Skip to main content

Table 1 Variables collected for algorithm building

From: Predicting acute kidney injury with an artificial intelligence-driven model in a pediatric cardiac intensive care unit

Type of variable

Name of variable

Admission and post-admission data (fixed data)

Sex, age, weight, PIM3, VIS, admission type, basal SCr, CPB duration, cross-clamp duration

Vital signs

SAP, DAP, MAP, O2 saturation, HR

Fluids

Diuresis, fluid input, fluid output, blood input, blood output

Blood gas analysis

BE, Na + , Cl-, Lactate, blood pH

Laboratory analysis

SCr, albumin, hemoglobin, platelets, LDH, aPTTs

Therapies

Adrenaline, milrinone, furosemide, levosimendan, vasopressin, ethacrynic acid

  1. Admission and post-admission data and Fluids are manually inserted into the electronic health record (EHR) by clinicians and nurses. Vital signs, Blood gas analysis, and Laboratory analysis are automatically collected from monitors and labs into the EHR. Therapies are prescribed by clinicians and validated by nurses. Once they are validated, data are automatically extracted from the EHR
  2. PIM3 Paediatric Index of Mortality3 score, VIS Vasoactive Inotropic Score, Admission type post-surgical/post cath lab/medical admission, CPB cardiopulmonary bypass, SCr serum creatinine, SAP systolic arterial pressure, DAP diastolic arterial pressure, MAP mean arterial pressure, HR heart rate, BE base excess, LDH lactic dehydrogenase, aPTTs activated partial thromboplastin time in seconds
Back to article page

Journal of Anesthesia, Analgesia and Critical Care

ISSN: 2731-3786

Contact us