A multi-wavelength adaptive transcranial light bio-modulation device based on non-invasive brain-computer interface

Abstract

The application belongs to the technical field of photobiomodulation, and particularly relates to a multi-wavelength self-adaptive transcranial photobiomodulation device based on a non-invasive brain-computer interface, which comprises: a zoom optical module, which is used for zooming collimated laser under the control of a main controller, and is used for detecting the size and energy of a light spot in real time to achieve the required size and energy of the light spot; and a spatial light modulation module, which is used for converting the zoomed light into linearly polarized light, and is used for modulating the linearly polarized light through a spatial light modulator under the control of the main controller, and irradiating the target area; wherein the main controller is used for comparing the newly-acquired electroencephalogram signal with a normal electroencephalogram signal, calculating a gray-scale image for correcting the difference value and the output power of each wavelength laser, and applying the electric signal obtained by converting the gray-scale image to the spatial light modulator to adjust the transverse gradient distribution and longitudinal focusing depth of the spatial light field of the output laser. The application can improve the accuracy and safety of transcranial photobiomodulation.

Description

Technical Field

[0001] This invention belongs to the field of photobiological modulation technology, and more specifically, relates to a multi-wavelength adaptive transcranial photobiological modulation device based on a non-invasive brain-computer interface. Background Technology

[0002] Depression is a mental disorder characterized by significant and persistent low mood, lethargy, irritability, and fatigue. According to the WHO, there are more than 300 million people with depression worldwide, and nearly 800,000 people with depression commit suicide each year. It is projected that by 2030, depression will become the leading cause of global disease burden, seriously affecting patients' quality of life and lifespan, and imposing a heavy economic burden on families and society.

[0003] Alzheimer's disease (AD) is a chronic neurodegenerative disease characterized by progressive cognitive impairment. According to Alzheimer's Disease International, a new AD patient is diagnosed globally every 3 seconds. As of 2021, the number of people with AD and other dementias in my country was approaching 17 million. It is projected that by 2050, the number of AD patients worldwide will increase to 139 million. AD patients experience memory loss or cognitive impairment, leading to a decline in their basic ability to live independently. This not only severely impacts their health and lives but also requires long-term family care, seriously endangering public health and hindering sustainable social development.

[0004] Current treatment standards for neuropsychiatric disorders primarily employ methods such as medication, psychotherapy, and physical therapy. Among these main treatment methods, medication is the first-line approach; however, its efficacy is unsatisfactory, with problems including low clinical cure rates, prominent residual symptoms, high relapse rates, and the development of drug resistance. Psychotherapy has drawbacks such as long treatment cycles and high relapse rates, while physical therapy can easily cause adverse reactions such as mild headaches, nausea, and fever. Therefore, further exploration of new therapeutic targets and the establishment of new and effective treatment measures are of significant clinical importance and practical necessity.

[0005] Low-intensity phototherapy, as a physical therapy method, is a novel treatment approach that utilizes low-power lasers or light-emitting diodes for neuromodulation. By stimulating nerve cells with light in the red to near-infrared wavelengths through the skull, it can regulate nerve cell function, producing beneficial therapeutic results, including pain or inflammation reduction, immune modulation, and improved cognitive function. It is a non-invasive, non-traumatic physical therapy, characterized by safety, convenience, and non-invasiveness, offering significant advantages in the treatment of neuropsychiatric disorders. Research has found that lasers of various wavelengths, such as 650nm, 808nm, 980nm, and 1064nm, have good effects on the treatment of depression. Therefore, multiple lasers of different wavelengths can be integrated into a single light source, allowing for the selection of different wavelengths to meet diverse needs.

[0006] In studies using low-intensity phototherapy to intervene in neuropsychiatric disorders, several irradiation methods exist depending on the site of light input: transcranial photobiomodulation, transnasal photobiomodulation, and peripheral skin photobiomodulation. Neuropsychiatric disorders are often accompanied by pathological changes in specific cerebral cortex functions. Transcranial photobiomodulation allows optical intervention devices to precisely and directly project light beams onto the target brain region based on anatomical positioning. This method can directly enhance the activity of cytochrome C oxidase (CCO) in the mitochondria of target cortical neurons, achieving truly precise regulation.

[0007] However, existing transcranial photobiomodulation devices based on low-intensity phototherapy lack spatial resolution and struggle to address the asymmetry of brain functional areas. Most existing headband-type transcranial photobiomodulation devices use fixed-array LEDs, producing uniformly distributed and relatively fixed-position irradiation. However, many neuropsychiatric diseases exhibit significant lateralization of brain regions; using existing devices for uniform irradiation may exacerbate this imbalance between the left and right hemispheres. Current technology requires a means to regulate energy distribution within the same irradiation area. Furthermore, infrared light has a strong thermal effect on the skin; high laser power may damage the skin, while low power results in insufficient power density to penetrate the skull. The laser focusing depth of existing devices is determined by the physical parameters of the lens and cannot be dynamically adjusted. Increasing penetration depth often requires increasing surface power, which can lead to overheating or even burns of the scalp. Existing devices lack irradiation depth adjustment mechanisms; increasing irradiation depth requires increasing laser output power, making selective intervention at penetration depth difficult.

