ABSTRACT
BACKGROUND: Parkinson's patients have significant autonomic dysfunction, early detect the disorder is a major challenge. To assess the autonomic function in the rat model of rotenone induced Parkinson's disease (PD), Blood pressure and ECG signal acquisition are very important. NEW METHOD: We used telemetry to record the electrocardiogram and blood pressure signals from awake rats, with linear and nonlinear analysis techniques calculate the heart rate variability (HRV) and blood pressure variability (BPV). we applied nonlinear analysis methods like sample entropy and detrended fluctuation analysis to analyze blood pressure signals. Particularly, this is the first attempt to apply nonlinear analysis to the blood pressure evaluate in rotenone induced PD model rat. RESULTS: HRV in the time and frequency domains indicated sympathetic-parasympathetic imbalance in PD model rats. Linear BPV analysis didn't reflect changes in vascular function and blood pressure regulation in PD model rats. Nonlinear analysis revealed differences in BPV, with lower sample entropy results and increased detrended fluctuation analysis results in the PD group rats. COMPARISON WITH EXISTING METHODS AND CONCLUSIONS: our experiments demonstrate the ability to evaluate autonomic dysfunction in models of Parkinson's disease by combining the analysis of BPV with HRV, consistent with autonomic impairment in PD patients. Nonlinear analysis by blood pressure signal may help in early detection of the PD. It indicates that the fluctuation of blood pressure in the rats in the rotenone model group tends to be regular and predictable, contributes to understand the PD pathophysiological mechanisms and to find strategies for early diagnosis.
ABSTRACT
Schizophrenia classification and abnormal brain network recognition have an important research significance. Researchers have proposed many classification methods based on machine learning and deep learning. However, fewer studies utilized the advantages of complementary information from multi feature to learn the best representation of schizophrenia. In this study, we proposed a multi-feature fusion network (MFFN) using functional network connectivity (FNC) and time courses (TC) to distinguish schizophrenia patients from healthy controls. DNN backbone was adopted to learn the feature map of functional network connectivity, C-RNNAM backbone was designed to learn the feature map of time courses, and Deep SHAP was applied to obtain the most discriminative brain networks. We proved the effectiveness of this proposed model using the combining two public datasets and evaluated this model quantitatively using the evaluation indexes. The results showed that the functional network connectivity generated by independent component analysis has advantage in schizophrenia classification by comparing static and dynamic functional connections. This method obtained the best classification accuracy (ACC=87.30%, SPE=89.28%, SEN=85.71%, F1 =88.23%, and AUC=0.9081), and it demonstrated the superiority of this proposed model by comparing state-of-the-art methods. Ablation experiment also demonstrated that multi feature fusion and attention module can improve classification accuracy. The most discriminative brain networks showed that default mode network and visual network of schizophrenia patients have aberrant connections in brain networks. In conclusion, this method can identify schizophrenia effectively and visualize the abnormal brain network, and it has important clinical application value.
