A red light therapy pod, power control adjustment method and system

Abstract

This invention relates to the field of red light therapy chamber technology, specifically to a red light therapy chamber, a power control and adjustment method, and a system. The system includes a sensing module integrated within the red light therapy chamber, comprising a multispectral imaging component, a three-dimensional depth vision component, and an infrared thermal imaging component; an execution module comprising a high-density array of miniature LED light sources; and a control module connected to the sensing and execution modules, used to construct a three-dimensional digital model and a safe power mapping map based on multidimensional biological characteristics and geometric morphological data, and to dynamically adjust the output parameters of each LED unit in the execution module using a time-division multiplexing sampling mechanism and a hierarchical clustering driven architecture; through real-time perception of skin pigmentation and curvature characteristics via multispectral imaging and three-dimensional vision, dynamically modulating the power distribution of the LED array, and combining dose reciprocity law and pulse driving strategy, to achieve precise photobiological regulation of deep tissues while ensuring epidermal safety.

Description

Technical Field

[0001] This invention relates to the field of red light therapy chamber technology, specifically to a red light therapy chamber, a power control and adjustment method, and a system. Background Technology

[0002] Red light therapy, as a non-invasive physical therapy and health management method, works by using high-energy red light of a specific wavelength to stimulate cell regeneration, promote blood circulation, relieve inflammation, and accelerate wound healing.

[0003] In practical applications, existing physiotherapy beds have a fixed light source layout that cannot be dynamically adjusted, making it impossible to flexibly adjust the position and density of the light beam according to the individual patient's lesion (such as local lesions or specific areas). This rigid design makes the treatment process lack targeting, and the light beam may not be able to fully cover the affected area, but instead scatter to non-target areas, which not only reduces the efficiency of physiotherapy, but may also cause energy waste or even cause patient discomfort.

[0004] Chinese patent CN115006732B discloses an IoT-based intelligent LED physiotherapy bed and phototherapy system. The system includes a bed frame and a support frame at the bottom. A physiotherapy chamber is located on the bed frame, with a pillow positioned above the front of the chamber. The upper part of the chamber is covered by a glass reclining panel, which is movably connected to the chamber for opening and closing. The chamber contains multiple LED physiotherapy mechanisms that can move freely to adjust the position and density of the light source. During LED irradiation therapy, to ensure adjustable light source density and position, a second drive motor rotates, driving gears to roll on a rack. Each LED physiotherapy mechanism can move along the Y-axis, allowing them to converge or disperse along this axis. A third drive motor rotates, driving rollers to roll within a groove, causing the physiotherapy components to converge or disperse along the X-axis. This ensures the light source is focused on the affected area.

[0005] While the aforementioned solution addresses the core issue of insufficient flexibility in traditional equipment by introducing intelligent, mobile LED therapy devices to dynamically adjust the distribution and intensity of the light source, in practical applications, the effectiveness of high-energy red light therapy depends on precise control of the treatment temperature. However, traditional phototherapy devices are often based on a constant light source power output, making it difficult to adaptively adjust according to the real-time light intensity applied to the body part. If the treatment temperature is too high, it may cause tissue damage or burns; while if the temperature is too low, it may reduce the treatment effect. Existing red light therapy chambers, especially those designed for whole-body irradiation, often rely on a constant light source power output, making it difficult to adaptively adjust the light intensity based on the real-time intensity applied to the affected body parts. The most prominent issue is the difference in skin color among users. According to the Fitzpatrick skin type, the density of epidermal melanin is determined, and melanin strongly absorbs light energy and converts it into heat. For individuals with darker skin, the higher melanin density leads to more light energy being absorbed by the epidermis, reducing the effective therapeutic energy reaching deeper target tissues and increasing the risk of thermal damage to the epidermis due to photothermal conversion. Conversely, for individuals with lighter skin, a fixed, lower power output may result in insufficient energy penetration, failing to achieve the desired biostimulation effect.