[0008] Therefore, the field urgently needs a more precise and safer transcranial photobiological modulation method. Summary of the Invention

[0009] In view of the above-mentioned defects or improvement needs of the existing technology, the present invention provides a multi-wavelength adaptive transcranial photobiomodulation device based on non-invasive brain-computer interface, which aims to improve the accuracy and safety of transcranial photobiomodulation.

[0010] To achieve the above objectives, according to one aspect of the present invention, a multi-wavelength adaptive transcranial photobiomodulation device based on a non-invasive brain-computer interface is provided, comprising: an optical fiber bundle-combining structure, an optical fiber bundle-expanding and collimating structure, a zoom optical module, a spatial light modulation module, an EEG signal feedback component for acquiring EEG signals, and a main controller. Among them, the fiber bundle combining structure is used to collimate and couple lasers of various wavelengths emitted by semiconductor lasers of multiple wavelengths in the same emission direction into the fiber; the fiber bundle expanding and collimating structure is used to expand and collimate the laser in the input fiber. The zoom optical module is used to zoom the laser beam after it is aligned with the main controller through the zoom component. The preset portion of the zoomed light is reflected into the spot collector through the reflector to detect the spot size and energy in real time and feed it back to the main controller to achieve the required spot size and energy. The remaining light passes through the reflector and enters the spatial light modulation module. The required spot size is the same as the size of the target illumination area on the forehead, and the energy is determined according to the real-time adjustment effect. The spatial light modulation module converts received light into linearly polarized light using a polarizer. Under the control of the main controller, the linearly polarized light is modulated by a spatial light modulator, and the modulated light illuminates the target illumination area. Specifically, the main controller compares newly acquired EEG signals with normal EEG signals, calculates the characteristic signal difference between the current brain state and the healthy brain state, and dynamically calculates the grayscale image for the next frame used to correct the difference and the output power of the current wavelength laser according to a preset mapping rule. The energy is determined by the output power of the current wavelength laser, and the output of the current wavelength laser is controlled. The grayscale image is converted into an electrical signal and applied to the spatial light modulator. After receiving the electrical signal, the spatial light modulator refreshes the state of its surface liquid crystal lattice and adjusts the lateral gradient distribution and longitudinal focusing depth of the spatial light field of its output laser.

[0011] Furthermore, the fiber optic bundle structure includes: a fast-axis collimating lens, a slow-axis collimating lens, a focusing lens, a fiber optic patch cord and an input fiber, a fused biconical taper fiber coupler and an output fiber, arranged sequentially after each wavelength of the semiconductor laser; A forehead illumination array is formed by arranging semiconductor lasers of multiple wavelengths with the same emission direction; The fast-axis collimating lens and the slow-axis collimating lens are used to reduce the fast-axis divergence angle and the slow-axis divergence angle of the laser, respectively; the focusing lens is used to focus the beam; the fiber jumper is used to introduce the focused beam into the input fiber; the fused taper fiber coupler is used to couple the laser in each input fiber by fused fiber stacking, and output it through the output fiber.

[0012] Furthermore, the plurality of wavelengths includes wavelengths of 650nm, 808nm, 850nm, 980nm and 1064nm.

[0013] Furthermore, the zoom optical module includes: a concave lens, a convex lens, a plano-convex lens and a reflector arranged sequentially along the light direction, a CCD camera, and a light spot energy and illumination area monitoring unit; The main controller controls the area of ​​the light spot input to the spatial light modulator by adjusting the positions of the concave lens and the plano-convex lens, ensuring that the light beam is parallel. Then, the light is output through a reflector that can transmit a preset percentage of the light beam, and the reflected light of the remaining beam enters the CCD camera to collect the light spot. The collected signal is input to the light spot energy and illumination area monitoring unit for further analysis to calculate the actual light spot energy and area, and is fed back to the main controller to adjust the positions of the concave lens and the plano-convex lens.

[0014] Furthermore, it also includes: bilateral temporal region laser irradiation structures; Each temporal laser irradiation structure includes a semiconductor laser tube of the aforementioned multiple wavelengths, a beam expander, and a focusing lens; Multiple single tubes of different wavelengths are evenly arranged in the target illumination area of ​​the temporal region. The beam is expanded by a beam expander and then focused by a focusing lens, so that the output spot of each single tube covers the target illumination area of ​​the temporal region. The main controller is also used to obtain the output power of each wavelength laser in the temporal region based on the feature signal difference mapping according to the preset mapping rules.

