People suffering from mild cognitive impairment (MCI) are at an increased risk of developing Alzheimer's disease (AD) or another dementia. High prevalence will possibly be reduced if early interventions could be applied to the stage of early MCI (eMCI). In network-based classification, brain functional networks are often constructed, relying on the entire time series. It can lead to the neglect of the complex and dynamic interaction relationships among brain regions. As a result, the features derived from this type of functional network may fail to serve as an effective disease biomarker. To address this problem, we proposed a multi-scale feature combination framework for the eMCI classification. In this framework, global static features, time-varying features, and more refined features could be able to flexibly extract from static functional networks, dynamic functional networks, and high-order functional networks, respectively. Then, they are utilized to train and test the classification model in the form of feature combination. The experimental results have verified that the proposed method achieves superior classification accuracy than other competed methods in the eMCI classification, indicating a great potential in understanding the dysfunction of the brain regions. INDEX TERMS Brain functional network, multi-scale feature combination, Alzheimer's disease (AD), early mild cognitive impairment (eMCI), classification accuracy.

The treatment of neurodegenerative diseases is expensive, and long-term treatment makes families bear a heavy burden. Accumulating evidence suggests that the high conversion rate can possibly be reduced if clinical interventions are applied at the early stage of brain diseases. Thus, a variety of deep learning methods are utilized to recognize the early stages of neurodegenerative diseases for clinical intervention and treatment. However, most existing methods have ignored the issue of sample imbalance, which often makes it difficult to train an effective model due to lack of a large number of negative samples. To address this problem, we propose a two-stage method, which is used to learn the compression and recover rules of normal subjects so that potential negative samples can be detected. The experimental results show that the proposed method can not only obtain a superb recognition result, but also give an explanation that conforms to the physiological mechanism. Most importantly, the deep learning model does not need to be retrained for each type of disease, which can be widely applied to the diagnosis of various brain diseases. Furthermore, this research could have great potential in understanding regional dysfunction of various brain diseases.

No abstract

No abstract

No abstract