Skip to main content
. 2023 Sep 22;16:77–81. doi: 10.1016/j.sopen.2023.09.013

Table 3.

Example patient data: in the first column, we present a list of variables used in the ML model; in the second column, we present the raw value for that variable, with NaN indicating missing data for that patient; in the third column, we present the continuous age-normalized score. Variables marked with an asterisk are not converted to deviance scores as they are equally informative for children and adults and are simply normalized at the time of training. Variables that have been bolded were transformed using the continuous SOFA transformations.

Raw value Age-normalized score
Age 1.1 yrs 1.1 yrs
C Reactive protein_mean 1.86 2.32
Creatinine_mean 1.58484163 4.28280543
Fshunt_mean* 5.7 5.7
MAP_max 53 1.12
MAP_mean 53 1.12
MAP_median 53 1.12
MAP_min 53 1.12
Oxygen Saturation_max* 100 100
Oxygen Saturation_mean* 98.4 98.4
Oxygen Saturation_median* 98.4 98.4
Oxygen Saturation_min* 96.8 96.8
Pulse_max NaN NaN
Pulse_mean NaN NaN
Pulse_median NaN NaN
Pulse_min NaN NaN
Pulse_std NaN NaN
Respiratory Rate_max 70 0.0008329
Respiratory Rate_mean 69 0.00074961
Respiratory Rate_median 69 0.00074961
Respiratory Rate_min 68 0.00066632
Temperature_max* 37 37
Temperature_mean* 36.8333333 36.8333333
Temperature_median* 37 37
Temperature_min* 36.5 36.5
Total Bilirubin_mean* 14.55 5
White Blood Cells_mean* NaN NaN
p02_max* 173 173
p02_mean* 124 124
p02_median* 124 124
p02_min* 75 75
paCO2_mean* 42.45 42.45