Dataset Open Access

Development, evaluation, and validation of machine learning models for COVID-19 detection based on routine blood tests

Cabitza, Federico; Campagner, Andrea; Ferrari, Davide; Di Resta, Chiara; Ceriotti, Daniele; Sabetta, Eleonora; Colombini, Alessandra; De Vecchi, Elena; Banfi, Giuseppe; Locatelli, Massimo; Carobene, Anna


MARC21 XML Export

<?xml version='1.0' encoding='UTF-8'?>
<record xmlns="http://www.loc.gov/MARC21/slim">
  <leader>00000nmm##2200000uu#4500</leader>
  <controlfield tag="005">20210414101611.0</controlfield>
  <controlfield tag="001">4081318</controlfield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">IRCCS Istituto Ortopedico Galeazzi, Orthopaedic Biotechnology Lab, Via Riccardo Galeazzi, 4, 20161, Milano, Italy</subfield>
    <subfield code="a">Campagner, Andrea</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">SCVSA Department, University of Parma, Parco Area delle Science 11/a, 43124, Parma, Italy</subfield>
    <subfield code="a">Ferrari, Davide</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Vita-Salute San Raffaele University; Unit of Genomics for Human Disease Diagnosis, Division of Genetics and Cell Biology., Via Olgettina 58, 20132, Milan, Italy</subfield>
    <subfield code="a">Di Resta, Chiara</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Laboratory Medicine, IRCCS San Raffaele Scientific Institute, Via Olgettina 60, 20132, Milan, Italy</subfield>
    <subfield code="a">Ceriotti, Daniele</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Laboratory Medicine, IRCCS San Raffaele Scientific Institute, Via Olgettina 60, 20132, Milan, Italy</subfield>
    <subfield code="a">Sabetta, Eleonora</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">IRCCS Istituto Ortopedico Galeazzi, Orthopaedic Biotechnology Lab, Via Riccardo Galeazzi, 4, 20161, Milano, Italy</subfield>
    <subfield code="a">Colombini, Alessandra</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">IRCCS Istituto Ortopedico Galeazzi, Orthopaedic Biotechnology Lab, Via Riccardo Galeazzi, 4, 20161, Milano, Italy</subfield>
    <subfield code="a">De Vecchi, Elena</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">IRCCS Istituto Ortopedico Galeazzi, Orthopaedic Biotechnology Lab, Via Riccardo Galeazzi, 4, 20161, Milano, Italy</subfield>
    <subfield code="a">Banfi, Giuseppe</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Laboratory Medicine, IRCCS San Raffaele Scientific Institute, Via Olgettina 60, 20132, Milan, Italy</subfield>
    <subfield code="a">Locatelli, Massimo</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">Laboratory Medicine, IRCCS San Raffaele Scientific Institute, Via Olgettina 60, 20132, Milan, Italy</subfield>
    <subfield code="a">Carobene, Anna</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">303721</subfield>
    <subfield code="z">md5:a0a7a04a687c019a17330c6b9f97fc66</subfield>
    <subfield code="u">https://zenodo.org/record/4081318/files/all_training.xlsx</subfield>
  </datafield>
  <datafield tag="542" ind1=" " ind2=" ">
    <subfield code="l">open</subfield>
  </datafield>
  <datafield tag="260" ind1=" " ind2=" ">
    <subfield code="c">2020-10-12</subfield>
  </datafield>
  <datafield tag="909" ind1="C" ind2="O">
    <subfield code="p">openaire_data</subfield>
    <subfield code="p">user-covid-19</subfield>
    <subfield code="o">oai:zenodo.org:4081318</subfield>
  </datafield>
  <datafield tag="100" ind1=" " ind2=" ">
    <subfield code="u">DISCo, Università degli Studi di Milano-Bicocca, Viale Sarca 336, Milano, 20126, Italy</subfield>
    <subfield code="a">Cabitza, Federico</subfield>
  </datafield>
  <datafield tag="245" ind1=" " ind2=" ">
    <subfield code="a">Development, evaluation, and validation of machine learning models for COVID-19 detection based on routine blood tests</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">user-covid-19</subfield>
  </datafield>
  <datafield tag="540" ind1=" " ind2=" ">
    <subfield code="u">https://creativecommons.org/licenses/by/4.0/legalcode</subfield>
    <subfield code="a">Creative Commons Attribution 4.0 International</subfield>
  </datafield>
  <datafield tag="650" ind1="1" ind2="7">
    <subfield code="a">cc-by</subfield>
    <subfield code="2">opendefinition.org</subfield>
  </datafield>
  <datafield tag="520" ind1=" " ind2=" ">
    <subfield code="a">&lt;p&gt;The .xlsx dataset includes all patients used for training, internal-external and external validation: these can be distinguished by looking at the ID (first column) in the dataset: those in format Axxxx-&amp;lt;Date&amp;gt; are the data used for the training, those in the format 20xx are the data used for the internal-external validation, while the remaining data were used for external validation.&lt;/p&gt;

