Multimodal MRI-based classification of migraine: using deep learning convolutional neural network

Abstract

BackgroundRecently, deep learning technologies have rapidly expanded into medical image analysis, including both disease detection and classification. As far as we know, migraine is a disabling and common neurological disorder, typically characterized by unilateral, throbbing and pulsating headaches. Unfortunately, a large number of migraineurs do not receive the accurate diagnosis when using traditional diagnostic criteria based on the guidelines of the International Headache Society. As such, there is substantial interest in developing automated methods to assist in the diagnosis of migraine.MethodsTo the best of our knowledge, no studies have evaluated the potential of deep learning technologies in assisting with the classification of migraine patients. Here, we used deep learning methods in combination with three functional measures (the amplitude of low-frequency fluctuations, regional homogeneity and regional functional correlation strength) based on rs-fMRI data to distinguish not only between migraineurs and healthy controls, but also between the two subtypes of migraine. We employed 21 migraine patients without aura, 15 migraineurs with aura, and 28 healthy controls.ResultsCompared with the traditional support vector machine classifier, which has an accuracy of 83.67%, our Inception module-based convolutional neural network approach showed a significant improvement in classification output (over 86.18%). Our data also indicate that the Inception module-based CNN performs better than the AlexNet-based CNN (Inception module-based CNN reached an accuracy of 99.25%). Finally, we also found that regional functional correlation strength (RFCS) could be regarded as the optimum input out of the three indices (ALFF, ReHo, RFCS).ConclusionsOverall, our study shows that combining the three functional measures of rs-fMRI with deep learning classification is a powerful method to distinguish between migraineurs and healthy individuals. Our data also highlight that deep learning-based frameworks could be used to develop more complicated models or systems to aid in clinical decision making in the future.Electronic supplementary materialThe online version of this article (10.1186/s12938-018-0587-0) contains supplementary material, which is available to authorized users.

Thyroid Diagnosis from SPECT Images Using Convolutional Neural Network with Optimization

“…Other CNN methods, including DenseNet121 [22], ResNet101 [23], InceptionV3 [24], VGG19 [25, 26], Modified VGG (MVGG) [27], GoogleNet [9,28–30], and SDAE [31–33], have also been implemented for comparison with the methods proposed in this paper.…”

“…ResNet is another efficient CNN architecture [24], and it is used to classify clinical 12 skin diseases in recent times [25]. Inception model is also reported for MRI-based classification of migraine in [26]. VGG architecture of CNN is employed in two-phase multimodel automatic brain tumour diagnosis system [27] and lung nodule classification between benign nodule and lung cancer [28].…”

The Spontaneous Activity Pattern of the Middle Occipital Gyrus Predicts the Clinical Efficacy of Acupuncture Treatment for Migraine Without Aura

“…Previous studies have found that patients with migraine displayed significant alterations in their cerebral spontaneous activity patterns ( 19 – 22 ), and acupuncture treatment could effectively regulate the disrupted cerebral functional activity of these patients ( 21 ). Furthermore, the regional cerebral functional activity and the interregional functional connectivity have been detected as the meaningful features for distinguishing patients with migraine from healthy individuals ( 23 , 24 ). Therefore, we hypothesized that the baseline cerebral spontaneous activity patterns of migraineurs could be used as a reliable predictor of acupuncture responsiveness after migraine treatment.…”

Deep learning for brain disorders: from data processing to disease treatment

“…Fu et al [172] performed natural language processing by using a CNN to detect individuals with silent brain infarction using radiological reports, as early detection can be useful for stroke prevention. Using different features extracted from functional MRI data, Yang et al [173] proposed to distinguish between migraine patients and healthy controls (but also between two subtypes of migraine) using an Inception CNN. Finally, MR angiography was used to detect cerebral aneurysms [174,175] using a custom CNN [174] or a ResNet-18 [175].…”