MicroCloud Hologram Inc. announced the launch of a multi-layer joint learning framework based on logistic regression models to construct a motion training system based on machine learning and SVM holographic brain-computer interface. Brain-computer interface is a communication technology that does not depend on people's normal peripheral nerves and muscle tissue. It is a direct connection pathway established between human or animal brain (or culture of brain cells) and external devices.

The motion training system of holographic brain-computer interface solves the difficult problem of exercise for patients with functional disorders, and stimulates, extracts and utilizes their active movement willingness. And strengthen the use of the affected limb, improve the motor function of the limb. By combining MEMS flexible microsensor array technology with BCI brain-computer interface technology, multi-source information fusion and adaptive feedback control technology, it can not only significantly improve the motor function of limbs, but also promote the reorganization of the functional dependent area of the cortex, thereby expanding the cortical motor control area of the affected limb, providing an effective tool for early rehabilitation training of patients with hand dysfunction.OLO also built a brain-computer interface experimental control platform based on holographic AR, which uses the holographic naked eye image as a visual stimulator to induce EEG signals, so that users do not need to perform visual stimulation in a fixed position, which can enhance the applicability in complex environments, so as to achieve more natural human-computer interaction.

Then the A/D sampling of the EEG signal is controlled by the motion training system of the holographic brain computer interface through digital signal processing, and the A/D sampling of The filtered EEG signal is then identified and matched by intelligent algorithm according to holographic data in holographic data tag library. Finally, the EEG holographic data is displayed and saved by complex algorithm and parallel communication. The holographic brain-computer interface motion training system based on machine Learning and SVM is composed of signal acquisition, feature extraction, feature classification and external control equipment: Signal acquisition: The brain computer interface collects signals of neuronal activity through microelectrodes implanted in the cerebral cortex; Feature extraction: The collected signals are decoded, then encoded, and converted into machine-readable instruction signals.

Common methods include fast Fourier transform (FFT), discrete Fourier transform (DFT), wavelet transform (WT), independent component analysis (ICA), common spatial mode (CSP) and some improved methods based on the above methods. Feature classification: The extracted feature signals are further classified. Commonly used classifiers include linear classifiers, support vector machines (SVM), neural networks and a combination of various classifiers.

External control device: The control process in the form of signals to the brain feedback to achieve human-computer interaction. In the field of rehabilitation medicine, the motion training system of holographic Brain-computer interface can effectively assist the rehabilitation training of neuromuscular patients such as stroke or spinal cord injury by controlling robotic arms and exoskeleton robots. With the continuous exploration of brain structure and function by modern medicine, human beings have more in-depth research on brain functional areas such as vision, heating, movement and language.

Micro-cloud holographic obtains information of these brain functional areas through brain-computer interface equipment and analyzes it, and lays out the diagnosis, screening, monitoring, treatment and rehabilitation of neurological and psychiatric diseases. The company is also exploring potential future research and application directions.