&lt;p&gt;As regards the features: for the Target feature the value 1 stands for &amp;quot;Positive to COVID-19&amp;quot; while the value 0 stands for &amp;quot;Negative to COVID-19&amp;quot;; while for the Sex feature the value 1 stands for &amp;quot;Male&amp;quot; while the value 0 stands for &amp;quot;Female&amp;quot;.&lt;/p&gt;

&lt;p&gt;The full article is available at: https://www.degruyter.com/view/journals/cclm/ahead-of-print/article-10.1515-cclm-2020-1294/article-10.1515-cclm-2020-1294.xml.&lt;/p&gt;

&lt;p&gt;A pre-print version of the article is also available on MedrXiv:&amp;nbsp;https://www.medrxiv.org/content/10.1101/2020.10.02.20205070v1&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;ABSTRACT&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Background&lt;/strong&gt; The rRT-PCR test, the current gold standard for the detection of coronavirus disease (COVID-19),&amp;nbsp;presents with known shortcomings, such as long turnaround time, potential shortage of reagents, false-negative&amp;nbsp;rates around 15&amp;ndash;20%, and expensive equipment. The hematochemical values of routine blood exams could&amp;nbsp;represent a faster and less expensive alternative.&amp;nbsp;&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Methods&lt;/strong&gt; Three different training data set of hematochemical values from 1,624 patients (52% COVID-19&amp;nbsp;positive), admitted at San Raphael Hospital (OSR) from February to May 2020, were used for developing machine&amp;nbsp;learning (ML) models: the complete OSR dataset (72 features: complete blood count (CBC), biochemical,&amp;nbsp;coagulation, hemogasanalysis and CO-Oxymetry values, age, sex and specific symptoms at triage) and two sub&amp;nbsp;datasets (COVID-specific and CBC dataset, 32 and 21 features respectively). 58 cases (50% COVID-19 positive)&amp;nbsp;from another hospital, and 54 negative patients collected in 2018 at OSR, were used for internal-external and external validation.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Results&lt;/strong&gt; We developed five ML models: for the complete OSR dataset, the area under the receiver operating&amp;nbsp;characteristic curve (AUC) for the algorithms ranged from 0.83 to 0.90; for the COVID-specific dataset from 0.83 15 to 0.87; and for the CBC dataset from 0.74 to 0.86. The validations also achieved good results: respectively, AUC 16 from 0.75 to 0.78; and specificity from 0.92 to 0.96.&amp;nbsp;&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Conclusions&lt;/strong&gt; ML can be applied to blood tests as both an adjunct and alternative method to rRT-PCR for the fast&amp;nbsp;and cost-effective identification of COVID-19-positive patients. This is especially useful in developing countries,&amp;nbsp;or in countries facing an increase in contagions.&lt;/p&gt;</subfield>
  </datafield>
  <datafield tag="024" ind1=" " ind2=" ">
    <subfield code="a">10.1515/cclm-2020-1294</subfield>
    <subfield code="2">doi</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">dataset</subfield>
  </datafield>
</record>
1,066
474
views
downloads
Views 1,066
Downloads 474
Data volume 144.0 MB
Unique views 962
Unique downloads 394

Share

Cite as