A multi-modal adaptive cursor control method and system

Abstract

The invention discloses a multimode adaptive cursor control method and system. The method comprises the following steps: respectively collecting a tongue motion magnetic sensor signal and an electroencephalogram signal; converting the tongue motion magnetic sensor signal into a first cursor control signal x1, and converting the electroencephalogram signal into a second cursor control signal x2; calculating the first cursor control signal x1 and the second cursor control signal x2 by adopting an adaptive weight fusion algorithm to obtain a fused cursor control signal (the formula is as shown in the specification), and controlling the cursor motion in a display according to the fused cursor control signal (the formula is as shown in the specification). By adoption of the method, the accuracy of cursor control is improved.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a multi-modal adaptive cursor control method and system. Background technique [0002] At present, amyotrophic lateral sclerosis (ALS) is a chronic progressive neurodegenerative disease involving the upper and lower motor neurons and the muscles of the trunk, limbs and head innervated by them. The general clinical manifestations of ALS are: the patient gradually loses the ability to move from the lower limbs to the upper limbs and trunk in a conscious state, which is very painful. The global incidence of ALS is 5-10 people / 100,000, and the prevalence is even higher in some areas. The number of patients in our country is more than 70,000, which brings a heavy burden to the family and society. Therefore, it is of great practical value to develop an effective medical assistance system to help ALS patients communicate with the outside world or achieve motor ...

AI Technical Summary

Problems solved by technology

In addition, a brain-computer interface (Brain-Computer Interface, BCI) can also be used. The brain-computer interface is an artificially built independent information transmission channel that does not depend on the peripheral nerve cells and muscle tissue of the brain. The brain can directly communicate with the outside world through the brain-computer interface. Devices interact with each other. At present, most brain-computer interfaces generate discrete control signals based on the classification of EEG signals, which is not convenient for flexible control of the cursor.

[0004] In the early stage of ALS, the tongue movement of the patient is normal, and the magnetic pole positioning technology can be used to decode the tongue movement parameters to realize cursor or mechanical arm control. The accuracy of cursor control gradually decreases until the ALS patient loses tongue movement, loses control of the cursor, and has less accuracy of cursor control

Although the use of brain-computer interface technology to achieve cursor control is not affected by peripheral muscle and nerve tissue damage, its limitations lie in: a small number of brain-computer interface studies have realized the decoding of upper limb movement parameters (including trajectory and speed) based on EEG signals, But there is still a certain distance from practicality

Because the training data of these studies were collected from healthy people, but for ALS patients, the limb movement training data is often not available, and the accuracy of cursor control achieved by the existing brain-computer interface technology is low. , For example, when the tongue and eyes are still able to move, it is not the best choice to only rely on noisy and unstable EEG signals to control the cursor

To sum up, the tongue movement tracking technology based on magnetic positioning technology and the limb movement parameter decoding technology based on EEG signals have their own advantages and disadvantages, but neither can take into account the needs of auxiliary control of ALS at different stages. low

Images