[0015] Furthermore, based on the difference in the characteristic signals, the main controller, through a preset mapping rule, if the parsed newly acquired EEG signal shows that the activity level of the left prefrontal cortex is significantly lower than that of the right side in the current brain state, generates a gradient light field instruction and converts the target light field parameters into a grayscale image, which is then loaded into each pixel unit of the spatial light modulator in the form of an electrical signal.

[0016] Furthermore, based on the difference in the feature signals, the main controller, through a preset mapping rule, if it is determined that the distribution of light energy in the tissue depth direction needs to be adjusted, selects a preset wavelength laser as the current working light source and generates a grayscale image corresponding to the target intervention depth, which is then loaded into each pixel unit of the spatial light modulator in the form of an electrical signal. Preferably, the preset wavelength laser is a 1064nm wavelength laser.

[0017] Furthermore, the preset mapping rule is a pre-trained deep learning model.

[0018] In summary, compared with the prior art, the technical solutions conceived by this invention have the following main advantages: 1. This invention proposes a multi-wavelength adaptive transcranial photobiomodulation device based on a non-invasive brain-computer interface. During operation, it follows a closed-loop process of "acquisition-analysis-decision-modulation." An EEG signal feedback component acquires EEG signals from the prefrontal cortex in real time (for biomodulation). By comparing newly acquired EEG signals with normal EEG signals, the characteristic signal difference between the current brain state and the healthy brain state is calculated. According to a preset mapping rule, a grayscale image is generated, which is used by a spatial light modulator to generate a target light intensity distribution, illuminating the frontal region. The introduction of a spatial light modulator allows the transcranial photobiomodulation device to move beyond fixed spot shapes and single focusing depths, enabling sub-millimeter-level light field manipulation. Furthermore, the laser input to the spatial light modulator is pre-modulated under the control of the main controller through an optical fiber combining structure, an optical fiber expanding and collimating structure, and a zoom optical module. The resulting laser possesses the required wavelength, spot size, and energy, improving the flexibility of photobiomodulation.

[0019] 2. In order to improve the adjustment effect, the present invention also proposes to have multiple wavelength laser tubes on both sides of the temporal region of the human body, and each tube is expanded and focused separately so that the output light spot of each wavelength tube can cover the entire temporal region. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of a multi-wavelength adaptive transcranial photobiomodulation device based on a non-invasive brain-computer interface, provided as an embodiment of the present invention.

[0021] In all the accompanying drawings, the same reference numerals are used to denote the same elements or structures, wherein: 1 is the forehead illumination array, 2 is the fast-axis collimating lens, 3 is the slow-axis collimating lens, 4 is the focusing lens, 5 is the fiber optic patch cord, 6 is the input fiber, 7 is the fused biconical fiber coupler, 8 is the output fiber, 9 is the negative lens, 10 is the cemented doublet lens, 11 is the guide rail, 12 is the movable frame, 13 is the concave lens, 14 is the convex lens, 15 is the fixed frame, 16 is the plano-convex lens, 17 is the reflecting mirror, 18 is the polarizer, 19 is the spatial light modulator, 20 is the CCD camera, 21 is the spot energy and illumination area monitoring unit, 22 is the semiconductor laser tube, 23 is the beam expander, 24 is the focusing lens, 25 is the temporal laser illumination structure, 26 is the electrode array, 27 is the wire, 28 is the signal output line, 29 is the bilateral temporal optical signal controller, 30 is the brain-computer interface signal processor, and 31 is the main controller. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0023] Example 1 A multi-wavelength adaptive transcranial photobiological modulation device based on a non-invasive brain-computer interface includes: an optical fiber bundle-combining structure, an optical fiber bundle-expanding and collimating structure, a zoom optical module, a spatial light modulation module, an EEG signal feedback component for acquiring EEG signals, and a main controller. Among them, the fiber bundle combining structure is used to collimate and couple lasers of various wavelengths emitted by semiconductor lasers of multiple wavelengths in the same emission direction into the fiber; the fiber bundle expanding and collimating structure is used to expand and collimate the laser in the input fiber. The zoom optical module is used to zoom the laser beam after it is aligned with the main controller through the zoom component. The preset portion of the zoomed light is reflected into the spot collector through the reflector to detect the spot size and energy in real time and feed it back to the main controller to achieve the required spot size and energy. The remaining light passes through the reflector and enters the spatial light modulation module. The required spot size is the same as the size of the target illumination area on the forehead, and the energy is determined according to the real-time adjustment effect. The spatial light modulation module converts received light into linearly polarized light using a polarizer. Under the control of the main controller, the linearly polarized light is modulated by a spatial light modulator, and the modulated light illuminates the target illumination area. Specifically, the main controller compares newly acquired EEG signals with normal EEG signals, calculates the characteristic signal difference between the current brain state and the healthy brain state, and dynamically calculates the grayscale image for the next frame used to correct the difference and the output power of the current wavelength laser according to a preset mapping rule. The energy is determined by the output power of the current wavelength laser, and the output of the current wavelength laser is controlled. The grayscale image is converted into an electrical signal and applied to the spatial light modulator. After receiving the electrical signal, the spatial light modulator refreshes the state of its surface liquid crystal lattice and adjusts the lateral gradient distribution and longitudinal focusing depth of the spatial light field of its output laser.

