai/details/publication/pub.1018563512
214 rdf:type schema:CreativeWork
215 https://doi.org/10.1111/j.0006-341x.2005.030814.x schema:sameAs https://app.dimensions.ai/details/publication/pub.1044144954
216 rdf:type schema:CreativeWork
217 https://doi.org/10.1111/j.1467-9868.2005.00503.x schema:sameAs https://app.dimensions.ai/details/publication/pub.1043971564
218 rdf:type schema:CreativeWork
219 https://doi.org/10.1111/j.1532-5415.1990.tb03450.x schema:sameAs https://app.dimensions.ai/details/publication/pub.1001041978
220 rdf:type schema:CreativeWork
221 https://doi.org/10.1111/j.2517-6161.1972.tb00899.x schema:sameAs https://app.dimensions.ai/details/publication/pub.1110457843
222 rdf:type schema:CreativeWork
223 https://doi.org/10.1136/bmj.313.7057.628 schema:sameAs https://app.dimensions.ai/details/publication/pub.1062777078
224 rdf:type schema:CreativeWork
225 https://doi.org/10.1177/0962280218766389 schema:sameAs https://app.dimensions.ai/details/publication/pub.1103284945
226 rdf:type schema:CreativeWork
227 https://doi.org/10.1177/1077558704263799 schema:sameAs https://app.dimensions.ai/details/publication/pub.1053805691
228 rdf:type schema:CreativeWork
229 https://doi.org/10.1177/183335830303100103 schema:sameAs https://app.dimensions.ai/details/publication/pub.1064077667
230 rdf:type schema:CreativeWork
231 https://doi.org/10.18637/jss.v039.i05 schema:sameAs https://app.dimensions.ai/details/publication/pub.1068672583
232 rdf:type schema:CreativeWork
233 https://doi.org/10.2202/1544-6115.1147 schema:sameAs https://app.dimensions.ai/details/publication/pub.1041046514
234 rdf:type schema:CreativeWork
235 https://doi.org/10.2307/2529885 schema:sameAs https://app.dimensions.ai/details/publication/pub.1069975475
236 rdf:type schema:CreativeWork
237 https://doi.org/10.2307/2982890 schema:sameAs https://app.dimensions.ai/details/publication/pub.1070160732
238 rdf:type schema:CreativeWork
239 https://doi.org/10.2307/3001968 schema:sameAs https://app.dimensions.ai/details/publication/pub.1102728208
240 rdf:type schema:CreativeWork
241 https://www.grid.ac/institutes/grid.414093.b schema:alternateName Hôpital Européen Georges-Pompidou
242 schema:name Assistance Publique-Hôpitaux de Paris, Internal Medicine Department, Georges Pompidou European Hospital, 20 Rue Leblanc, 75015, Paris, France
243 INSERM UMRS 970, Université Paris Descartes, 56 rue Leblanc, 75015, Paris, France
244 rdf:type schema:Organization
245 https://www.grid.ac/institutes/grid.417925.c schema:alternateName Centre de Recherche des Cordeliers
246 schema:name Assistance Publique-Hôpitaux de Paris, Biomedical Informatics and Public Health Department, European Georges Pompidou Hospital, 20 Rue Leblanc, 75015, Paris, France
247 INSERM UMRS 1138, Eq22, Centre de Recherche des Cordeliers, Université Paris Descartes, 15 Rue de l’École de Médecine, 75006, Paris, France
248 rdf:type schema:Organization
249 https://www.grid.ac/institutes/grid.4444.0 schema:alternateName French National Centre for Scientific Research
250 schema:name LAMME, Univ Evry, CNRS, Université Paris-Saclay, 23 boulevard de France, 91025, Evry, France
251 rdf:type schema:Organization
252 https://www.grid.ac/institutes/grid.462265.1 schema:alternateName Centre de Mathématiques Appliquées
253 schema:name CMAP, UMR 7641 École Polytechnique CNRS, Route de Saclay, 91128, Palaiseau, France
254 Laboratoire de Probabilités Statistique et Modélisation (LPSM), UMR 8001, Sorbonne University, 4 Place Jussieu, 75005, Paris, France
255 rdf:type schema:Organization
256 https://www.grid.ac/institutes/grid.462844.8 schema:alternateName Sorbonne University
257 schema:name Laboratoire de Probabilités Statistique et Modélisation (LPSM), UMR 8001, Sorbonne University, 4 Place Jussieu, 75005, Paris, France
258 rdf:type schema:Organization
 




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


...