Volumetric measurement of intracranial meningiomas: a comparison between linear, planimetric, and machine learning with multiparametric voxel-based morphometry methods

Mathematical modeling of meningioma volume change after radiation treatment

OBJECTIVE

Meningiomas are the most common primary central nervous tumor and are often treated with radiation therapy. This study examines the long-term volumetric changes of intracranial meningiomas in response to radiation therapy. The objective is to analyze and model the volumetric changes following treatment.

METHODS

Data from a retrospective single-institution database (2005-2015) were used, with inclusion criteria being patients with a diagnosis of meningiomas, along with additional inclusion criteria consisting of treatment with radiation, having at least three magnetic resonance imaging (MRI) scans with one or more before and after radiation treatment, and the patients following up for at least eighteen months. Exclusion criteria consisted of patients less than 18 years old, patients receiving surgery and/or adjuvant chemotherapy following radiation, and patients without any available details regarding radiation treatment parameters. Tumor volumes were measured via T1-weighted post-contrast MRI and calculated using the ABC/2 ellipsoidal approximation, a method allowing for the measurement of non-linear growth volume reduction.

RESULTS

Of 48 meningioma patients considered, 10 % experienced post-radiation growth, while 75 % witnessed a ≥50 % decrease in volume over a follow-up period of 0.3-14.9 years. Median decay rate was 0.81, and within 1.17 years, 90 % achieved the predicted volume reduction. Predicted vs. actual volumes showed a mean difference of 0.009 ± 0.347 cc. Initial tumor volumes strongly correlated (Pearson's R=0.98, R-squared=0.96) with final asymptotic volumes, which had a median of 1.50 cc, with interquartile range (IQR) = [0.39, 3.67].

CONCLUSION

90 % of patients achieved tumor-volume reduction at 1.17 years post-treatment, reaching a non-zero asymptote strongly correlated with initial tumor volume, and 75 % experienced at least a 50 % volume decrease. Individual volume changes for responsive meningiomas can be modeled and predicted using exponential decay curves.

Performance Test of a Well-Trained Model for Meningioma Segmentation in Health Care Centers: Secondary Analysis Based on Four Retrospective Multicenter Data Sets

BACKGROUND

Convolutional neural networks (CNNs) have produced state-of-the-art results in meningioma segmentation on magnetic resonance imaging (MRI). However, images obtained from different institutions, protocols, or scanners may show significant domain shift, leading to performance degradation and challenging model deployment in real clinical scenarios.

OBJECTIVE

This research aims to investigate the realistic performance of a well-trained meningioma segmentation model when deployed across different health care centers and verify the methods to enhance its generalization.

METHODS

This study was performed in four centers. A total of 606 patients with 606 MRIs were enrolled between January 2015 and December 2021. Manual segmentations, determined through consensus readings by neuroradiologists, were used as the ground truth mask. The model was previously trained using a standard supervised CNN called Deeplab V3+ and was deployed and tested separately in four health care centers. To determine the appropriate approach to mitigating the observed performance degradation, two methods were used: unsupervised domain adaptation and supervised retraining.

RESULTS

The trained model showed a state-of-the-art performance in tumor segmentation in two health care institutions, with a Dice ratio of 0.887 (SD 0.108, 95% CI 0.903-0.925) in center A and a Dice ratio of 0.874 (SD 0.800, 95% CI 0.854-0.894) in center B. Whereas in the other health care institutions, the performance declined, with Dice ratios of 0.631 (SD 0.157, 95% CI 0.556-0.707) in center C and 0.649 (SD 0.187, 95% CI 0.566-0.732) in center D, as they obtained the MRI using different scanning protocols. The unsupervised domain adaptation showed a significant improvement in performance scores, with Dice ratios of 0.842 (SD 0.073, 95% CI 0.820-0.864) in center C and 0.855 (SD 0.097, 95% CI 0.826-0.886) in center D. Nonetheless, it did not overperform the supervised retraining, which achieved Dice ratios of 0.899 (SD 0.026, 95% CI 0.889-0.906) in center C and 0.886 (SD 0.046, 95% CI 0.870-0.903) in center D.

CONCLUSIONS

Deploying the trained CNN model in different health care institutions may show significant performance degradation due to the domain shift of MRIs. Under this circumstance, the use of unsupervised domain adaptation or supervised retraining should be considered, taking into account the balance between clinical requirements, model performance, and the size of the available data.