clinical study

The Effect of Artificial Intelligence-Based Intracranial Hemorrhage Detection on Study Turnaround Time for Emergent Care Non- Contrast Head CT – A Prospective Randomized Clinical Trial

Materials & Methods

A total of 1,936 consecutive non-contrast head CT scans from two CT scanners used for inpatient and emergency room patients at a large academic hospital were prospectively acquired over a time period of 47 consecutive days. Study turnaround time (TAT) was measured automatically as the time difference between study completion time (=study accessible to radiologists for reporting) to study reporting time (=first report visible to clinicians, regardless of whether preliminary or final).


A total of 381 ICH-AI+ cases were found, of which 190 cases were flagged. TATs for flagged cases (52 ± 25 min.) were significantly lower than TATs for non-flagged (82 ± 31 min.) cases (p<0.05, Wilcoxon signed-rank test). 


Notifying radiologists on automatically detected ICH statistically significantly reduces TAT for reporting ICH to clinicians in emergency setting. Reduced TAT may expedite clinically indicated therapeutic interventions.