Comparison of Machine Learning Algorithms for the Automated Detection of Functional Brain Networks in fMRI (Preprint)
BACKGROUND The functional MRI (fMRI) is an essential tool for the presurgical planning of brain tumor removal, allowing the identification of functional brain networks in order to preserve the patient’s neurological functions. One fMRI technique used to identify the functional brain network is the resting-state-fMRI (rsfMRI). However, this technique is not routinely used because of the necessity to have a expert reviewer to identify manually each functional networks. OBJECTIVE We aimed to automatize the detection of brain functional networks in rsfMRI data using machine learning algorithms. METHODS We used the rsfMRI data of 30 healthy patients to test the diagnostic performance of 10 machine learning algorithms compared to the reference functional networks identified manually by 2 expert reviewers. Then we selected the most fitted algorithm that we trained and tuned to optimize the diagnostic performance. RESULTS The comparison of the diagnostic performance of the machine learning algorithms identified the artificial neuron network using a scale conjugate gradient backpropagation as the most fitted algorithm. After training and fine tuning of the hyperparameters, the selected machine learning algorithm was able to identify correctly the different functional networks with an accuracy between 89 and 100%. CONCLUSIONS The artificial neural network using a scaled conjugate gradient backpropagation was the most performant machine learning algorithm. The use of this machine learning to automatize the functional networks detection in rsfMRI may allow to broaden the use of the rsfMRI, allowing the presurgical identification of these networks and thus help to preserve the patient’s neurological status.