Automated Segmentation and Volumetry for Meningioma Using Deep Learning

Artificial Intelligence Clinical Trial

Official title:

Automated Meningioma Segmentation and Volumetry Using a nnU-Net Based Architecture on Contrast-enhanced MRI

Clinical Trial Details — Status: Completed

Administrative data

• NCT number: NCT05093751
• Other study ID #: SNUH-MNG-AI001
• Status: Completed
• Start date: March 23, 2013
• Est. completion date: September 30, 2021

Study information

• Verified date: October 2021
• Source: Seoul National University Hospital
• Contact: n/a
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

U-Net-based architectures will be applied to 500 contrast-enhanced axial MR images of different patients from a single institution after manual segmentation of meningioma, of which 50 were used for testing. Tumor volumetry after autosegmentation by trained U-Net-based architecture is final goal.

Description:

U-Net-based architectures will be applied to 500 contrast-enhanced axial MR images of different patients from a single institution after manual segmentation of meningioma, of which 50 were used for testing. After preprocessing with Z-isotropification and intensity normalization of images, 3 U-Net-based networks (2D U-Net, Attention U-Net, 3D U-Net) and 3 nnU-Net-based networks (2D nnU-Net, Attention nnU-Net, 3D nnU-Net) will be trained with meningioma-segmented images. For applying to 3D networks, sagittal and coronal images will be reconstructed using axial images. After prediction, the cut-off of the probability function, which is a trade-off, will be obtained with the Gaussian Mixture Modeling algorithm using the probability density function. The voxels having a probability function higher than that will be finally predicted as meningioma. Tumor volume is calculated as the sum of the product of segmented area and thickness of axial images. For performance evaluation, dice similarity coefficient (DSC), precision, and recall will be evaluated compared with manually segmented voxels for validation datasets. The results of volumetry of each model will be compared with manual segmentation-based volume through Pearson's correlation analysis.

Recruitment information / eligibility

• Status: Completed
• Enrollment: 600
• Est. completion date: September 30, 2021
• Est. primary completion date: September 30, 2021
• Accepts healthy volunteers: No
• Gender: All
• Age group: 18 Years and older

Eligibility

Inclusion Criteria:

• Radiologically diagnosed meningioma by MRI

Exclusion Criteria:

• under 18 years old

• Multiple meningiomas

• Orbital meningioma

• Any prior treatment for intracranial meningioma before registration

Related Conditions & MeSH terms

• Artificial Intelligence
• Meningioma

Intervention

Other:
• Observation
This study does not involve any intervention to subjects.

Locations

Country	Name	City	State
n/a

Sponsors (1)

Lead Sponsor	Collaborator
Seoul National University Hospital

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Accuracy compared with ground truth	As a primary endpoint, we will examine the ability of U-Net and nnU-Net to segment meningioma in brain MR compared with ground truth. Ground truth is defined as area on MR drawn by two neurosurgeons. Accuracy of autosegmentation of meningioma will be assessed in dice similarity coefficient, recall, and precision.	10-01-2020 until 09-30-2021