DBS closed-loop learning method in Parkinson's state based on generalized generative adversarial network

Abstract

The invention provides a DBS closed-loop learning method in a Parkinson's state based on a generalized generative adversarial network. According to the method, a generative network and a basal nucleus-thalamus-cortex neuron network built by MATLAB are creatively defined as the generalized generative network, and a closed-loop DBS deep learning algorithm based on the generalized generative adversarial network is designed. The method comprises the steps that a required data set is obtained through a basal nucleus-thalamus-cortex neuron network built by a field programmable gate array, repeated adversarial training is conducted on a generalized generative network and a discrimination network, when random noise is input to the generalized generative network, the generalized generative networkcan output a data sequence of ''falsifying the truth with the falsification'', so that a control signal capable of effectively inhibiting the Parkinson's state is obtained, and the effective control over the Parkinson state is achieved. The algorithm adopts the idea of deep learning and uses the adversarial network algorithm based on generalized generative network to carry out closed-loop DBS modulation on Parkinson's state, so as to promote the possibility of adaptive DBS technology to improve the Parkinson's state.

Description

technical field [0001] The invention relates to the fields of biomedical engineering technology and deep learning, in particular to a DBS closed-loop deep learning algorithm based on a generalized generative confrontation network in Parkinson's state. Background technique [0002] Parkinson's disease (PD) is a degenerative neurological disease caused by functional degradation of the central nervous system. Patients with Parkinson's disease present with symptoms such as resting tremor, muscle stiffness, bradykinesia, and abnormal posture and gait. Studies have shown that Parkinson's disease is associated with abnormal electrophysiological activity of neuronal network circuits in the basal ganglia of the brain. The basal ganglia region in the human brain mainly includes three parts: Subthalamic nucleus (STN), lateral globus pallidus (Globus Pallidus externa, GPe) and medial globus pallidus (Globus Pallidus, GPi), and cortical neurons can be divided into cortical vertebral bodi...

AI Technical Summary

Problems solved by technology

According to the real-time local field potential signal changes, the DBS is adaptively modulated by the generalized generative confrontation network, and the closed-loop control of the Parkinson's state is realized by using the generalized generative confrontation network, which solves the control of the traditional generative confrontation network in the Parkinson's state It is difficult to apply to the problem, and it also provides a new idea for the closed-loop control of Parkinson's state, making personalized closed-loop DBS modulation possible

Images