Intracranial ring lesions: can we differentiate by computed tomography?

Area of Science:

• * Neurology and Radiology
• * Medical Imaging Analysis

Background:

• * Distinguishing between brain abscess, primary/metastatic brain tumors, and cerebral vascular accidents (CVAs) is critical for effective patient management.
• * Ring-enhancing lesions on medical imaging can present similarly across these diverse pathologies, posing diagnostic challenges.

Purpose of the Study:

• * To evaluate the efficacy of using patient age, specific ring lesion image parameters, and mathematical algorithms in differentiating brain abscess, brain tumor, and CVA.
• * To identify key imaging parameters and their diagnostic contribution for each condition.

Main Methods:

• * Analysis of eleven patients with ring-enhancing brain lesions.
• * Utilized patient age, ten image parameters, Bayes theorem, and linear discriminant function analysis.
• * Identified a subset of six parameters for classification and explored disease-specific parameter subsets.

Main Results:

• * Overall classification accuracy reached 86%.
• * Individual disease classification accuracies: brain abscess (84%), brain tumor (96%), and CVA (17%).
• * Key differentiating parameters included ring thickness variability, patient age, outside ring diameter, CT numbers in ring center, maximum ring thickness, and lesion-to-ring ratio.

Conclusions:

• * A combination of patient age and specific imaging parameters, analyzed with mathematical algorithms, can effectively differentiate brain abscess and tumors.
• * The model showed lower accuracy for CVA, highlighting the need for further refinement for this condition.
• * Identified disease-specific parameters can guide further research and clinical investigation into lesion characteristics.