Join an AFSMC mission to Israel and show your support and love
Donate
The study demonstrated that machine learning analysis of head computed tomography (CT) scans upon admission to the emergency department can enable earlier detection of ICH, reducing the overall mortality rate associated with the haemorrhage, a press release states.
Along with Sheba neurology and radiology experts, David Orion (Sheba Medical Center, Tel Aviv, Israel) conducted the retrospective cohort study across 587 patients with a confirmed diagnosis of ICH in a Level 1 trauma centre.
The study compared 289 patients between January 2017 and January 2018 who did not have their CT scan analysed by AI, to 298 patients between January 2019 and January 2020 who had an AI analysis, examining the impact on 30- and 120-day all-cause mortality in both groups.
Findings from the trial demonstrated that AI analysis of CT scans drove earlier detection of ICH, enabling physicians to begin administering therapeutic interventions sooner. And, as a result of these early interventions, the study demonstrated a relative decrease of more than 30% in the rate of absolute mortality in the AI-scanned cohort.
“Intracerebral haemorrhages are one of the most critical medical conditions in emergency care, with early detection vital for preventing loss of life and further morbidities,” said Orion. “This study shows the tremendous impact of AI to drive improved patient outcomes, helping to save lives and improve overall quality of life.”
The study also examined the impact of AI analysis and early treatment initiation on patient comorbidities, and found that patients in the AI cohort had a significantly lower modified Rankin scale (mRS) score—a measure of the degree of disability or dependence in the daily activities of patients with neurological disabilities—at discharge than those in the non-AI group.
Published September 21, 2023, NeuroNews
