Download PDFOpen PDF in browser

Prediction and Analysis of Sepsis by Using Machine Learning Algorithms (XG Boost & Light GBM)

EasyChair Preprint no. 7495

10 pagesDate: February 23, 2022


Sepsis is a fatal condition that develops from blood poisoning. It occurs when the immune system attacks the body as it fights off infection. Sepsis is a medical emergency that should be treated soon as it develops. We like to frame two processing methods that are the mean processing method and the feature generation method by machine learning algorithms like XG Boost and Light GBM. These are designed to predict sepsis 6 hours in advance. XG Boost and Light GBM algorithm both play an admirable role in prediction performance (AUC:910~0.979), whereas Light GBM is the fastest acting in performance. It is powerful on multidimensional data. The key factor to predict early sepsis are WBC, platelets, and PTT.

Keyphrases: bacterial infections., feature generation method, Light GBM, linear regression, mean processing method, platelets, PTT, Sepsis, SHAP value, Systematic Inflammatory Response Syndrome ICU database, WBC, XG Boost

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
  author = {Sivasankari Kannan and Priyadharshini Subramanian and Bharathi Arivalagan and Murugeshwari Adhiappan},
  title = {Prediction and Analysis of Sepsis by Using Machine Learning Algorithms (XG Boost & Light GBM)},
  howpublished = {EasyChair Preprint no. 7495},

  year = {EasyChair, 2022}}
Download PDFOpen PDF in browser