Risk Stratification with Extreme Learning Machine: A Retrospective Study on Emergency Department Patients

<table>The distance based scoring system (DIST) triage architecture. Firstly, the selected variables are converted into interval <svg height="11.7327pt" id="M1" style="vertical-align:-2.208321pt" version="1.1" viewbox="0 -9.52438 35.6673 11.7327" width="35.6673pt" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink"><g transform="matrix(.0135,0,0,-0.0135,0,0)"><path d="M290 -163V-135C183 -126 181 -122 181 -44V583C181 662 184 666 290 675V703H120V-163H290Z" id="g113-92"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="345" vert-adv-y="345"></glyph.data></g><g transform="matrix(.0135,0,0,-0.0135,4.674,0)"><path d="M535 230V280H52V230H535Z" id="g117-33"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="587" vert-adv-y="587"></glyph.data></g><g transform="matrix(.0135,0,0,-0.0135,12.627,0)"><path d="M384 0V27C293 34 287 42 287 114V635C232 613 172 594 109 583V559L157 557C201 555 205 550 205 499V114C205 42 199 34 109 27V0H384Z" id="g113-50"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="480" vert-adv-y="480"></glyph.data></g><g transform="matrix(.0135,0,0,-0.0135,19.13,0)"><path d="M95 130C70 130 46 113 46 88C46 72 54 64 59 64C93 55 121 33 121 -3C121 -41 93 -68 44 -88L55 -117C117 -98 186 -56 186 22C186 91 131 130 95 130Z" id="g113-45"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="228" vert-adv-y="228"></glyph.data></g><g transform="matrix(.0135,0,0,-0.0135,24.49,0)"><path d="M384 0V27C293 34 287 42 287 114V635C232 613 172 594 109 583V559L157 557C201 555 205 550 205 499V114C205 42 199 34 109 27V0H384Z" id="g113-50"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="480" vert-adv-y="480"></glyph.data></g><g transform="matrix(.0135,0,0,-0.0135,30.993,0)"><path d="M226 -163V703H56V676C162 667 165 662 165 584V-43C165 -122 162 -126 56 -136V-163H226Z" id="g113-94"></path><glyph.data ascent="3473" descent="-2876" horiz-adv-x="345" vert-adv-y="345"></glyph.data></g></svg> with min-max normalization. Secondly, cluster centers for both positive samples (patients with cardiac arrest within 72 h) and negative samples (patients without cardiac arrest within 72 h) are calculated based on Euclidean distance, and an initial score for a testing sample is derived by measuring distances between the testing sample and two cluster centers. Lastly, the classification algorithm is implemented to fine-tune the risk score.</table>

Mathematical Problems in Engineering

fig1

Figure 1

Figure 1: Risk Stratification with Extreme Learning Machine: A Retrospective Study on Emergency Department Patients 