[0024] A spatial light modulator (SLM) is an advanced optical device capable of real-time modulation of the spatial distribution of light waves. It controls the state of liquid crystal molecules or micromirrors via electrical signals, thereby altering the amplitude, phase, polarization state, or intensity of the incident light beam at each pixel on its cross-section. This allows for complex shaping of the beam's wavefront without loss of light energy. This embodiment introduces a spatial light modulator into the field of laser medicine, meaning that treatment devices are no longer limited to fixed spot shapes and single focusing depths, but instead possess the ability to manipulate the light field at the sub-millimeter level.

[0025] Brain-computer interface (BCI) is a cutting-edge technology that establishes a direct communication channel between the brain and external devices. In non-invasive BCI systems, electroencephalography (EEG) is the most crucial and widely used signal detection method. EEG technology directly records the weak electrical signals generated by the synchronous synaptic activity of pyramidal cell groups in the cerebral cortex through electrodes placed on the scalp. EEG boasts ultra-high temporal resolution (milliseconds). Cognitive activity, emotional fluctuations, and transient responses to neuromodulation interventions often occur within tens to hundreds of milliseconds, and EEG can capture these dynamic changes with extremely low latency. Furthermore, the frequency domain characteristics of EEG signals (such as the power spectral density of the Delta, Theta, Alpha, Beta, and Gamma bands) have a clear mapping relationship with human arousal levels, attention allocation, and emotional valence. For example, the power asymmetry in the prefrontal alpha band (8-13Hz) plays an important role in the diagnosis and assessment of depression and in guiding treatment. Introducing high temporal resolution EEG technology into devices provides core hardware support for achieving instantaneous perception of brain states.

[0026] The method in this embodiment is based on a non-invasive brain-computer interface, which acquires EEG signals in real time as feedback for photobiological regulation. The main controller adjusts the light power and light field distribution in real time based on the feedback EEG signals to achieve the desired adaptive real-time regulation effect.

[0027] This can be used as a preferred implementation method, such as Figure 1 As shown, the fiber optic bundle structure includes: a fast-axis collimating mirror 2, a slow-axis collimating mirror 3, a focusing lens 4, a fiber optic patch cord 5, and an input fiber 6, a fused biconical taper fiber coupler 7, and an output fiber 8, arranged sequentially after each wavelength of the semiconductor laser; Multiple semiconductor lasers of different wavelengths are arranged in the same emission direction to form a forehead irradiation array 1; The fast-axis collimating lens and the slow-axis collimating lens are used to reduce the fast-axis divergence angle and the slow-axis divergence angle of the forehead illumination array, respectively; the focusing lens is used to focus the beam; the fiber optic patch cord is used to introduce the focused beam into the input fiber; the fused taper fiber coupler is used to couple the laser in each input fiber in a fused taper manner, and output it through the output fiber.

[0028] As a preferred embodiment, the aforementioned wavelengths include 650nm, 808nm, 850nm, 980nm, and 1064nm. The focusing lens, fiber optic patch cord, input fiber, fused biconical taper fiber coupler, and output fiber are constructed by coupling each single tube of an array of these five wavelengths into the input fiber, then bundling the fibers together using a fused biconical taper, and outputting through the same output fiber. Single-wavelength or multi-wavelength combinations can be selected to be activated according to treatment needs.

[0029] In practical implementation, the fiber optic beam expanding and collimating structure may include a negative lens 9 and a cemented doublet lens 10. The negative lens 9 expands the beam output from the fiber for subsequent beam shaping. The cemented doublet lens 10 is a composite lens made of two lenses of different materials cemented together, which can effectively correct chromatic aberration. Furthermore, the negative lens 9 and the cemented doublet lens 10 work together to achieve collimation.

[0030] As a preferred embodiment, the zoom optical module includes: a guide rail 11, a movable lens frame 12, a fixed lens frame 15, a concave lens 13, a convex lens 14, a plano-convex lens 16 and a reflector 17 arranged sequentially along the light direction, a CCD camera 20, and a light spot energy and illumination area monitoring unit 21.

