Fast fully automatic heart fat segmentation in computed tomography datasets.

Area of Science:

• Medical Imaging
• Cardiology
• Artificial Intelligence in Medicine

Background:

• Heart diseases are a major global health concern, often linked to cardiac fat accumulation.
• Accurate segmentation of cardiac fat from medical images is crucial for diagnosis and treatment planning.
• Existing segmentation methods can be time-consuming, limiting their clinical utility.

Purpose of the Study:

• To develop a novel, efficient, and automated approach for segmenting cardiac fat from Computed Tomography (CT) images.
• To significantly reduce the time required for cardiac fat segmentation compared to existing methods.
• To evaluate the accuracy and efficiency of the proposed segmentation technique as a potential medical diagnostic aid.

Main Methods:

• Utilized the Floor of Log (FoL) clustering algorithm for cardiac fat segmentation.
• Employed Support Vector Machine (SVM) to optimize FoL algorithm parameters.
• Incorporated mathematical morphology techniques for noise reduction in CT images.

Main Results:

• Achieved an average segmentation time of 2.01 seconds, a substantial improvement over traditional methods exceeding one hour.
• Reported high performance metrics: 93.45% accuracy and 95.52% specificity.
• The developed approach demonstrated automation and reduced computational requirements.

Conclusions:

• The proposed FoL-based cardiac fat segmentation method is highly efficient, offering rapid analysis times.
• The technique shows significant potential as a medical diagnostic aid, enabling faster and more accurate assessments.
• This approach can assist medical experts in improving the diagnosis and management of heart diseases.