Yimit Yasen, Yasin Parhat, Hao Yue, Tuersun Abudouresuli, Huang Chencui, Zou Xiaoguang, Qiu Ya, Wang Yunling, Nijiati Mayidili
Department of Radiology, The First People's Hospital of Kashi Prefecture, Kashgar, China.
The Xinjiang Key Laboratory of Artificial Intelligence Assisted Imaging Diagnosis, Varanasi, China.
Front Mol Biosci. 2025 Apr 28;12:1570860. doi: 10.3389/fmolb.2025.1570860. eCollection 2025.
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.
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.
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.
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 ( < 0.001), highlighting its enhanced discriminative ability.
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.
儿童髓母细胞瘤（MB）和室管膜瘤（EM）在年龄组、肿瘤位置和临床表现方面存在相似之处，这使得临床诊断和区分它们具有挑战性。
本研究旨在探讨基于T2加权磁共振成像（MRI）的深度学习（DL）结合临床影像特征区分MB和EM的有效性。
从三个研究中心的201例患者中获取的轴向T2加权MRI序列用于模型训练和测试。感兴趣区域由经验丰富的神经放射科医生在资深放射科医生的监督下手动勾勒。我们使用预训练的AlexNet架构开发了一个DL分类器，并在我们的数据集上进行了微调。为了减轻类别不平衡，我们实施了数据增强并采用K折交叉验证来提高模型的泛化能力。对于患者分类，我们使用了两种投票策略：硬投票策略，即从单个图像切片中选择多数预测；软投票策略，即对切片预测分数进行平均，阈值为0.5。此外，通过将DL分类器与临床和影像特征相结合构建了一个多模态融合模型。使用数据集的7:3随机分割分别进行训练和验证来评估模型性能。计算了敏感性、特异性、阳性预测值、阴性预测值、F1分数、受试者操作特征曲线下面积（AUC）和准确性等关键指标，并使用DeLong检验进行统计比较。此后，将MB分类为阳性，将EM分类为阴性。
采用硬投票策略的DL模型在训练集上的AUC值为0.712（95%置信区间（CI）：0.625 - 0.797），在测试集上为0.689（95%CI：0.554 - 0.826）。相比之下，多模态融合模型表现更优，在训练集上的AUC值为0.987（95%CI：0.974 - 0.996），在测试集上为0.889（95%CI：0.803 - 0.949）。DeLong检验表明，与DL模型相比，融合模型的AUC值有统计学显著提高（<0.001），突出了其增强的判别能力。
基于T2加权MRI的DL结合多模态临床和影像特征可有效区分儿童的MB和EM。因此，决策树分类器中的决策树结构有望在日常实践中极大地帮助临床医生。
