Intelligent laser acne treatment system

Abstract

The invention provides an intelligent laser acne treatment system, which comprises a laser module 10, a main control module 40, a laser output device 30, a treatment cooling device 20, an information feedback unit 50 and a temperature control unit 60, and is characterized in that the main control module 40 controls the laser module 10 to generate personalized treatment light beams based on an acne treatment neural network model; personalized treatment light beams generated by the laser module 10 are output through a laser window of the laser output device 30; the information feedback unit 50 collects the real-time temperature of the treatment cooling device 20 in real time, and the main control module 40 generates a cooling signal based on an intelligent temperature adjusting model according to the real-time temperature and sends the cooling signal to the temperature control unit 60. The acne treatment neural network model and the intelligent temperature regulation model are creatively constructed, the treatment effect is better, and the patient can be prevented from being burnt.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence and intelligent medical technology, and in particular to an intelligent laser acne treatment system. BACKGROUND

[0002] Acne is a skin disease that mainly occurs in adolescents. According to research reports, up to 90% of adolescents and nearly 50% of adults suffer from acne. Factors such as genetics, living environment, hormone levels, daily diet, and skin care can increase the risk of acne. The occurrence of acne significantly affects the appearance, psychology, social activities, and quality of life of patients. Traditional treatment methods mainly include topical administration and antibiotic therapy. However, drug treatment can cause significant side effects such as skin irritation and teratogenicity. Acne is a chronic disease, and early intervention treatment can effectively reduce the risk of scarring and pigmentation. Laser as a new treatment method has begun to make a name for itself in the treatment of acne. Laser therapy does not require any drugs and can well solve the side effects of traditional drugs. With the widespread application of laser in the treatment of skin diseases, a new treatment method for acne has been provided. The earliest treatment for acne is ultraviolet light, but exposure to ultraviolet light can increase skin aging and have the risk of causing skin deterioration. In addition, short-wave near-infrared light in the 1320-1450nm band is also used to treat acne. This band of light has a significant effect on acne treatment, and the number of acne can be reduced by 80% to 90% after multiple treatments. However, short-wave near-infrared light can cause pain during treatment, causing persistent erythema and scabbing due to skin damage. During treatment, patients usually need to be injected with anesthetics. Although short-wave near-infrared light can achieve certain therapeutic effects, it cannot target the sebaceous glands. The absorption of the sebaceous glands at 1726nm is stronger than that of water (ratio of 1.8:1), and early studies have shown that 1726nm laser treatment of the sebaceous glands can last for two years.

[0003] Current treatment equipment cannot accurately output laser that meets the treatment requirements, and the temperature cooling performance of the treatment equipment in the prior art is low, which makes it difficult to provide a constant treatment temperature and easily burns the patient's skin. Therefore, how to accurately generate laser that meets the requirements and accurately control the temperature is a technical problem. SUMMARY

[0004] The present application is directed to one or more technical defects in the prior art as described above, and the following technical solutions are proposed.

[0005] The intelligent laser acne treatment system comprises a laser module, a main control module, a laser output device, a treatment cooling device, an information feedback unit and a temperature control unit, the main control module controls the laser module to generate a personalized treatment light beam according to an acne treatment neural network model running on the main control module;

[0006] The laser module is connected with the laser output device, and the personalized treatment light beam generated by the laser module is output through the laser window of the laser output device.

[0007] The information feedback unit is connected with the main control module, the information feedback unit collects real-time temperature of the treatment cooling device in real time, and sends the real-time temperature to the main control module, the main control module sets and adjusts the parameters such as power, pulse width, repetition frequency, working mode and running time of the laser module output laser, and controls the output light beam shape and size of the laser output device, the main control module generates a cooling signal based on an intelligent temperature adjustment model according to the real-time temperature and sends it to the temperature control unit, and the temperature control unit controls the cooling temperature of the treatment cooling device based on the cooling signal.

[0008] Further, the acne treatment neural network model is a convolutional neural network model, and the intelligent temperature adjustment model is an LSTM network model.

[0009] Further, the construction process of the acne treatment neural network model is as follows:

[0010] The training sample set for the acne treatment neural network model was obtained through simulation. The construction process of the training sample set included: constructing a human body with acne type, acne grade, skin color, gender, and age; using a geometric model of the human skin tissue structure containing sebaceous glands as the simulation domain; simplifying the skin model to a cube, and then representing the sebaceous gland model as a sphere located inside the skin; simulating the treatment beam incident on the target treatment area through a laser window; describing the light transmission and distribution within the skin tissue based on the diffuse approximation equation; and comparing the light distribution at different depths, different pulse widths, and different laser energies using the Pennes biological heat transfer equation; and simulating the photo-induced thermal effect using the Pennes biological heat transfer equation. Based on the cooling effect of the treatment cooling device, boundary conditions and relevant thermal parameters are set; light energy density, pulse width, and cooling temperature are input; the Arrhenius kinetic model is used to simulate tissue thermal damage caused by high temperature and to evaluate sebaceous gland damage; it is determined whether the epidermal temperature is below 50°C and whether the sebaceous gland has reached the thermal damage threshold. If "yes", the light energy density, pulse width, cooling temperature, treatment target area, acne type, acne grade, skin color, gender, and age are saved as a sample. If the answer is "no", the light energy density, pulse width, and cooling temperature are re-entered; after repeating N times, a training sample set including N training samples is obtained, where N≥2, and the treatment target area includes the treatment location;

