Table 5.

Performance comparison between Final, INRG risk factors and INRG consensus pretreatment schema classifiers

Classifier	Accuracy a	Recall b	Precision c	Specificity d	Negative Predictive Value e
Final f (Good vs Poor)	86%	91%	90%	70%	70%
INRG risk factors g (Good vs Poor)	82%	85%	92%	70%	53%
INRG Pretreatment Schema h (VLR/LR/IR vs HR)	84%	83%	97%	80%	56%

^a Accuracy is the fraction of correctly classified patients and overall classified patients.

^b Recall is the fraction of correctly classified good outcome patients and the overall predicted good outcome patients.

^C Precision is the fraction of correctly classified good outcome patients and the predicted good outcome patients.

^d Specificity is the fraction of correctly classified poor outcome patients and the overall poor outcome patients.

^e Negative predictive value is the fraction of correctly classified poor outcome patients and the overall predicted poor outcome patients.

^f Final classifier indicates the classifier in Table 4 obtained utilizing Rulex.

^g INRG risk factors classifier indicates the classifier in Table 2 obtained utilizing Rulex.

^h INRG Pretreatment Schema classifier indicates the INRG pretreatment classification schema.

VLR = Very Low Risk; HR = High Risk; IR = Intermediate Risk; LR = Low Risk. We computed the performance of INRG pretreatment schema classifier on the 51 patients' cohort as follows.

We assigned Very low, Low, Intermediate, High risk utilizing the INRG consensus pretreatment classification schema considering only the Age at diagnosis, INSS stage and MYCN status risk factors.

We then created two groups to make the risk assignment comparable with outcome.

The first included Very low, Low and Intermediate risk patients and the second included High risk patients.

The first group was associated with a favorable outcome and the second was associated with an unfavourable outcome.