Neuro / Head & NeckAI / InformaticsResearch

Attention U-Net using contrast-enhanced T1, T1, and FLAIR MRI achieves high meningioma detection and segmentation

Journal of imaging informatics in medicineyesterday

In 149 patients, an Attention U-Net using contrast-enhanced T1-weighted, T1-weighted, and FLAIR MRI detected meningiomas with detection rate 0.98 (SD 0.03) and Dice 0.82 (SD 0.04), compared with contrast-enhanced T1 alone at 0.79 and Dice 0.67.

  • Among single-sequence models, contrast-enhanced T1-weighted MRI performed best, with detection rate 0.79±0.11 and Dice 0.67±0.10.
  • Models trained on multiple MRI sequences up to three consistently outperformed single-sequence models; T1-CE + T1 + FLAIR had the highest detection and segmentation performance.
  • The study retrospectively analyzed 149 patients with histologically diagnosed meningioma, using manual segmentation by one experienced neuroradiologist as the reference standard.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.