Principal component analysis of dynamic fluorescence diffuse optical tomography images.

Area of Science:

• Biomedical optics
• Medical imaging
• Pharmacokinetics

Background:

• Accurate visualization of drug distribution in small animals is crucial for preclinical research.
• Current fluorescence diffuse optical tomography (FDOT) methods face challenges in resolving organ-specific drug kinetics.
• Developing advanced imaging techniques is essential for understanding drug biodistribution and metabolism.

Purpose of the Study:

• To present a novel method using principal component analysis (PCA) for enhanced detection and visualization of organs with distinct metabolic kinetics.
• To improve the analysis of time-series FDOT data for better drug distribution mapping.
• To validate the proposed PCA-based approach through simulations and experimental studies.

Main Methods:

• Utilizing principal component analysis (PCA) on time-series FDOT images to identify spatial patterns related to different kinetic behaviors.
• Simulating and imaging Indocyanine green (ICG) metabolic processes using FDOT.
• Applying PCA to extract and represent kinetic variations within the acquired FDOT data.

Main Results:

• PCA successfully identified principal components representing distinct kinetic behaviors of ICG.
• The method demonstrated the ability to extract and illustrate changes in ICG kinetic behavior between the heart and lungs.
• Both simulation and experimental studies validated the performance of the PCA-based FDOT algorithm.

Conclusions:

• The proposed PCA method offers a significant advancement in resolving organ-specific drug distributions using FDOT.
• This technique enhances the visualization of kinetic differences, aiding in the understanding of drug behavior in vivo.
• The findings support the utility of PCA for analyzing complex dynamic processes in biomedical imaging.