Related Experiment Video
Updated: May 15, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
MRI-based deep learning with clinical and imaging features to differentiate medulloblastoma and ependymoma in
Yasen Yimit1,2, Parhat Yasin3, Yue Hao1
1Department of Radiology, The First People's Hospital of Kashi Prefecture, Kashgar, China.
Background:
Medulloblastoma (MB) and ependymoma (EM) in children share similarities in terms of age group, tumor location, and clinical presentation, which makes it challenging to clinically diagnose and distinguish them.
Purpose:
The present study aims to explore the effectiveness of T2-weighted magnetic resonance imaging (MRI)-based deep learning (DL) combined with clinical imaging features for differentiating MB from EM.
Methods:
Axial T2-weighted MRI sequences obtained from 201 patients across three study centers were used for model training and testing. The regions of interest were manually delineated by an experienced neuroradiologist with supervision by a senior radiologist. We developed a DL classifier using a pretrained AlexNet architecture that was fine-tuned on our dataset. To mitigate class imbalance, we implemented data augmentation and employed K-fold cross-validation to enhance model generalizability. For patient classification, we used two voting strategies: hard voting strategy in which the majority prediction was selected from individual image slices; soft voting strategy in which the prediction scores were averaged across slices with a threshold of 0.5. Additionally, a multimodality fusion model was constructed by integrating the DL classifier with clinical and imaging features. The model performance was assessed using a 7:3 random split of the dataset for training and validation, respectively. The key metrics like sensitivity, specificity, positive predictive value, negative predictive value, F1 score, area under the receiver operating characteristic curve (AUC), and accuracy were calculated, and statistical comparisons were performed using the DeLong test. Thereafter, MB was classified as positive, while EM was classified as negative.
Results:
The DL model with the hard voting strategy achieved AUC values of 0.712 (95% confidence interval (CI): 0.625-0.797) on the training set and 0.689 (95% CI: 0.554-0.826) on the test set. In contrast, the multimodality fusion model demonstrated superior performance with AUC values of 0.987 (95% CI: 0.974-0.996) on the training set and 0.889 (95% CI: 0.803-0.949) on the test set. The DeLong test indicated a statistically significant improvement in AUC values for the fusion model compared to the DL model (p < 0.001), highlighting its enhanced discriminative ability.
Conclusion:
T2-weighted MRI-based DL combined with multimodal clinical and imaging features can be used to effectively differentiate MB from EM in children. Thus, the structure of the decision tree in the decision tree classifier is expected to greatly assist clinicians in daily practice.
Insights
This study shows that combining deep learning with T2-weighted MRI and clinical data effectively distinguishes medulloblastoma (MB) from ependymoma (EM) in children. This approach aids in accurate diagnosis for pediatric brain tumors.
Area of Science:
- Pediatric neuro-oncology
- Medical imaging analysis
- Artificial intelligence in healthcare
Background:
- Medulloblastoma (MB) and ependymoma (EM) are common pediatric brain tumors with overlapping clinical presentations.
- Distinguishing between MB and EM is challenging due to similar age groups, tumor locations, and clinical symptoms.
Purpose of the Study:
- To evaluate the efficacy of deep learning (DL) models utilizing T2-weighted MRI and clinical data for differentiating pediatric MB from EM.
- To develop and validate a multimodal fusion model integrating imaging and clinical features.
Main Methods:
- A deep learning classifier (AlexNet) was trained on T2-weighted MRI scans from 201 pediatric patients across three centers.
- Data augmentation and K-fold cross-validation were employed to improve model generalizability and address class imbalance.
- A multimodal fusion model combined the DL classifier with clinical and imaging features, with performance assessed using sensitivity, specificity, AUC, and accuracy.
Main Results:
- The standalone DL model achieved an Area Under the Curve (AUC) of 0.689 on the test set.
- The multimodal fusion model significantly outperformed the DL model, achieving an AUC of 0.889 on the test set (p < 0.001).
- The fusion model demonstrated superior discriminative ability in differentiating MB from EM.
Conclusions:
- T2-weighted MRI-based deep learning combined with multimodal clinical and imaging features provides an effective method for differentiating pediatric medulloblastoma from ependymoma.
- This integrated approach is expected to significantly assist clinicians in the daily diagnosis of these challenging pediatric brain tumors.