[0006] When the skin area to be irradiated has localized pigmentation changes such as tattoos or melasma, the fixed output of traditional equipment can cause safety issues when irradiating the tattooed or melasma areas on the human body. Specifically, tattoo ink, especially dark ink, selectively absorbs light energy of specific wavelengths, which can cause local temperature rise and burns. For active pigmented areas such as melasma, improperly controlled irradiation energy can lead to the aggravation of post-inflammatory hyperpigmentation. Summary of the Invention

[0007] The purpose of this invention is to provide a red light therapy chamber, a power control and adjustment method and system, in order to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a red light therapy power control and adjustment system, comprising: The sensing module, integrated within the red light therapy chamber, is used to acquire multidimensional biological characteristics and geometric morphological data of the treatment area. The sensing module includes a multispectral imaging component, a three-dimensional depth vision component, and an infrared thermal imaging component. The execution module includes a high-density array of miniature LED light sources for projecting therapeutic red light; The control module, connected to the sensing module and the execution module, is used to construct a three-dimensional digital model and a safe power mapping map based on the multidimensional biological characteristics and geometric morphology data, and to dynamically adjust the output parameters of each LED unit in the execution module by adopting a time-division multiplexing sampling mechanism and a hierarchical clustering driving architecture.

[0009] Preferably, the time-division multiplexing sampling mechanism specifically comprises: The control module drives the LED array in pulse width modulation mode and uses the moment when the LED is in the off interval as the sampling window to synchronously trigger the multispectral imaging component to perform exposure sampling in order to capture the intrinsic reflectance spectrum data of the skin after eliminating the interference of treatment light.

[0010] Preferably, the control module incorporates a near-field array light field superposition model for: Based on the surface point cloud data obtained by the 3D depth vision component and the light field distribution function of each LED unit, the radiation illuminance distribution of the treatment area is calculated, the power adjustment amount is generated, and the illuminance unevenness caused by the curvature fluctuation of the human body surface is compensated to form a uniform initial radiation illuminance field.

[0011] Preferably, the control module further includes security threshold processing logic, used for: The data collected by the multispectral imaging component in the sampling window is analyzed, the endogenous melanin index and exogenous pigment distribution information of the irradiated area are extracted, and the melanin index and exogenous pigment distribution information are converted into the maximum allowable irradiance of each area to generate a safe power ceiling mapping map that defines the maximum instantaneous light power density.

[0012] Preferably, the safety threshold processing logic sets a stricter safety threshold for the identified exogenous pigment deposition area than for areas with the same melanin index, thereby reducing the safety power ceiling of the area to avoid thermal damage caused by local overheating.

[0013] Preferably, the control module further integrates a dose reciprocity law compensation algorithm, used for: Based on the total energy density set for the treatment target and the attenuation characteristics of skin tissue, the total energy required for the body surface is calculated and compared with the energy accumulated within the preset standard treatment time under the safe power ceiling. If the total energy required for the body surface is large, the irradiation time is automatically extended without exceeding the total treatment time of a single session.

[0014] Preferably, for tattoo areas with high light absorption, the control module activates a high-frequency narrow pulse width drive mode for: Based on the difference in thermal relaxation time between pigment particles and surrounding tissue, the driving current is adjusted into a series of pulses with a pulse width close to the thermal relaxation time of the pigment particles and a turn-off interval longer than the pulse width, so as to promote the diffusion of heat generated by the pigment particles to the surrounding matrix tissue and inhibit heat accumulation.

[0015] Preferably, for heat-sensitive skin lesion areas, the control module employs a spatial discretization driving strategy for: A virtual grid is constructed within the skin projection corresponding to this area. Specific LED units are activated based on the grid nodes, while the remaining unselected LED units remain off, forming tiny light spots separated from each other and reserved channels for heat diffusion.

[0016] Preferably, the hierarchical clustering adaptive driving architecture specifically comprises: The control module dynamically divides the granularity of the logical control clusters by calculating the feature gradient between adjacent pixels. In the feature gradient flat region, adjacent pixels are merged to form a large-scale control cluster and assigned the same driving parameters. In the feature gradient transition region, the cluster granularity is shrunk to the smallest unit, and a multi-level array is used to realize the distribution and control of the driving signal.

[0017] The present invention also provides a red light therapy chamber, wherein the red light therapy chamber is equipped with a red light therapy chamber power control and adjustment system.

