What question did this study set out to answer?

The aim is to develop a deep learning model for fully automated 3D segmentation and volumetric assessment of meningioma burden.

A deep learning algorithm for fully automated volumetric measurement of meningioma burden

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial developed a deep learning model for volumetric assessment of meningioma burden, suggesting improved accuracy and efficiency.

The aim is to develop a deep learning model for fully automated 3D segmentation and volumetric assessment of meningioma burden.
Utilized 450 post-contrast T1-weighted brain MRIs from 104 patients.
Trained a U-Net model with a joint Dice-cross entropy loss function.
Included diverse meningioma grades and treatment statuses in the dataset.
Achieved a median Dice of 0.741 and HD95 of 26 mm on high-grade tumors.
Achieved a median Dice of 0.848 and HD95 of 1.41 mm on low-grade tumors.
Generalized well to external data with a median Dice of 0.923 and HD95 of 2.24 mm.

Cleveland et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial developed a deep learning model for volumetric assessment of meningioma burden, suggesting improved accuracy and efficiency.

The aim is to develop a deep learning model for fully automated 3D segmentation and volumetric assessment of meningioma burden.
Utilized 450 post-contrast T1-weighted brain MRIs from 104 patients.
Trained a U-Net model with a joint Dice-cross entropy loss function.
Included diverse meningioma grades and treatment statuses in the dataset.
Achieved a median Dice of 0.741 and HD95 of 26 mm on high-grade tumors.
Achieved a median Dice of 0.848 and HD95 of 1.41 mm on low-grade tumors.
Generalized well to external data with a median Dice of 0.923 and HD95 of 2.24 mm.

Cleveland et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark