Computer-based assessment for facioscapulohumeral dystrophy diagnosis

Area of Science:

• Medical imaging analysis
• Biomedical engineering
• Neuromuscular disease research

Background:

• Facioscapulohumeral dystrophy (FSHD) diagnosis relies on clinical evaluation and muscle imaging.
• Quantifying muscle fat infiltration and edema is crucial for assessing FSHD progression.
• Accurate muscle region segmentation in MRI is a prerequisite for reliable quantification.

Purpose of the Study:

• To develop and validate a novel computer-based method for automated muscle segmentation in MRI for FSHD assessment.
• To quantify fat and edema percentages within segmented muscle regions in FSHD patients.
• To evaluate the accuracy of the proposed segmentation method against manual segmentation by medical specialists.

Main Methods:

• A multi-slice, live-wire-based technique was developed for segmenting muscle regions in T1-weighted MRI.
• An exponential cost function incorporating edge information from an edge-enhancement algorithm was utilized.
• Fuzzy c-mean clustering was applied to quantify fat and edema in T1-weighted and T2-STIR MRI, respectively.
• The Zijdenbos similarity index was used to compare automated and manual segmentations.

Main Results:

• The proposed live-wire-based method demonstrated high accuracy in muscle region segmentation, as validated by the Zijdenbos similarity index.
• Quantification of fat and edema percentages was successfully performed on T1-weighted and T2-STIR MRI scans of 10 FSHD patients.
• The computer-based assessment provides a quantitative measure of muscle pathology in FSHD.

Conclusions:

• The developed automated segmentation method is accurate and reliable for muscle region analysis in FSHD.
• This computer-based approach facilitates objective quantification of fat and edema, aiding in FSHD diagnosis and monitoring.
• The study highlights the potential of advanced image analysis techniques in improving neuromuscular disease assessment.