[0018] The technical effects and advantages of this invention are as follows: 1. This invention, through a sensing array integrating multispectral imaging, three-dimensional depth vision, and infrared thermal imaging modules, can acquire in real time the skin pigment distribution, exogenous pigment deposition, and three-dimensional curvature data of the human body surface in the treatment area. Based on this information, it dynamically adjusts the output power of the high-density micro LED light source array, effectively overcoming the problem of uneven energy absorption and distribution caused by individual skin color differences, pigment deposition such as tattoos, and fluctuations in body surface curvature. By calibrating the uniformity of radiation irradiance on a three-dimensional curved surface and setting a "safe power ceiling" based on clinical safety data for different areas, it fundamentally avoids the risk of local overheating or burns.

[0019] 2. This invention introduces a dose reciprocity law and a thermal relaxation time control strategy to meet the energy requirements of deep target tissues while ensuring epidermal safety through compensated irradiation time. For dark skin or specific lesions, automatic duration compensation and minimum effective power monitoring ensure that light energy effectively reaches deep target tissues and produces a bio-activation effect. For tattoo areas, narrow pulse drive combined with a thermal diffusion gap achieves selective heating and instantaneous cooling of pigment particles. For heat-sensitive melasma areas, spatial discretization lattice drive technology is used to reserve physical channels for lateral heat diffusion, avoiding side effects such as pigment particle bursting or post-inflammatory hyperpigmentation. Attached Figure Description

[0020] Figure 1 This is a structural diagram of the red light therapy chamber of the present invention; Figure 2 This is a system architecture diagram of the present invention; Figure 3 This is a schematic diagram of the dose reciprocity law logic of the present invention; Figure 4 This is a diagram of the hierarchical clustering adaptive driving architecture of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. Example

[0022] In existing red light therapy chambers, the output power is basically a fixed light field, which cannot cope with the huge differences in energy absorption and distribution caused by exogenous pigment deposition such as skin color, tattoos, and the natural curvature of different parts of the body. For dark skin or pigmented areas, using the same light power as the whole may lead to excessive absorption by the epidermis, resulting in heat accumulation or even burn risk. At the same time, since the human body surface is not flat, a fixed power output will cause the actual radiation irradiation received by raised areas to be too high, while the irradiation of recessed areas will be insufficient, resulting in a serious uneven spatial distribution of treatment energy, which poses both safety hazards and affects the efficacy.

[0023] Reference Figures 1 to 3 As shown, in view of this, this embodiment proposes a red light therapy chamber power adjustment system, which is integrated into the red light therapy chamber. Its hardware components include a sensing array consisting of a multispectral imaging module, a three-dimensional depth vision sensor and an infrared thermal imaging module, as well as a high-density micro LED light source array. When the physiotherapy chamber is in use, the system first performs a hardware self-test and collaborative initialization process. The multispectral imaging module calibrates its optical parameters at different wavelengths, including the red light band and the visible and near-infrared bands used for analysis. The 3D depth vision sensor undergoes internal calibration to ensure the accuracy of spatial coordinate measurements. The infrared thermal imaging module performs non-uniformity correction to eliminate pixel response differences. The LED light source array is driven to be illuminated instantaneously at extremely low power. An optical power meter assists in completing the initial light intensity calibration, establishing a correspondence model between the driving current and the output optical power. This ensures that the sensing and execution units operate on a unified benchmark, laying the foundation for subsequent data fusion and control.

[0024] To address the saturation interference of high-intensity therapeutic red light on sensing elements, the system employs a rigorous "dark moment" time-division multiplexing sampling mechanism. Specifically, the system control circuit drives the LED array in high-frequency pulse width modulation (PWM) mode. Within one PWM cycle, when the LEDs are in a microsecond-level off interval, the system synchronously triggers the multispectral imaging module to perform exposure sampling. This time window without active illumination is the "dark moment." The image acquired during this period contains only ambient diffuse reflection light and the skin's own weak luminescence, but its main information is the true reflection / scattering spectrum of the skin surface to ambient light and the deep tissue to therapeutic light after eliminating the direct interference of the therapeutic light. By superimposing and denoising the data acquired from multiple consecutive "dark moments," the system can reconstruct an image of the skin's intrinsic optical characteristics with a high signal-to-noise ratio.

