Dr Abeyratne's paper: "Exhaustive mathematical analysis of...

Dr Abeyratne's paper: "Exhaustive mathematical analysis of simple clinical measurements for childhood pneumonia diagnosis" which appeared in the peer reviewed World Journal of Pediatrics (March 2017) gives some insight into the complexity of the process (which I oversimplified terribly). In their research, and in order to diagnose pneumonia at the high levels of sensitivity and specificity they obtained, the researchers systematically and exhaustively examined combinations of 17 clinical features using LOO-CV and data from 134 subjects. Altogether they built and tested 131 071 algorithm models, from which the best performing was selected and, I imagine, used at the time as the basis for the Resappdx model.