Predicting Mortality of ICU Patients: The PhysioNet/Computing in Cardiology Challenge 2012 1.0.0
(686 bytes)
function [S]=event2score(data)
%[S]=event2score(data) calculates minimum of the SE and PPV
%
%data -(Nx2) where N is the number of cases (4000 for each of sets A, B, and C),
% the first column is the true outcome (data(:,2) from the Outcomes file),
% and the second column is the estimated risk (from the challenge
% entry outputs for event2)
%
% Calculate sensitivity (Se) and positive predictivity (PPV)
TP=sum(data(data(:,2)==1,1));
FN=sum(~data(data(:,2)==1,1));
FP=sum(data(data(:,2)==0,1));
Se=TP/(TP+FN);
PPV=TP/(TP+FP);
% The event 1 score is the smaller of Se and PPV.
S = min(Se, PPV);
end