K Means Segmentation of Alzheimers Disease in PET scan datasets: An implementation

TL;DR

This paper explores the application of K Means clustering for segmenting Alzheimer's disease in PET scan images, comparing tools and implementations to improve automation and accuracy in medical image analysis.

Contribution

It provides an implementation of K Means clustering for PET scan segmentation and discusses toolboxes for reproducing and extending this method in medical imaging.

Findings

K Means clustering effectively segments Alzheimer's regions in PET scans.

Comparison of toolboxes aids in selecting suitable tools for medical image segmentation.

Implementation demonstrates potential for automation in Alzheimer's diagnosis.

Abstract

The Positron Emission Tomography (PET) scan image requires expertise in the segmentation where clustering algorithm plays an important role in the automation process. The algorithm optimization is concluded based on the performance, quality and number of clusters extracted. This paper is proposed to study the commonly used K Means clustering algorithm and to discuss a brief list of toolboxes for reproducing and extending works presented in medical image analysis. This work is compiled using AForge .NET framework in windows environment and MATrix LABoratory (MATLAB 7.0.1)