Automated segmentation of pituitary adenomas, pituitary gland, and internal carotid arteries on routine coronal contrast-enhanced T1-weighted MRI: a single-sequence feasibility study - Takeaways - MDSpire

Automated segmentation of pituitary adenomas, pituitary gland, and internal carotid arteries on routine coronal contrast-enhanced T1-weighted MRI: a single-sequence feasibility study

  • By

  • Woojae Hong

  • Ho Kang

  • Jong Ha Hwang

  • Seong-Min Kim

  • Yong Hwy Kim

  • Hyunggun Kim

  • June 30, 2026

  • 0 min

Share

  • 1

    This study evaluated the feasibility of automated segmentation techniques for pituitary adenomas, pituitary gland, and internal carotid arteries using T1CE coronal MRI.

  • 2

    A cohort of 255 patients with surgically confirmed pituitary adenomas was analyzed, utilizing manual annotations by expert neurosurgeons as ground truth.

  • 3

    Swin-Unet and nnU-Net demonstrated comparable voxel-wise performance, with mean Dice similarity coefficients of 0.70 and 0.69, respectively.

  • 4

    Segmentation accuracy was lower for the pituitary gland compared to the pituitary adenoma and internal carotid arteries.

  • 5

    The mean computation time for segmentation was 78.4 seconds per patient, facilitating approximate 3D reconstructions for qualitative assessment.

Original Source(s)

Automated segmentation of pituitary adenomas, pituitary gland, and internal carotid arteries on routine coronal contrast-enhanced T1-weighted MRI: a single-sequence feasibility study

  • by Woojae Hong, Ho Kang, Jong Ha Hwang, Seong-Min Kim, Yong Hwy Kim, Hyunggun Kim

  • June 30, 2026

Related Content

Pre-treatment CEMRI habitat radiomics and deep learning for noninvasive prediction of the VETC pattern in hepatocellular carcinoma: an exploratory radiogenomic analysis

Incisura groove depth is associated with reduced rotational fibular displacement after syndesmotic injury: cadaveric validation of an automated three-dimensional incisura morphometric pipeline

Noninvasive MRA-derived fractional flow for intracranial stenosis: methodological evaluation and hemodynamic insights