Improved prediction of postoperative pediatric cerebellar mutism syndrome using an artificial neural network
Jai Sidpra1,2,3, Adam P Marcus4, Ulrike Löbel3
1University College London Medical School, London, UK.
Insights
An artificial neural network (ANN) accurately predicts postoperative pediatric cerebellar mutism syndrome (pCMS) after posterior fossa tumor resection. This AI model shows superior performance compared to existing methods, improving patient outcomes.
Area of Science:
- Neuroscience
- Oncology
- Medical Artificial Intelligence
Background:
- Postoperative pediatric cerebellar mutism syndrome (pCMS) is a significant complication following posterior fossa tumor resection in children.
- Previous attempts to predict pCMS preoperatively have yielded inconsistent results.
- This study investigates the efficacy of artificial neural networks (ANNs) for more accurate pCMS prediction.
Purpose of the Study:
- To evaluate the generalization of existing pCMS predictive models.
- To develop and assess an ANN for improved preoperative prediction of pCMS.
- To compare the ANN's predictive performance against conventional methods.
Main Methods:
- A systematic review identified key risk factors for pCMS.
- A retrospective dataset was compiled based on these risk factors from pediatric patients undergoing posterior fossa tumor resection.
- An ANN was trained and its performance evaluated using receiver operator characteristic (ROC) curves, comparing it with logistic regression and other predictive indices.
Main Results:
- The study included 204 children, with 80 developing pCMS.
- The ANN achieved a high performance with an Area Under the Curve (AUC) of 0.949 and 90.9% accuracy.
- The ANN significantly outperformed logistic regression and previously established external predictive models (P < .05 and P < .001, respectively).
Conclusions:
- Artificial neural networks demonstrate superior accuracy in predicting pCMS compared to traditional methods.
- The developed ANN offers a promising tool for identifying children at high risk of pCMS.
- Improved prediction can facilitate timely interventions and potentially mitigate the severity of pCMS.
Background:
Postoperative pediatric cerebellar mutism syndrome (pCMS) is a common but severe complication that may arise following the resection of posterior fossa tumors in children. Two previous studies have aimed to preoperatively predict pCMS, with varying results. In this work, we examine the generalization of these models and determine if pCMS can be predicted more accurately using an artificial neural network (ANN).
Methods:
An overview of reviews was performed to identify risk factors for pCMS, and a retrospective dataset was collected as per these defined risk factors from children undergoing resection of primary posterior fossa tumors. The ANN was trained on this dataset and its performance was evaluated in comparison to logistic regression and other predictive indices via analysis of receiver operator characteristic curves. The area under the curve (AUC) and accuracy were calculated and compared using a Wilcoxon signed-rank test, with P < .05 considered statistically significant.
Results:
Two hundred and four children were included, of whom 80 developed pCMS. The performance of the ANN (AUC 0.949; accuracy 90.9%) exceeded that of logistic regression (P < .05) and both external models (P < .001).
Conclusion:
Using an ANN, we show improved prediction of pCMS in comparison to previous models and conventional methods.
