Related Experiment Video
Updated: Jun 3, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
Development and routine implementation of deep learning algorithm for automatic brain metastases segmentation on MRI
Loïse Dessoude1, Raphaëlle Lemaire2, Romain Andres2
1Radiotherapy Department, Centre François Baclesse, Caen 14000, France.
Rationale And Objectives:
The RANO-BM criteria, which employ a one-dimensional measurement of the largest diameter, are imperfect due to the fact that the lesion volume is neither isotropic nor homogeneous. Furthermore, this approach is inherently time-consuming. Consequently, in clinical practice, monitoring patients in clinical trials in compliance with the RANO-BM criteria is rarely achieved. The objective of this study was to develop and validate an AI solution capable of delineating brain metastases (BM) on MRI to easily obtain, using an in-house solution, RANO-BM criteria as well as BM volume in a routine clinical setting.
Materials (Patients) And Methods:
A total of 27,456 post-Gadolinium-T1 MRI from 132 patients with BM were employed in this study. A deep learning (DL) model was constructed using the PyTorch and PyTorch Lightning frameworks, and the UNETR transfer learning method was employed to segment BM from MRI.
Results:
A visual analysis of the AI model results demonstrates confident delineation of the BM lesions. The model shows 100 % accuracy in predicting RANO-BM criteria in comparison to that of an expert medical doctor. There was a high degree of overlap between the AI and the doctor's segmentation, with a mean DICE score of 0.77. The diameter and volume of the BM lesions were found to be concordant between the AI and the reference segmentation. The user interface developed in this study can readily provide RANO-BM criteria following AI BM segmentation.
Conclusion:
The in-house deep learning solution is accessible to everyone without expertise in AI and offers effective BM segmentation and substantial time savings.
Insights
This study introduces an AI solution for segmenting brain metastases (BM) on MRI, accurately assessing RANO-BM criteria and lesion volume. The accessible tool saves time and improves clinical trial monitoring for brain metastases.
Area of Science:
- Radiology and Medical Imaging
- Artificial Intelligence in Medicine
- Oncology
Background:
- Current RANO-BM criteria for brain metastases (BM) assessment are time-consuming and imperfect due to lesion heterogeneity.
- Clinical trials rarely achieve full compliance with RANO-BM criteria in routine practice.
- Accurate and efficient monitoring of BM is crucial for patient management and clinical trial success.
Purpose of the Study:
- To develop and validate an AI solution for delineating brain metastases (BM) on MRI.
- To enable easy and routine acquisition of RANO-BM criteria and BM volume using an in-house AI tool.
- To overcome the limitations of manual RANO-BM assessment in clinical settings.
Main Methods:
- A deep learning (DL) model utilizing the UNETR transfer learning method was developed.
- The model was trained and validated on 27,456 post-Gadolinium-T1 MRI scans from 132 patients with BM.
- PyTorch and PyTorch Lightning frameworks were used for model construction.
Main Results:
- The AI model achieved 100% accuracy in predicting RANO-BM criteria compared to expert assessment.
- A high overlap between AI and expert segmentation was observed, with a mean DICE score of 0.77.
- AI-derived diameter and volume measurements of BM lesions were concordant with reference segmentations.
Conclusions:
- The developed in-house deep learning solution provides effective BM segmentation and RANO-BM criteria assessment.
- The AI tool offers substantial time savings and is accessible without AI expertise.
- This solution facilitates routine clinical application and improves monitoring of brain metastases.
