Method for regulating a multi-modal hyperbaric chamber and multi-modal hyperbaric chamber

Abstract

The application provides a multi-modal micro-high-pressure oxygen cabin adjusting method and a multi-modal micro-high-pressure oxygen cabin, which can be applied to the technical field of micro-high-pressure oxygen cabins. The adjusting method comprises the following steps: synchronously acquiring blood oxygen saturation, myoelectric signals, electroencephalogram signals and electrocardiogram signals of an object in the cabin; constructing input features according to the blood oxygen saturation, the myoelectric signals, the electroencephalogram signals and the electrocardiogram signals of the object in the cabin; processing the input features by using a neural network to obtain metabolic indexes of different levels; the metabolic indexes represent metabolic states of the object in the cabin; and outputting control signals according to the metabolic indexes, which are used for adjusting cabin environmental parameters of the multi-modal micro-high-pressure oxygen cabin.

Claims

1. A method of regulating a multi-modal micro-hyperbaric chamber, characterized in that, The adjustment method includes: Simultaneously acquire blood oxygen saturation, electromyography (EMG) signals, electroencephalogram (EEG) signals, and electrocardiogram (ECG) signals of the subjects inside the cabin; Input features are constructed based on the blood oxygen saturation, electromyography (EMG) signals, electroencephalography (EEG) signals, and electrocardiogram (ECG) signals of the objects inside the cabin; The input features are processed using a neural network to obtain metabolic indicators at different levels; these metabolic indicators characterize the metabolic state of the object inside the cabin. Based on the metabolic indicators, a control signal is output to adjust the internal environmental parameters of the multimodal micro hyperbaric oxygen chamber. The step of constructing input features based on the blood oxygen saturation, electromyography (EMG), electroencephalography (EEG), and electrocardiography (ECG) signals of the object inside the cabin includes: Based on the blood oxygen saturation, electromyography (EMG), electroencephalography (EEG), and electrocardiogram (ECG) signals of the objects inside the cabin, calculate the characteristic parameters of each of the following: The input features are constructed based on the characteristic parameters of blood oxygen saturation, electromyography (EMG), electroencephalography (EEG), and electrocardiography (ECG). The step of calculating the characteristic parameters of each of the blood oxygen saturation, electromyography (EMG), electroencephalography (EEG), and electrocardiogram (ECG) signals based on the blood oxygen saturation, EMG, EEG, and ECG signals of the object inside the cabin includes: A first characteristic parameter is calculated based on the blood oxygen saturation of the object inside the cabin; the first characteristic parameter includes the trend slope of blood oxygen saturation. A second characteristic parameter is calculated based on the electromyographic signals of the object inside the cabin; the second characteristic parameter includes muscle oxygen consumption rate. A third characteristic parameter is calculated based on the EEG signals of the object inside the cabin; the third characteristic parameter includes the energy ratio of the first frequency band and the energy ratio of the second frequency band. Based on the electrocardiogram signal of the object inside the cabin, a fourth characteristic parameter is calculated; the fourth characteristic parameter includes ST segment offset, root mean square of the time interval difference between consecutive heartbeats, percentage of the number of time intervals between adjacent consecutive heartbeats with a time interval difference greater than 50ms to the total number of time intervals between consecutive heartbeats, and the ratio of high-frequency wave power to low-frequency wave power.

2. The conditioning method of claim 1, wherein, The input features are constructed based on the characteristic parameters of blood oxygen saturation, electromyography (EMG), electroencephalography (EEG), and electrocardiography (ECG), including: The characteristic parameters of blood oxygen saturation, electromyography, electroencephalography and electrocardiography are preprocessed. Based on the characteristic parameters of the preprocessed blood oxygen saturation, electromyography, electroencephalography and electrocardiography signals, the input features are constructed with a preset time step.

3. The conditioning method of claim 2, wherein, The preprocessing of characteristic parameters of blood oxygen saturation, electromyography (EMG), electroencephalography (EEG), and electrocardiography (ECG) signals includes: Subtract the baseline values ​​from the characteristic parameters of blood oxygen saturation, electromyography, electroencephalography, and electrocardiography, and divide by the standard deviation of the baseline values. Wherein, the baseline value of each feature parameter is the expected value of each feature parameter when the object inside the cabin rests for a preset period of time, and the standard deviation of the baseline value of each feature parameter is the standard deviation of the expected value of each feature parameter when the object inside the cabin rests for a preset period of time.

4. The adjustment method according to claim 2, characterized in that, The preprocessing of characteristic parameters of blood oxygen saturation, electromyography (EMG), electroencephalography (EEG), and electrocardiography (ECG) signals includes: If one or more characteristic parameters of blood oxygen saturation, electromyography, electroencephalography, and electrocardiography are missing, interpolation is performed on the one or more characteristic parameters.

5. The adjustment method according to claim 2, characterized in that, The input features are constructed based on the characteristic parameters of preprocessed blood oxygen saturation, electromyography, electroencephalography, and electrocardiography signals, with a preset time step, including: The time features of the preprocessed second feature parameter, the preprocessed third feature parameter, and the preprocessed fourth feature parameter are time-aligned with the time features of the preprocessed first feature parameter.

6. The adjustment method according to claim 1, characterized in that, The process of using a neural network to process the input features to obtain metabolic indicators includes: One-dimensional convolution is used to extract local correlation features of the input features; Using an encoder, the delayed correlation features of the input features are extracted; The local correlation features and the delayed correlation features are weighted and fused using an attention weight layer to obtain multidimensional fused features; The metabolic index is obtained by decoding the multidimensional fusion features using a decoder.

7. The adjustment method according to claim 1, characterized in that, The adjustment method further includes: Based on a preset duration, the arterial stiffness of the objects inside the cabin is periodically determined; Based on the preset duration and preset pressure gradient, the pressure of the micro hyperbaric oxygen chamber is periodically increased; If the arterial stiffness is greater than or equal to a preset increase in arterial stiffness in the previous cycle, a protection strategy is triggered; the protection strategy includes reducing the preset pressure gradient.

8. The adjustment method according to claim 7, characterized in that, The method of periodically determining the arterial stiffness of the object inside the cabin based on a preset time interval includes: Calculate pulse wave conduction velocity based on the blood oxygen saturation and electrocardiogram signal of the object inside the cabin; The arterial stiffness of the object inside the cabin is determined based on the pulse wave propagation velocity.

9. A multimodal micro hyperbaric oxygen chamber regulated using the regulation method of any one of claims 1-8, characterized in that, The multimodal micro-hyperbaric oxygen chamber includes: The chamber of a micro-hyperbaric oxygen chamber; The physiological multi-parameter acquisition unit is used to acquire blood oxygen saturation, electromyography signals, electroencephalogram signals, and electrocardiogram signals of the subjects inside the cabin; An oxygen generator is used to supply oxygen to the objects inside the cabin. The main control unit is used to control the physiological multi-parameter acquisition unit using the fusion control unit to simultaneously acquire the blood oxygen saturation, electromyography (EMG), electroencephalography (EEG), and electrocardiogram (ECG) signals of the subject inside the chamber; to construct input features based on the blood oxygen saturation, EMG, EEG, and ECG signals of the subject inside the chamber; to process the input features using a neural network to obtain metabolic indicators at different levels; the metabolic indicators characterize the metabolic state of the subject inside the chamber; and to output control signals based on the metabolic indicators to adjust the environmental parameters inside the multimodal hyperbaric oxygen chamber.

Images