Search

Abstract Introduction There remains a critical need for precise localization of the epileptogenic foci in individuals with drug-resistant epilepsy (DRE). 18F-Fluorodeoxyglucose positron emission tomography (FDG-PET) imaging can reveal hypometabolic regions during the interval between seizures in patients with epilepsy. However, visual-based qualitative analysis is time-consuming and strongly influenced by physician experience. CortexID Suite is a quantitative analysis software that helps to evaluate PET imaging of the human brain. Therefore, we aimed to evaluate the efficacy of CortexID quantitative analysis in the localization of the epileptogenic zone in patients with temporal lobe epilepsy (TLE). Methods A total of 102 patients with epilepsy who underwent 18F-FDG-PET examinations were included in this retrospective study. The PET visual analysis was interpreted by two nuclear medicine physicians, and the quantitative analysis was performed automatically using CortexID analysis software. The assumed epileptogenic zone was evaluated comprehensively by two skilled neurologists in the preoperative assessment of epilepsy. The accuracy of epileptogenic zone localization in PET visual analysis was compared with that in CortexID quantitative analysis. Results The diagnostic threshold for the difference in the metabolic Z-score between the right and left sides of medial temporal lobe epilepsy (MTLE) was calculated as 0.87, and that for lateral temporal lobe epilepsy (LTLE) was 2.175. In patients with MTLE, the area under the curve (AUC) was 0.922 for PET visual analysis, 0.853 for CortexID quantitative analysis, and 0.971 for the combined diagnosis. In patients with LTLE, the AUC was 0.842 for PET visual analysis, 0.831 for CortexID quantitative analysis, and 0.897 for the combined diagnosis. These results indicate that the diagnostic efficacy of CortexID quantitative analysis is not inferior to PET visual analysis (p > 0.05), while combined analysis significantly increases diagnostic efficacy (p

In order to reduce the error of automatic frequency control of transmission machinery under different disturbances, an automatic frequency control system of transmission machinery based on back-propagation (BP) neural network algorithm is designed. The gradient descent method and Newton method are combined to optimize the BP neural network. Aiming at the periodicity and trend of the actual mechanical vibration frequency time series, the first order backward difference is processed by the difference method, and the autoregressive sequence is derived to obtain the control model. The experimental results show that the system can realize the automatic control of transmission mechanical frequency with small error under different types of interference, and the control effect is ideal. [ABSTRACT FROM AUTHOR]