Adaptive threshold method for recovered images of FMT

Area of Science:

• Biomedical Imaging
• Optical Imaging
• Medical Physics

Background:

• Fluorescence molecular tomography (FMT) is a powerful imaging technique.
• Recovered FMT images often contain noise and artifacts.
• Effective post-processing is crucial for accurate interpretation of FMT data.

Purpose of the Study:

• To develop and validate an adaptive post-processing strategy for FMT images.
• To objectively enhance the clarity and reduce spurious information in reconstructed FMT images.
• To provide a user-friendly method for improving FMT image quality.

Main Methods:

• An adaptive thresholding method was developed to process recovered FMT images.
• The threshold value is automatically determined from the reconstructed image data.
• Simulation and physical phantom experiments were conducted to evaluate the method.
• Image clarity was assessed visually and quantitatively using full width at half-maximum (FWHM) and contrast-to-noise ratio (CNR).

Main Results:

• The proposed adaptive thresholding method significantly improved the clarity of recovered FMT images compared to no post-processing.
• Visualization confirmed clearer images after applying the adaptive threshold.
• Quantitative analysis using FWHM and CNR demonstrated the method's effectiveness in removing spurious information.
• The method proved objective as the threshold was derived directly from reconstructed image data.

Conclusions:

• The adaptive thresholding strategy is an effective post-processing technique for fluorescence molecular tomography.
• This method objectively enhances image quality, making it more convenient for users.
• Improved image clarity facilitates more accurate analysis and interpretation of FMT data.