[0025] Due to the complex curvature of the human body, the actual irradiance projected onto the skin surface by the LED array varies significantly depending on the incident angle and distance. To address this, the system uses point cloud data of the human body surface within the treatment area, acquired in real time by a 3D depth vision sensor, to construct a 3D digital model of the human body surface. Simultaneously, the system pre-configures a near-field array light field superposition model. This model calculates the sum of irradiance generated by all LED units at any point on the body surface by superimposing the point cloud data of the human body surface acquired by the 3D depth sensor with the light field distribution function of each LED unit, thereby predicting the irradiance distribution of the entire treatment area. The model combines pre-stored LED light field parameters with real-time 3D scanning data to calculate the power adjustment amount of each LED unit required to compensate for the uneven irradiance caused by surface undulations. Based on this calculation result, the output of the LED array is dynamically controlled to calibrate the uniformity of irradiance on the 3D curved surface, thus forming a uniform initial irradiance field that conforms to the preset benchmark value on the complex 3D body surface.

[0026] After obtaining a uniform initial light field, the system begins to analyze the multispectral data collected during the "dark moment". Using a pre-established algorithm model, the system calculates the melanin index of each pixel or small region. Based on the melanin index, the system further identifies the distribution, type and density of exogenous pigments.

[0027] It is important to note that exogenous pigments such as tattoo ink have specific absorption spectral characteristics that distinguish them from endogenous pigments in the skin. The system analyzes the differences in absorption and reflection characteristics in the visible to near-infrared bands by comparing measured spectra with a built-in pigment spectral library, thereby identifying different types of tattoo inks and estimating their relative density distribution on the skin surface based on the intensity of their reflectance spectra. For endogenous pigmentation areas such as melasma, although their spectral characteristics differ from normal skin, the system can distinguish them from exogenous pigments such as tattoo inks through more refined spectral analysis.

[0028] The system incorporates a safety threshold model based on clinical data. This model converts melanin index and exogenous pigment information into maximum permissible epidermal irradiance. A higher melanin index indicates greater light absorption by the skin, leading to a greater risk of heat accumulation. Therefore, the relative safety power ceiling is set lower. For identified tattoo areas or other areas with high light absorption of exogenous pigment deposits, the absorption coefficient of pigment particles may be higher than that of skin melanin, resulting in a shorter thermal relaxation time. Therefore, the system will employ a stricter safety threshold than for areas with the same melanin index, setting a relatively lower safety power ceiling for these areas to prevent localized overheating and thermal damage.

[0029] Ultimately, the system generates a "safe power ceiling" map for the entire treatment area; this map defines the maximum instantaneous photoelectric power density that each tiny area on the skin surface can withstand, in order to fundamentally prevent epidermal thermal damage caused by excessive radiation.

[0030] During use, after initialization, the system first sends synchronization commands to the LED driver circuit and the sensing system, putting the multispectral imaging module and the 3D depth vision sensor into a ready state. The LED driver circuit operates in a high-frequency pulse width modulation mode, and its specific microsecond-level turn-off interval is precisely defined as the "moment of darkness." Within this interval, the control core synchronously triggers the multispectral imaging module to perform exposure sampling to capture the intrinsic reflectance spectrum data of the skin, which is completely unaffected by the treatment light. Simultaneously, the 3D depth vision sensor scans the treatment area to acquire high-precision 3D point cloud data of the human body surface.

[0031] Based on the real-time acquired 3D point cloud data, the system's built-in near-field array light field superposition model is activated. This model combines the spatial position and radiation characteristics of each LED unit to calculate the theoretical irradiance distribution at each point on the curved surface of the human body. The control core compares the theoretical irradiance with the preset uniform target irradiance and generates a real-time power compensation coefficient for each LED unit. By adjusting its driving current, it compensates for the irradiance unevenness caused by differences in distance and incident angle, thereby constructing an initial uniform radiation field on the 3D body surface.

[0032] While achieving a uniform distribution of the basic light field, the system analyzes the multispectral data acquired during the "moment of darkness" to obtain the optical properties of the skin. The image data captured by the multispectral imaging module within the microsecond-level off interval of the LED is transmitted to the system's signal processing unit. This unit first preprocesses the raw spectral data, including background noise filtering, dark current correction, and compensation for non-uniformity caused by differences in the response of each pixel of the sensor, in order to improve the signal-to-noise ratio.