[0031] The main controller 31 adjusts the position of the concave lens 13 and plano-convex lens 16 by adjusting the movable lens frame 12 to control the output light spot area and ensure that the output beam is a parallel beam. The output beam is then output through a reflector 17 that can transmit, for example, 99% of the beam, with 1% of the reflected light entering the CCD camera 20 to collect the light spot. The signal is then input to the light spot energy and illumination area monitoring unit 21 for further analysis, calculating the actual light spot energy and area, and feeding back to the main controller to adjust the position of the concave lens and plano-convex lens.

[0032] The spatial light modulation module includes a polarizer 18 and a spatial light modulator 19; the polarizer 18 modulates the input beam (99% transmitted light) into linearly polarized light, which then enters the spatial light modulator 19 for further modulation. The main controller receives EEG signals from the EEG signal feedback component, compares the new EEG signals with normal EEG signals, converts the calculated target light field parameters into a grayscale image, and then loads them as electrical signals to each pixel unit of the spatial light modulator. The incident linearly polarized light hits the liquid crystal modulation surface of the spatial light modulator 19. Each pixel independently changes its refractive index according to the received electrical signal, thereby independently delaying the phase or weakening the amplitude of each pixel region on the laser wavefront. The reflected beam after being modulated at the pixel level undergoes coherent superposition in space, and finally presents a target light field with a preset spatial intensity gradient on a macroscopic scale. After the beam with the target light field shines on the target forehead, it generates a new EEG signal feedback (through the brain-computer signal feedback structure). The main controller generates a new grayscale image and the output power of the selected wavelength laser based on the comparison between the new EEG signal and the normal EEG signal, realizing the adaptive real-time adjustment of the spatial light modulator.

[0033] Therefore, by coarse adjustment of the zoom optical module (light energy coverage of the target illumination area and energy) and fine adjustment of the spatial light modulator (adjusting the light intensity distribution in this area), the theoretically required light spot is output.

[0034] As a preferred embodiment, it may also include: a bilateral temporal laser irradiation structure 25 and a bilateral temporal optical signal controller 29; Each temporal laser irradiation structure includes a semiconductor laser single tube 22 with multiple wavelengths, a beam expander 23, and a focusing lens 24. Multiple single tubes of different wavelengths are evenly arranged in the temporal target illumination area. The beam is expanded by a beam expander and then focused by a focusing lens, so that the output spot of each single tube covers the temporal target illumination area. The main controller is also used to obtain the output power of each wavelength laser in the temporal region based on the aforementioned characteristic signal difference mapping according to a preset mapping rule. The dual temporal optical signal controller 29 works in conjunction with the main controller to control the dual temporal laser illumination structure 25.

[0035] In the pathological evolution of neuropsychiatric disorders (such as major depressive disorder, Alzheimer's disease, and sleep disorders), the prefrontal cortex is not an isolated lesion, but is often accompanied by overall functional disorders of the limbic system and the lateral fissure core brain network. The temporal region is the closest transcranial optical window to the hippocampus (responsible for memory and emotional processing) and the amygdala (responsible for fear and anxiety). Through temporal irradiation, highly penetrating near-infrared lasers can directly penetrate the relatively thin temporal bone, modulating the metabolic level of the limbic system, thereby synergizing with frontal regulation to alleviate anxiety, depression, and cognitive decline, and enhance photobiological modulation effects.

[0036] In addition, in specific implementation, the brain-computer signal feedback component includes an electrode array 26, wires 27, signal output lines 28, and a brain-computer signal processor 30. The electrode array 26 is placed at specific points on the head to collect brain signals. The collected signals are processed by the brain-computer signal processor 30 to remove noise and extract signal features. The main controller extracts feature signals from the real-time brain signals and compares them with a preset normal healthy brain database (or the patient's own baseline target state) to calculate the difference between the current brain state and the healthy state. The main controller inputs the difference into the built-in optimization control algorithm, and according to the predetermined mapping rules, dynamically calculates the holographic grayscale image for the next frame used to correct the difference and the output power of each wavelength laser. After receiving a new signal, the SLM refreshes the surface liquid crystal lattice state and adjusts the lateral gradient distribution and longitudinal focusing depth of the spatial light field.

[0037] For example, if the pathological features are extremely significant (large difference in characteristic signals), the driving current of the semiconductor laser is automatically increased. The main controller adjusts the laser output parameters within a preset output power range based on the deviation of the EEG characteristics, so as to quickly activate the CCO with a relatively high-energy beam. As the real-time EEG gradually approaches the normal value (the difference decreases), the output power is smoothly attenuated to prevent excessive irradiation from causing stimulation tolerance. This process is continuously iterated until the real-time EEG characteristics converge to the normal range.