[0011] Each training sample in the training sample set, which includes N training samples, is used to treat patients with the same treatment target area, acne type, acne grade, skin color, gender, and age as those training samples. The light energy density, pulse width, and cooling temperature in the training sample are calibrated based on the actual treatment effect. After all training samples have been calibrated, a calibrated training sample set is generated.

[0012] The acne treatment neural network model is trained using the calibrated training sample set to obtain the trained acne treatment neural network model.

[0013] Furthermore, the acne treatment neural network model undergoes self-updating during use, including: adjusting the weights of the acne treatment neural network model based on the treatment effects of M actual patients, and inputting the treatment target area, acne type, acne grade, skin color, gender, and age of the M patients into the acne treatment neural network model to output the corresponding light energy density (LE). i Pulse width FW i and cooling temperature CT i Based on the light energy density (FLE) used in actual treatment. i Pulse width FFW i and cooling temperature FCT ia light energy density LE corresponding to an output of the acne treatment neural network model i a pulse width FW i a cooling temperature CT i updating an output of the acne treatment neural network model;

[0014] a light energy density ULE of the updated output of the acne treatment neural network model is:

[0015]

[0016] a pulse width UFW of the updated output of the acne treatment neural network model is:

[0017]

[0018] a cooling temperature UCT of the updated output of the acne treatment neural network model is:

[0019]

[0020] wherein LE, FW and CT are light energy density, pulse width and cooling temperature output by the acne treatment neural network model inputted with treatment target area, acne type, acne grade, skin color, gender and age of a patient.

[0021] Further, the treatment cooling device comprises a sapphire light guide crystal, a cold guide block, a TEC refrigeration sheet, a heat sink, a heat dissipation fan and a temperature measuring element; the temperature measuring element transmits a measured temperature signal to an information feedback unit, and the cooling temperature is adjusted by a temperature control unit.

[0022] The laser module comprises a plurality of single-tube lasers, a fiber coupling device, an indicating light source, a heat dissipation module and a laser driving power supply, and the plurality of single-tube lasers are outputted through fiber coupling; the laser module further comprises a red light indicating light source for displaying the spot shape, size and position of the treatment light beam; the fiber coupling device couples a plurality of laser beams into a transmission optical fiber, couples the treatment light source into a laser output device through the transmission optical fiber, and outputs the treatment light beam through the laser output device.

[0023] Further, the fiber coupling device comprises an optical focusing lens group and a fiber coupling device, the optical focusing lens group focuses the laser and couples it with the indicating light through the fiber coupling device, and the fiber coupling device is connected with a fiber connector.

[0024] Further, the laser output device integrates a beam shaping lens group, a scanning galvanometer and a beam shrinking assembly, shapes, focuses and forms a spot array of the treatment light source in the transmission optical fiber, and outputs through a laser window.

[0025] Furthermore, a contact pressure sensor is embedded in the laser window to sense whether the laser output window is in contact with the skin and transmits the pressure information to the main control module to directly control whether the laser is output.

[0026] Furthermore, the output wavelength range of the single-tube laser is 1710-1730nm, the average power of the single-tube laser is 0-100W, and the pulse width is modulated to 0-60ms.

[0027] Furthermore, the system also includes a touch screen for human-computer interaction.

