Fig. 3: Impact of the proposed model on clinical practice for patients with suspicion of AIS in the emergency department. | Nature Communications

Fig. 3: Impact of the proposed model on clinical practice for patients with suspicion of AIS in the emergency department.

From: A clinically applicable deep-learning model for detecting intracranial aneurysm in computed tomography angiography images

Fig. 3: Impact of the proposed model on clinical practice for patients with suspicion of AIS in the emergency department.The alternative text for this image may have been generated using AI.

In the cohort of patients with suspected AIS (Internal cohort 5), who were prescribed to perform head CTA examination, 86.8% patients diagnosed as aneurysm-negative cases by our model, among which 99.0% were true-negative, demonstrating high confidence in identifying negative cases by our model. As a result, only 13.2% of patients were categorized as high-risk, to whom the radiologists can pay more-intense attention and reduce their workload in detecting aneurysm in AIS patients. AIS acute ischemic stroke, IA intracranial aneurysm.

Back to article page