[0038] In summary, the multi-wavelength adaptive transcranial photobiological modulation device of this embodiment includes a light source section, an optical path shaping section, a modulation section, and a control feedback section. The laser emitted by the multi-wavelength laser is transmitted through an optical fiber; after collimation and beam expansion, it enters the zoom optical module, and after adjusting the spot size, it is incident on the modulation surface of the spatial light modulator; the spatial light modulator reflects and modulates the incident light according to the loaded grayscale image, and the modulated beam is projected onto the user's forehead area through the output lens.

[0039] The brain-computer interface (BCI) signal feedback structure employs a full-head mesh flexible support structure, with 32 electrode channels arranged according to the international 10-20 standard system. These 32 channels cover key areas of the entire brain, aiming to achieve precise extraction of cortical electrical activity characteristics through high spatial resolution signal acquisition. The selection of 32 channels not only enables real-time monitoring of the direct response in the prefrontal cortex (the treatment target area), but also allows for the construction of a brain network connectivity map using whole-brain signals, assessing the regulatory effects of phototherapy on distal brain regions and overall function. Therefore, high spatial resolution whole-brain signal acquisition enables focused monitoring of treatment feedback signals.

[0040] The electrode support structure uses transparent material or a ring structure near the Fp1 and Fp2 sites to ensure that the laser can pass through the electrode gap and accurately hit the forehead, and that the electrodes do not block the beam.

[0041] The multi-wavelength adaptive transcranial photobiological modulation device operates following a closed-loop process of "acquisition-analysis-decision-modulation". The EEG signal feedback component acquires EEG signals from the prefrontal cortex (for biomodulation) and EOG signals from the eye muscles in real time (for retinal protection). The brain-computer interface signal processor 30 performs noise removal on the signals and then uses a deep learning model to extract and analyze features from the EEG signals.

[0042] The main controller (which can use a pre-trained deep learning model for decision-making) generates corresponding control commands based on the analyzed brain state: If the analysis shows that the activity level in the left prefrontal cortex is significantly lower than that on the right, the main controller generates a gradient light field command, converts the target light field parameters into a grayscale image, and then loads it into each pixel unit of the spatial light modulator in the form of an electrical signal. This causes the light intensity distribution emitted from the spatial light modulator to exhibit a "stronger on the left and weaker on the right" distribution, thereby enhancing stimulation on the left side and inhibiting excessive activity on the right side. If the analysis indicates that the distribution of light energy in the tissue depth direction needs to be adjusted, the main controller selects a 1064nm laser as the current working light source and generates a corresponding grayscale image based on the preset target intervention depth, loading the grayscale image into the spatial light modulator. The spatial light modulator adjusts the phase of the incident beam at different positions according to the grayscale image, causing the emitted beam to converge or diverge, thereby changing the convergence position of the beam before entering the head tissue and achieving adjustment of the light energy depth distribution. The main controller can change the convergence degree of the beam by updating the grayscale image to adapt to the irradiation requirements of different target intervention areas.

[0043] In a preferred implementation, regardless of the adjustment mode, a set of rapidly changing random interference patterns is overlaid on the base grayscale image. The main controller continuously switches these patterns at a frequency of 60Hz or higher. Although the microstructure of these patterns changes randomly in each frame, the total light intensity they impart to the skin remains consistent macroscopically. This high-frequency switching allows the local hotspots generated by the laser at the microscopic level to cancel each other out, effectively preventing local heat accumulation and ensuring skin safety.

[0044] While the light field is being output, the safety mechanism operates independently. Based on general facial anatomy data, a black area with a brightness of 0 is loaded by default in the area of ​​the SLM display image (output light field) that coincides with the human eye, physically blocking light from reaching the eye socket. During the process, if the user is detected to have continuous high-amplitude electrooculogram (EOG) signals (indicating photophobia or blinking), it is determined that the spatial position of the preset black area boundary is too low, and light is touching the edge of the retina. The main controller automatically corrects the edge coordinates of the black area in the display image (output light field) by adjusting the grayscale image, expanding the black screen area upwards until the EOG signal returns to normal. This process does not require camera intervention and utilizes biofeedback to achieve personalized retinal protection zone settings.

[0045] Through the above mechanism, non-invasive precision phototherapy for neuropsychiatric disorders (such as depression and Alzheimer's disease) is achieved. Compared with traditional devices, this device can dynamically project a corresponding fine light field based on the patient's real-time brain activity topography, and ensures high safety and deep effectiveness of multi-wavelength laser therapy.

