[Progress in computer-assisted Alberta stroke program early computer tomography score of acute ischemic stroke based...

Area of Science:

• Neurology
• Radiology
• Medical Imaging Analysis

Background:

• Non-contrastive computed tomography (NCCT) is crucial for diagnosing stroke types and guiding treatment via the Alberta Stroke Program Early CT Score (ASPECTS).
• Early-stage acute ischemic stroke (AIS) presents diagnostic challenges on NCCT due to subtle infarction and indistinct brain region boundaries, complicating accurate ASPECTS assessment.
• Clinical ASPECTS scoring often suffers from inter-observer variability.

Purpose of the Study:

• To review the challenges in clinical ASPECTS assessment for AIS.
• To summarize the application of machine learning and deep learning in computer-aided ASPECTS.
• To explore future research directions for AIS-assisted assessment.

Main Methods:

• Review of current literature on machine learning and deep learning techniques applied to AIS ASPECTS.
• Analysis of how these technologies address challenges in identifying cerebral infarction and segmenting brain regions.
• Discussion of quantitative scoring capabilities offered by AI.

Main Results:

• Machine learning and deep learning methods demonstrate potential for accurate and rapid identification of cerebral infarction areas and quantitative ASPECTS scoring.
• Computer-aided systems can significantly reduce inconsistency in clinical ASPECTS evaluations.
• Multi-modal imaging-based AI systems offer enhanced comprehensiveness and accuracy in AIS assessment.

Conclusions:

• AI-powered tools are vital for improving the accuracy and consistency of ASPECTS in AIS.
• Future research should focus on developing multi-modal imaging-based computer-aided systems for AIS assessment.
• These advancements hold promise for establishing new research avenues in AIS-assisted diagnosis and management.