Deep Learning Approach to Predict Hemorrhage in Moyamoya Disease

TL;DR

This study develops machine learning models, especially an artificial neural network, to predict hemorrhage risk in moyamoya disease patients using clinical and radiographic data, achieving over 75% accuracy.

Contribution

It introduces a novel ANN-based predictive model for hemorrhage risk in MMD, outperforming SVM and random forest methods.

Findings

ANN achieved 75.7% accuracy in prediction.

Radiographic and clinical features effectively inform hemorrhage risk.

ANN outperformed other machine learning models in this task.

Abstract

Objective: Reliable tools to predict moyamoya disease (MMD) patients at risk for hemorrhage could have significant value. The aim of this paper is to develop three machine learning classification algorithms to predict hemorrhage in moyamoya disease. Methods: Clinical data of consecutive MMD patients who were admitted to our hospital between 2009 and 2015 were reviewed. Demographics, clinical, radiographic data were analyzed to develop artificial neural network (ANN), support vector machine (SVM), and random forest models. Results: We extracted 33 parameters, including 11 demographic and 22 radiographic features as input for model development. Of all compared classification results, ANN achieved the highest overall accuracy of 75.7% (95% CI, 68.6%-82.8%), followed by SVM with 69.2% (95% CI, 56.9%-81.5%) and random forest with 70.0% (95% CI, 57.0%-83.0%). Conclusions: The proposed ANN framework can be a potential effective tool to predict the possibility of hemorrhage among adult MMD patients based on clinical information and radiographic features.