Artificial Intelligence-Enabled Medical Analysis for Intracranial Cerebral Hemorrhage Detection and Classification

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Neurology

Background:

• Intracranial hemorrhage (ICH) is a life-threatening condition with a high mortality rate.
• Early detection and accurate classification of ICH from computed tomography (CT) images are vital for effective treatment and improved patient prognosis.
• Manual diagnosis of ICH is time-consuming, necessitating automated solutions.

Purpose of the Study:

• To introduce an AI-enabled medical analysis tool for the detection and classification of intracranial hemorrhage (ICH) using CT images.
• To develop an automated system, AIMA-ICHDC, for identifying the presence and grading of ICH.
• To enhance diagnostic accuracy and efficiency in managing ICH cases.

Main Methods:

• The study designed an AI-enabled medical analysis tool for ICH detection and classification (AIMA-ICHDC).
• Segmentation was performed using a glowworm swarm optimization with fuzzy entropy clustering (GSO-FEC) technique.
• Feature extraction utilized the VGG-19 model, and classification was achieved with an optimal mixed-kernel-based extreme learning machine (OMKELM) optimized by the coyote optimization algorithm (COA).

Main Results:

• The AIMA-ICHDC technique demonstrated effective detection and grading of ICH from single CT image samples.
• The GSO-FEC method provided efficient segmentation of ICH regions.
• Comparative analyses showed the superiority of the AIMA-ICHDC technique over existing state-of-the-art ICH classification methods.

Conclusions:

• The developed AIMA-ICHDC tool provides an accurate and efficient automated approach for ICH detection and classification.
• The integration of GSO-FEC, VGG-19, and COA-optimized OMKELM contributes to improved diagnostic performance.
• This AI-driven solution holds significant potential for enhancing clinical decision-making in managing intracranial hemorrhage.