Fig. 2From: Evaluation of techniques to improve a deep learning algorithm for the automatic detection of intracranial haemorrhage on CT head imagingROC and PR curves for the deep learning model on the CQ500 dataset. The top two graphs show the ROC curves (a intracranial haemorrhage; b each haemorrhage subtype). The bottom two graphs show the PR curves (c intracranial haemorrhage; d each haemorrhage subtype). Ninety-five per cent confidence intervals are provided in parentheses for each AUC. AUC Area under the curve, EDH Extradural haemorrhage, ICH Intracerebral haemorrhage, IVH Intraventicular haemorrhage, PR Precision-recall, ROC Receiver operating characteristic, SAH Subarachnoid haemorrhage, SDH Subdural haemorrhageBack to article page