[0033] Subsequently, based on the pre-established algorithm model, the system calculates the melanin index (MI) of each pixel or small region from the processed spectral data and draws a "melanin index distribution map" of the entire treatment area.

[0034] The melanin index can be calculated based on the skin's reflectivity to specific wavelengths of light. For example, the skin's reflectivity can be measured in red light (e.g., 660nm) and infrared light (e.g., 880nm) bands, and then substituted into the melanin index formula MI=500 / log(5)*(log(infrared light reflectance / red light reflectance)+log(5)). The MI value ranges from 0 to 999. The higher the value, the higher the melanin content of the skin. The formula for calculating the melanin index is existing technology and will not be elaborated further here.

[0035] The distribution map is matched with a safety threshold model that has been experimentally calibrated in advance, and a maximum allowable irradiance value, i.e., the "safe power ceiling", is dynamically set for each pixel or small area. For exogenous pigment deposition areas determined by the image recognition algorithm, the system will call a more conservative safety threshold. This ceiling value serves as a hard upper limit for red light output power.

[0036] When controlling the output of the LED array, the system strictly follows the power compensation coefficient based on three-dimensional calibration and the "safe power ceiling" mapping based on pigment distribution to ensure that the output light power density of each LED unit is always lower than the safe threshold of its corresponding region.

[0037] Meanwhile, the system's infrared thermal imaging branch continuously monitors the surface temperature distribution and compares the real-time temperature data with a preset safe temperature threshold. Once any abnormal temperature rise trend or exceeding the safe threshold is detected in any area, the system will immediately trigger protection logic, dynamically reducing the driving power of the corresponding LED cluster or temporarily shutting down its output, thus forming a basic thermal safety closed loop based on real-time feedback. Example

[0038] While the "safe power ceiling" in Example 1 prevents epidermal burns, for areas with darker skin and thicker subcutaneous tissue, simply following the "safe power ceiling" in Example 1 may result in the energy density reaching the deep target tissue falling below the bioactivation threshold, rendering the treatment ineffective. Conversely, blindly increasing the power in pursuit of deeper therapeutic effects will directly exceed the epidermal safety limits. Furthermore, for areas with special photothermal response characteristics, such as tattoos and melasma, conventional power or time adjustment strategies are ineffective or even dangerous. Tattoo ink particles absorb a large amount of light energy instantly, generating intense heat and posing a risk of explosion; while for heat-sensitive lesions such as melasma, the cumulative heat from continuous light field superposition can easily trigger post-inflammatory hyperpigmentation.

[0039] Reference Figures 1 to 3 As shown, in view of this, this embodiment proposes a red light therapy chamber power adjustment system, and introduces the dose reciprocity law in photobiological regulation, and adds key constraints to it.

[0040] The dose reciprocity law is expressed as:

[0041] Where H is the total energy density, E is the irradiance, and t is the irradiation time. During use, the user or system sets a required total energy density for the deep target tissue based on the treatment goals. Then, based on the tissue thickness and optical properties estimated by the three-dimensional model, the system calculates the attenuation coefficient of light energy reaching deeper tissues, thereby deducing the total energy required to reach the body surface. Next, the system compares... With the "safe power ceiling" (i.e. Under the standard treatment duration The energy that can be accumulated inside ( ).

[0042] If the total energy required to reach the body surface is greater than the energy that can be accumulated within the standard treatment duration, that is... If time compensation is applied, specifically, the irradiation time for that area will be automatically extended, provided that the maximum permissible total duration of a single treatment does not exceed the limit. This is to ensure that the energy accumulated under the "safe power ceiling" and over a longer irradiation period in that area is roughly equivalent to the total energy required to reach the body surface. This ensures the energy needs of deep tissues are met.

[0043] Based on the above scheme, this embodiment also sets a minimum effective power density threshold. The minimum effective power density threshold is determined based on the nonlinear characteristics of photobiological modulation effects, when the irradiance is lower than... Even if the treatment time is extended indefinitely, the irradiated area cannot be effectively treated. Therefore, before performing time compensation, the system will make a second judgment. If the available power calculated to avoid the risk of skin burn is less than the minimum effective power density threshold, time compensation will not be performed, and an early warning will be triggered, prompting the operator that the area cannot be effectively treated due to dark skin color or safety restrictions. It is recommended to switch to a special low-power, longer wavelength safety mode designed for dark skin.

