An EM approach to MAP solution of segmenting tissue mixtures: a numerical analysis

Area of Science:

• Medical Image Processing
• Computational Biology
• Statistical Modeling

Background:

• The partial volume effect is a significant challenge in medical image analysis, leading to quantitative imprecision.
• Accurate segmentation of mixed tissues within image voxels is crucial for reliable quantitative analysis.

Purpose of the Study:

• To develop and present an iterative expectation-maximization (EM) approach for the maximum a posteriori (MAP) solution to segmenting tissue mixtures within image voxels.
• To provide a theoretical solution to the partial volume effect in medical image processing.

Main Methods:

• An iterative expectation-maximization (EM) algorithm is employed to find the maximum a posteriori (MAP) solution.
• Tissue types are modeled using normal distributions across the field-of-view (FOV), assuming independence.
• A Markov random field model is used for underlying tissue distributions, enabling the computation of conditional expectations.

Main Results:

• The proposed MAP-EM framework iteratively refines tissue mixture estimations.
• Numerical analysis indicates accurate and efficient estimation of tissue mixtures.
• The method offers a theoretical solution to the partial volume effect.

Conclusions:

• The iterative EM approach to MAP solution effectively segments tissue mixtures within voxels.
• This framework has the potential to enhance quantitative precision in medical image processing by addressing the partial volume effect.
• The method demonstrates accuracy and efficiency in numerical analyses.