[0046] Therefore, the multi-wavelength adaptive transcranial photobiological modulation device proposed in this embodiment has the function of continuous gradient light field modulation based on frontal lobe asymmetry. The main controller calculates the activity difference between the left and right prefrontal lobes through electroencephalogram signals. When an imbalance in excitability between the left and right hemispheres is detected, a grayscale image with a continuous gradient change is generated based on the mapping rule and loaded onto the spatial light modulator to project a gradient light field with gradually changing light intensity from left to right or from right to left onto the forehead to reshape the brain region balance.

[0047] Furthermore, the multi-wavelength adaptive transcranial photobiological modulation device proposed in this embodiment also possesses a depth-selective modulation function based on wavefront shaping. The main controller, according to the depth requirements of the treatment target (such as the superficial cortex or deep sulci), generates a corresponding grayscale image using a spatial light modulator based on mapping rules, thereby achieving focused or defocused control of light energy at different tissue depths.

[0048] Furthermore, as a preferred embodiment, the multi-wavelength adaptive transcranial photobiological modulation device proposed in this embodiment has a thermally safe laser speckle elimination function; the main controller calculates multiple grayscale images for the same target light intensity distribution. Each image forms the same or approximately the same target light intensity distribution macroscopically, but produces different microscopic speckle patterns, which are used to suppress speckle noise and prevent the accumulation of local skin hotspots.

[0049] Furthermore, as a preferred embodiment, the multi-wavelength adaptive transcranial photobiological modulation device proposed in this embodiment also possesses a safety mechanism to protect the retina. The main controller has a pre-stored facial anatomical model and, by default, overlays completely black pixels below the light field pattern loaded by the spatial light modulator, forming a dark area targeting the orbital region. Simultaneously, it uses forehead electrodes to monitor electrooculography (EOG) signals in real time. When characteristic high-frequency blinking or extraocular muscle tension potentials are detected, it is determined that the current illumination boundary is too low and interferes with vision. The control unit immediately triggers a safety algorithm to adjust the coverage height of the black pixel area upwards until the EOG signal returns to the baseline level, thereby achieving personalized retinal protection.

[0050] Therefore, the device in this embodiment achieves precise control of the spatial light field. Existing devices mostly use LED arrays or fixed-focal-length lasers, which cannot accurately match the complex geometry and functional heterogeneity of the cerebral cortex. This invention, by introducing a spatial light modulator, achieves power distribution control in the lateral direction and overcomes the limitation of fixed focal length in traditional optical systems. Utilizing wavefront phase shaping technology, it achieves selective intervention in the depth direction. This allows light energy to be precisely focused on specific brain regions, significantly improving the targeting and energy utilization of photobiological modulation.

[0051] Furthermore, this invention establishes a closed-loop control mechanism based on brain-computer interface signal feedback. Most existing devices operate in an "open-loop" mode, unable to adjust parameters according to the user's real-time neurophysiological state, and their monitoring methods are limited, making it difficult to support complex neural feedback logic. By using a 32-channel full-head mesh EEG acquisition system and employing deep learning algorithms to analyze brain-computer interface signals in real time, this invention can sense the instantaneous fluctuations in the brain's state during treatment. These features are then analyzed and output as modulation parameters for the SLM (Spiritual Light Modulation), thereby achieving adaptive adjustment of the light field and more effectively treating neuropsychiatric disorders.

[0052] Furthermore, this invention also incorporates an active safety protection system and a thermal damage suppression mechanism. Existing laser treatment devices rely heavily on physical shielding and passive heat dissipation for safety, and the high coherence of the laser easily generates speckle hotspots, posing a risk of localized burns. By pre-setting the eye position in the software and acquiring electrooculography (EOG) signals, dual safety protection is achieved, ensuring retinal safety without the need for eye protection. Utilizing the high-frequency phase refresh characteristics of the SLM to disrupt the temporal coherence of the laser, localized heat accumulation caused by the speckle effect is suppressed, keeping the skin surface temperature within a safe range and improving the safety and comfort of long-term treatments.