[0044] For areas like tattoos containing highly light-absorbing exogenous pigments, the ink particles will generate a rapid temperature rise after absorbing a large amount of light energy in an instant, which may cause them to explode or burn. Simple power limits and time compensation cannot meet the actual needs. To this end, after identifying the tattoo area through multispectral recognition, the system will activate an ultra-high frequency, narrow pulse width driving mode. By utilizing the huge difference in thermal relaxation time between pigment particles and surrounding normal tissue, the driving current is adjusted into a series of extremely narrow pulses. The pulse width is shorter than or close to the thermal relaxation time of the pigment particles. Correspondingly, the turn-off interval between pulses is much larger than the pulse width itself.

[0045] It is important to note that the difference in thermal relaxation time between pigment particles and surrounding normal tissue is that the thermal relaxation time of pigment particles is extremely short (microseconds to milliseconds), while the thermal relaxation time of normal tissue is longer due to its lower thermal diffusivity.

[0046] During the pulse, the system drives the LED light source to output high-intensity light energy, causing the tattoo pigment particles to be selectively and rapidly heated within this short period due to their inherently high absorption cross-section. Because the pulse width is strictly controlled, the heat is primarily confined within the pigment particles and their immediate vicinity during the pulse duration. The subsequent off-period provides a time window for heat diffusion into the surrounding matrix tissue. This interval, longer than the pulse width, ensures that the heat accumulated in the previous pulse cycle has sufficient time to effectively diffuse from the heated pigment particles into the surrounding tissue with a larger heat capacity. This allows the temperature of the pigment particles themselves and their immediate vicinity to drop significantly, preventing heat accumulation before the next pulse.

[0047] It should be noted that in the tattoo area, although the instantaneous power of a single pulse is high, the average power density is still below the safety ceiling due to the extremely low duty cycle.

[0048] For skin lesions like melasma that are highly sensitive to heat and prone to post-inflammatory hyperpigmentation, this embodiment identifies them by analyzing multispectral data. Specifically, the multispectral imaging module captures the intrinsic reflectance spectrum of the skin within the microsecond-level off interval of the LED, covering multiple bands from visible light to near-infrared. By analyzing this spectral data, the melanin index of each pixel is calculated, mainly based on the differences in the absorption and reflection characteristics of the skin at specific wavelengths (such as the 400-450nm band, which is sensitive to melanin). As an endogenous hyperpigmentation, melasma is characterized by dark areas distributed in patches or reticular patterns under orange light (brown area image) at wavelengths of 590-625nm, forming a sharp contrast with the surrounding normal skin.

[0049] The pigment distribution pattern of melasma often appears as brown reticular or diffuse pigmentation under dermoscopy, and may be accompanied by slight telangiectasia. In contrast, exogenous pigments such as those from tattoos are composed of artificial ink particles, and their spectral absorption characteristics are often different from those of the skin's own melanin. By comparing the measured spectra with the built-in spectral library, the system analyzes the shape of the absorption peaks and reflectance curves in each band, which can distinguish between the endogenous pigmentation of melasma and the exogenous pigment deposition of tattoo ink.

[0050] Once the system identifies the melasma area, it suppresses heat accumulation through a spatial discretization-driven strategy. Based on the melasma boundary determined by multispectral data, it constructs a virtual grid within the skin projection area it covers.

[0051] It should be noted that in the core area where the pigment concentration is high and the thermal risk is high, the grid division is more refined, the corresponding discrete illumination lattice density is set relatively low, and the spacing between the dots is larger; while in the pigment edge transition area, the lattice density can be appropriately increased to ensure the continuity of treatment.

[0052] After the virtual grid is defined, the system calculates the specific LED units that need to be activated based on the position of the grid nodes. These LED units are lit up during the treatment cycle under the drive of control signals, while the unselected LED units around them remain off. This causes the light projected onto the surface of the melasma skin to form a series of tiny, separate light spots. The dark areas between the light spots are reserved channels for heat diffusion. When the discrete light spots heat the skin tissue, the generated heat energy can be conducted laterally to these adjacent areas with lower temperatures, thus avoiding the formation of a large area of ​​continuous heat accumulation in the entire melasma lesion area.

