Automatic lymphedema segmentation in T2-STIR MRI using an unsupervised clustering method

Key Points

• Segmented regions demonstrate high congruity with expert clinician assessments, achieving a 92% agreement rate.
• The unsupervised clustering method used for image segmentation allows for efficient identification of lymphedema areas without manual annotation.
• Utilization of T2-STIR MRI images facilitates improved visualization of fluid accumulation, aiding in lymphedema diagnosis.
• Highlights the need for automation in MRI analysis, suggesting enhancements in clinical workflow and patient care.