Time sequence image analysis of positron emission tomography using wavelet transformation

Area of Science:

• Medical Imaging
• Signal Processing
• Radiochemistry

Background:

• Positron Emission Tomography (PET) is a crucial imaging modality for diagnosing various diseases.
• Analyzing dynamic PET scans, particularly for abdominal organs, presents challenges in accurate region of interest (ROI) selection.
• Current methods for ROI identification can be time-consuming and prone to human error.

Purpose of the Study:

• To develop an automated image analysis technique for dynamic PET scans using wavelet transformation.
• To enable automatic selection of the region of interest (ROI) for abdominal organs in dynamic PET imaging.
• To improve the accuracy and efficiency of lesion detection in PET scans.

Main Methods:

• Dynamic PET scans of the abdominal cavity were acquired using [18F]Fluoro-2-deoxy-2-D-glucose (18F-FDG).
• Wavelet transformation was applied to dynamic PET images for organ selection and automatic ROI identification.
• Image segmentation involved sampling, wavelet transformation, erosion, dilation, and superimposition to create Wavelet Constructed Image (WCI) contours.

Main Results:

• The developed image segmentation technology successfully identified organ contours from dynamic PET sequences.
• Automatic ROI selection was achieved, demonstrating the effectiveness of the wavelet transformation method.
• The technique facilitated accurate localization of lesions within organs.

Conclusions:

• The wavelet transformation method offers an efficient and accurate approach for analyzing dynamic PET images.
• This technology can significantly reduce human operational time and minimize errors in medical image analysis.
• Automated ROI selection in PET scans has the potential to enhance diagnostic capabilities and patient care.