[0053] Although photons are discretely input at the epidermal entry point, the skin tissue's strong scattering characteristics for visible and near-infrared light cause multiple scatterings and overlaps after entering the skin tissue. By optimizing the spacing of the photons in the lattice, a relatively uniform and effective light field can be formed at the target depth. This ensures the photon diffusion coverage required for treatment. However, in the epidermis and superficial layers, because the light source itself is discrete, the heat energy is also dispersed across multiple isolated points. This fundamentally avoids regional hotspots caused by the superposition of continuous light fields, strictly controlling the local temperature rise below the critical line that would induce post-inflammatory hyperpigmentation without sacrificing treatment efficacy. Example

[0054] While the above embodiments can achieve personalized driving for each tiny area, if tens of thousands of micro LED units are controlled independently and point-to-point with precision, a massive amount of real-time data processing will be required, resulting in an unbearable burden on the system's data bandwidth, a significant increase in computing latency, and a sharp rise in overall power consumption.

[0055] Reference Figures 1 to 4 As shown, in view of this, this implementation proposes a hierarchical clustering adaptive driven architecture. Specifically, the system acquires biological characteristics and geometric morphological data of the treatment area in real time through a multispectral imaging module and a 3D depth vision sensor. The multispectral data is acquired within the microsecond-level off-time of the LED light source, and after preprocessing, it is used to calculate the melanin index of each pixel and identify the distribution of exogenous pigments. The 3D depth data is used to construct a spatial coordinate model of the human body surface through point cloud modeling. These data are aligned through a coordinate transformation module to ensure that the optical characteristics of each skin region accurately correspond to its 3D position, providing an input basis for subsequent dynamic clustering.

[0056] The clustering decision algorithm is based on the calculation of feature gradients between adjacent pixels. The system traverses the registered data in a sliding window manner, calculating in real time the Euclidean distance difference between adjacent pixels in terms of melanin index, as well as the rate of change of local surface curvature. When the gradient value is lower than a set threshold, the region is considered to have uniform features; if the gradient value exceeds the threshold, it is marked as a feature abrupt change boundary. For example, in flat areas such as the back, the pigment distribution and curvature change are gradual, and the gradient values ​​are generally low; while on the nose or the edge of a tattoo, the local gradients of pigment concentration and surface normal vectors will significantly increase.

[0057] Based on gradient analysis results, the system dynamically divides the granularity of logic control clusters. For regions with persistently low gradient values, the clustering algorithm automatically merges physically adjacent and similar-featured pixels to form a large-scale control cluster covering dozens to hundreds of LED units. The LEDs within these clusters are assigned the same driving parameters (such as reference power and PWM duty cycle). Through batch control with a single instruction, the number of units that need to be addressed independently is reduced to less than 1 / 10 of the traditional point-to-point mode. In feature gradient transition regions (such as pigmentation edges), the cluster granularity shrinks to the smallest unit. When the difference in melanin index between adjacent pixels exceeds a safety threshold, the system automatically switches to pixel-level control mode to ensure that each LED unit is independently controlled.

[0058] This embodiment achieves adaptive adjustment of cluster size and hardware acceleration through a multi-level FPGA array. Specifically, the first-level FPGA handles the processing of the lowest-level sensor data stream, performing preliminary signal processing on the raw data received from multispectral and 3D sensors, including noise reduction and coordinate transformation. Simultaneously, it performs preliminary and rapid control cluster partitioning calculations. The second-level FPGA receives the results processed by the first-level FPGA, namely the pre-partitioned control clusters and their characteristic data. It then runs a complex algorithm to calculate personalized driving parameters for each cluster, including dose reciprocity compensation and special mode selection as described in Embodiment 2. The third-level FPGA converts the driving parameters calculated by the second-level FPGA into specific, time-sequential driving signals, which are then distributed to the LED driving circuits corresponding to each control cluster. These driving parameters include, for example, PWM waveforms and current values. The three FPGAs communicate with each other via a high-speed data bus.

