Accuracy of ICH volumetry on NCCT​ - 97% sensitivity in Swedish Stroke Registry

A total of 1649 stroke patients were included. The study showed excellent agreement in volume quantification between the fully automated software qER–NCCT and manual segmentation of ICH on NCCT. The qER–NCCT would be an important additive tool by aiding in early diagnostics and prognostication for patients with ICH and in providing volumetry on a population-wide level. Further refinement of the software should address the underestimation of ICH volume seen in a portion of large, heterogenous, irregularly shaped ICHs.

AI Detection of critical findings on NCCT​ - Over 90% AUC for all findings​

The study was conducted on 21 095 scans, wherein the algorithm achieved an AUC of 0·92 for detecting intracranial hemorrhage (0·90 [0·89–0·91] for intraparenchymal, 0·96 [0·94–0·97] for intraventricular, 0·92 [0·90–0·93] for subdural, 0·93 [0·91–0·95] for extradural, and 0·90 [0·89–0·92] for subarachnoid). AUCs were 0·92 (0·91–0·94) for calvaria fractures, 0·93 (0·91–0·94) for midline shift, and 0·86 (0·85–0·87) for mass effect. This proves that the deep learning algorithms can accurately identify head CT scan abnormalities requiring urgent attention, opening up the possibility to use these algorithms to automate the triage process