[A computerized method for detection of acute cerebral infarction on CT images]

Area of Science:

• Medical Imaging
• Computer-Aided Diagnosis
• Neurology

Background:

• Acute cerebral infarction (ACI) detection on CT images is crucial for timely treatment.
• Existing methods may lack the precision required for early and accurate diagnosis.

Purpose of the Study:

• To develop and evaluate a computerized method for automated detection of ACI on CT images.
• To assess the performance of the proposed method in terms of sensitivity and false positive rates.

Main Methods:

• Brain parenchyma segmentation using thresholding after image correction.
• Selection of regions-of-interest (ROIs) in the middle cerebral artery (MCA) region.
• Feature extraction using statistical, co-occurrence, and run length matrices.
• Classification of ROIs with ACI using linear discriminant analysis based on feature differences in symmetrical ROIs.

Main Results:

• The computerized method achieved a sensitivity of 88.0% for ACI detection.
• An average of 4.6 false positives per case was recorded.
• The method demonstrated relatively high performance in detecting ACI.

Conclusions:

• The proposed computerized method shows significant potential for the automated detection of ACI on CT images.
• This approach could serve as a valuable algorithm for computer-aided diagnosis systems.
• Further validation may enhance its clinical applicability in stroke detection.