[0059] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A power control and adjustment system for a red light therapy chamber, characterized in that, include: The sensing module, integrated within the red light therapy chamber, is used to acquire multidimensional biological characteristics and geometric morphological data of the treatment area. The sensing module includes a multispectral imaging component, a three-dimensional depth vision component, and an infrared thermal imaging component. The execution module includes a high-density array of miniature LED light sources for projecting therapeutic red light; The control module, connected to the sensing module and the execution module, is used to construct a three-dimensional digital model and a safe power mapping map based on the multidimensional biological characteristics and geometric morphology data, and to dynamically adjust the output parameters of each LED unit in the execution module by adopting a time-division multiplexing sampling mechanism and a hierarchical clustering driving architecture.

2. The red light therapy chamber power control and adjustment system according to claim 1, characterized in that, The time-division multiplexing sampling mechanism is specifically as follows: The control module drives the LED array in pulse width modulation mode and uses the moment when the LED is in the off interval as the sampling window to synchronously trigger the multispectral imaging component to perform exposure sampling in order to capture the intrinsic reflectance spectrum data of the skin after eliminating the interference of treatment light.

3. The red light therapy chamber power control and adjustment system according to claim 1, characterized in that, The control module incorporates a near-field array light field superposition model, used for: Based on the surface point cloud data obtained by the 3D depth vision component and the light field distribution function of each LED unit, the radiation illuminance distribution of the treatment area is calculated, the power adjustment amount is generated, and the illuminance unevenness caused by the curvature fluctuation of the human body surface is compensated to form a uniform initial radiation illuminance field.

4. The red light therapy chamber power control and adjustment system according to claim 1, characterized in that, The control module also includes security threshold processing logic, used for: The data collected by the multispectral imaging component in the sampling window is analyzed, the endogenous melanin index and exogenous pigment distribution information of the irradiated area are extracted, and the melanin index and exogenous pigment distribution information are converted into the maximum allowable irradiance of each area to generate a safe power ceiling mapping map that defines the maximum instantaneous light power density.

5. The red light therapy chamber power control and adjustment system according to claim 4, characterized in that, The safety threshold processing logic sets a stricter safety threshold for identified exogenous pigment deposition areas than for areas with the same melanin index, thereby lowering the safety power ceiling for that area to avoid thermal damage caused by local overheating.

6. The red light therapy chamber power control and adjustment system according to claim 4, characterized in that, The control module also integrates a dose reciprocity law compensation algorithm, used for: Based on the total energy density set for the treatment target and the attenuation characteristics of skin tissue, the total energy required for the body surface is calculated and compared with the energy accumulated within the preset standard treatment time under the safe power ceiling. If the total energy required for the body surface is large, the irradiation time is automatically extended without exceeding the total treatment time of a single session.

7. The red light therapy chamber power control and adjustment system according to claim 5, characterized in that, For tattoo areas with high light absorption, the control module activates a high-frequency, narrow-pulse-width drive mode for: Based on the difference in thermal relaxation time between pigment particles and surrounding tissue, the driving current is adjusted into a series of pulses with a pulse width close to the thermal relaxation time of the pigment particles and a turn-off interval longer than the pulse width, so as to promote the diffusion of heat generated by the pigment particles to the surrounding matrix tissue and inhibit heat accumulation.

8. The red light therapy chamber power control and adjustment system according to claim 1, characterized in that, For heat-sensitive skin lesions, the control module employs a spatial discretization driving strategy for: A virtual grid is constructed within the skin projection corresponding to this area. Specific LED units are activated based on the grid nodes, while the remaining unselected LED units remain off, forming tiny light spots separated from each other and reserved channels for heat diffusion.

9. The red light therapy chamber power control and adjustment system according to claim 1, characterized in that, The hierarchical clustering adaptive driven architecture is specifically as follows: The control module dynamically divides the granularity of the logical control clusters by calculating the feature gradient between adjacent pixels. In the feature gradient flat region, adjacent pixels are merged to form a large-scale control cluster and assigned the same driving parameters. In the feature gradient transition region, the cluster granularity is shrunk to the smallest unit, and a multi-level array is used to realize the distribution and control of the driving signal.

10. A red light therapy chamber, characterized in that, The red light therapy chamber is equipped with a red light therapy chamber power control and adjustment system as described in any one of claims 1-9.

Images