[0053] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A multi-wavelength adaptive transcranial photobiological modulation device based on a non-invasive brain-computer interface, characterized in that, include: The system includes an optical fiber bundle combining structure, an optical fiber expander and collimator structure, a zoom optical module, a spatial light modulation module, an EEG signal feedback component for acquiring EEG signals, and a main controller. Among them, the fiber bundle combining structure is used to collimate and couple lasers of various wavelengths emitted by semiconductor lasers of multiple wavelengths in the same emission direction into the fiber; the fiber bundle expanding and collimating structure is used to expand and collimate the laser in the input fiber. The zoom optical module is used to zoom the laser beam after it is aligned with the main controller through the zoom component. The preset portion of the zoomed light is reflected into the spot collector through the reflector to detect the spot size and energy in real time and feed it back to the main controller to achieve the required spot size and energy. The remaining light passes through the reflector and enters the spatial light modulation module. The required spot size is the same as the size of the target illumination area on the forehead, and the energy is determined according to the real-time adjustment effect. The spatial light modulation module converts received light into linearly polarized light using a polarizer. Under the control of the main controller, the linearly polarized light is modulated by a spatial light modulator, and the modulated light illuminates the target illumination area. Specifically, the main controller compares newly acquired EEG signals with normal EEG signals, calculates the characteristic signal difference between the current brain state and the healthy brain state, and dynamically calculates the grayscale image for the next frame used to correct the difference and the output power of the current wavelength laser according to a preset mapping rule. The energy is determined by the output power of the current wavelength laser, and the output of the current wavelength laser is controlled. The grayscale image is converted into an electrical signal and applied to the spatial light modulator. After receiving the electrical signal, the spatial light modulator refreshes the state of its surface liquid crystal lattice and adjusts the lateral gradient distribution and longitudinal focusing depth of the spatial light field of its output laser.

2. The multi-wavelength adaptive transcranial photobiological modulation device as described in claim 1, characterized in that, The fiber optic bundle structure includes: a fast-axis collimating lens, a slow-axis collimating lens, a focusing lens, a fiber optic patch cord and an input fiber, a fused biconical taper fiber coupler and an output fiber, arranged sequentially after each wavelength of semiconductor laser; A forehead illumination array is formed by arranging semiconductor lasers of multiple wavelengths with the same emission direction; The fast-axis collimating lens and the slow-axis collimating lens are used to reduce the fast-axis divergence angle and the slow-axis divergence angle of the laser, respectively; the focusing lens is used to focus the beam; the fiber jumper is used to introduce the focused beam into the input fiber; the fused taper fiber coupler is used to couple the laser in each input fiber by fused fiber stacking, and output it through the output fiber.

3. The multi-wavelength adaptive transcranial photobiological modulation device as described in claim 1, characterized in that, The multiple wavelengths include 650nm, 808nm, 850nm, 980nm and 1064nm wavelengths.

4. The multi-wavelength adaptive transcranial photobiological modulation device as described in claim 1, characterized in that, The zoom optical module includes: a concave lens, a convex lens, a plano-convex lens and a mirror arranged sequentially along the light direction, a CCD camera, and a light spot energy and illumination area monitoring unit; The main controller controls the area of ​​the light spot input to the spatial light modulator by adjusting the positions of the concave lens and the plano-convex lens, ensuring that the light beam is parallel. Then, the light is output through a reflector that can transmit a preset percentage of the light beam, and the reflected light of the remaining beam enters the CCD camera to collect the light spot. The collected signal is input to the light spot energy and illumination area monitoring unit for further analysis to calculate the actual light spot energy and area, and is fed back to the main controller to adjust the positions of the concave lens and the plano-convex lens.

5. The multi-wavelength adaptive transcranial photobiological modulation device as described in claim 1, characterized in that, Also includes: Bilateral temporal region laser irradiation structures; Each temporal laser irradiation structure includes a semiconductor laser tube of the aforementioned multiple wavelengths, a beam expander, and a focusing lens; Multiple single tubes of different wavelengths are evenly arranged in the target illumination area of ​​the temporal region. The beam is expanded by a beam expander and then focused by a focusing lens, so that the output spot of each single tube covers the target illumination area of ​​the temporal region. The main controller is also used to obtain the output power of each wavelength laser in the temporal region based on the feature signal difference mapping according to the preset mapping rules.

6. The multi-wavelength adaptive transcranial photobiological modulation device as described in claim 1, characterized in that, Based on the difference in the characteristic signals, the main controller, through a preset mapping rule, if the parsed newly acquired EEG signal shows that the activity level of the left prefrontal cortex is significantly lower than that of the right side in the current brain state, generates a gradient light field instruction and converts the target light field parameters into a grayscale image, which is then loaded into each pixel unit of the spatial light modulator in the form of an electrical signal.

7. The multi-wavelength adaptive transcranial photobiological modulation device as described in claim 1, characterized in that, Based on the difference in the characteristic signals, the main controller, through a preset mapping rule, if it is determined that the distribution of light energy in the tissue depth direction needs to be adjusted, selects a preset wavelength laser as the current working light source and generates a grayscale image corresponding to the target intervention depth, which is then loaded into each pixel unit of the spatial light modulator in the form of an electrical signal. Preferably, the preset wavelength laser is a 1064nm wavelength laser.

8. The multi-wavelength adaptive transcranial photobiological modulation device according to any one of claims 1 to 7, characterized in that, The preset mapping rules are based on a pre-trained deep learning model.

Images