[0028] The technical advantages of this invention are as follows: This invention provides an intelligent laser acne treatment system, comprising a laser module 10, a main control module 40, a laser output device 30, a treatment cooling device 20, an information feedback unit 50, and a temperature control unit 60. The main control module 40 controls the laser module 10 to generate a personalized treatment beam based on an acne treatment neural network model running on the main control module 40. The laser module 10 is connected to the laser output device 30, and the personalized treatment beam generated by the laser module 10 is output through the laser window of the laser output device 30. The information feedback unit 50 is connected to the main control module 40. The control module 40 is connected to the information feedback unit 50, which collects the real-time temperature of the treatment cooling device 20 and sends the real-time temperature to the main control module 40. The main control module 40 sets and adjusts parameters such as the power, pulse width, repetition frequency, working mode, and running time of the laser output of the laser module 10, and controls the shape and size of the output beam of the laser output device 30. The main control module 40 generates a cooling signal based on the real-time temperature and an intelligent temperature regulation model, and sends it to the temperature control unit 60. The temperature control unit 60 controls the cooling temperature of the treatment cooling device 20 based on the cooling signal.This invention creatively constructs an acne treatment neural network model and an intelligent temperature regulation model. The main control module 40 controls the laser module 10 to generate a personalized treatment beam based on the acne treatment neural network model running on the main control module 40. The main control module 40 sets and adjusts parameters such as the power, pulse width, repetition frequency, operating mode, and running time of the laser output from the laser module 10, and controls the shape and size of the output beam from the laser output device 30. Based on the real-time temperature and the intelligent temperature regulation model, the main control module 40 generates a cooling signal and sends it to the temperature control unit 60. The temperature control unit 60 controls the cooling temperature of the treatment cooling device 20 based on the cooling signal, thereby generating laser light that meets treatment requirements and preventing burns to the patient. A key inventive concept of this invention is the generation process of the training sample set for training the acne treatment neural network model. This involves first generating multiple training samples through simulation, then using these training samples to treat actual patients, and calibrating the parameters in the training samples during the treatment process. This makes the training samples more accurate and avoids the problem of setting parameters from 0 during actual patient treatment, improving the accuracy and generation speed of the training samples, and making the generated neural network model more in line with the needs of patients. In order to avoid the degradation of the acne treatment neural network model during use, it needs to be updated. This invention updates the model based on the parameters of the historical laser beam and the actual treatment parameters. This update method has a small amount of computation, avoids retraining the neural network model, saves training time, and improves the output accuracy of the model. As shown in the schematic diagram of the connection between the single tube chip 120 and the fast axis collimating lens 1211 and the slow axis collimating lens 1212, the output laser is collimated. The collimated light is then reflected by the reflector 1213 into the cemented polarization combiner 1214 for beam combining. The cemented polarization combiner 1214 is formed by cementing a half-wave plate 1214a and a polarization coupling prism 1214b. The polarization state of the first beam rotates by 90° after passing through the half-wave plate 1214a. The two beams are then polarized and coupled by the polarization coupling prism 1214b. This makes the generated laser beam more stable and has higher beam precision. In this invention, the laser module 10 is connected to the laser output device 30 via a transmission fiber 19, the fiber core of which is 400 μm and the numerical aperture is 0.22 NA. When the treatment light source is transmitted to the beam shaping mirror group 31 for beam expansion and collimation, the scanning galvanometer 32 swings at a specific angle in the X and Y axes to form a beam array after expansion and collimation. Finally, the beam is focused and collimated by the beam shrinking assembly 33 and output through the laser window. This generates a beam array that meets the treatment requirements, resulting in better treatment effects. Attached Figure Description

[0029] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings.

[0030] Figure 1 This is a structural block diagram of an intelligent laser acne treatment system according to the present invention;

[0031] Figure 2 This is a structural block diagram of the laser module described in this invention;

[0032] Figure 3 This is a schematic diagram of the internal structure of the multi-single-tube laser of the present invention;

[0033] Figure 4 This is a schematic diagram of the internal structure of the optical fiber combining device of the present invention;

[0034] Figure 5 This is a schematic diagram of the laser output device and treatment cooling device of the present invention.

[0035] Wherein: the labels in the attached figures

[0036] 10-Laser module; 11-Laser driver power supply; 12-Multi-single-tube laser; 120-Single-tube laser chip; 121 Optical lens group; 1211-Fast axis collimating lens; 1212-Slow axis collimating lens; 1213-Reflecting mirror; 1214-Cemented polarization combiner; 1214a-Half-wave plate; 1214b-Polarization coupling prism; 13-Fiber optic bundle combiner; 130-Optical focusing lens group; 131-Fiber optic coupling device; 132-Fiber optic connector; 133-Replaceable protective window; 14-Indicator light source; 15-Heat dissipation module; 16-Temperature sensor; 17-Optical power sensor; 18-Fiber optic in-situ probe; 19-Transmission fiber;

[0037] 20-Therapeutic cooling device; 21-Sapphire light guide crystal; 22-Cooling plate; 23-TEC cooling plate; 24-Heat sink; 25-Cooling fan; 26-Temperature sensing element;

[0038] 30-Laser output device; 31-Beam shaping mirror assembly; 32-Scanning galvanometer; 33-Beam constriction assembly; 34-Contact pressure sensing element;

[0039] 40 - Main control module;

[0040] 50 - Information Feedback Unit;

[0041] 60 - Temperature control unit;

[0042] 70-Substrate; Detailed Implementation

[0043] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0044] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0045] Figure 1 The present invention illustrates an intelligent laser acne treatment system, which includes a laser module 10, a main control module 40, a laser output device 30, a treatment cooling device 20, an information feedback unit 50, and a temperature control unit 60. The main control module 40 controls the laser module 10 to generate personalized treatment beams according to an acne treatment neural network model running on the main control module 40.

[0046] The laser module 10 is connected to the laser output device 30, and the personalized treatment beam generated by the laser module 10 is output through the laser window of the laser output device 30.

[0047] The information feedback unit 50 is connected to the main control module 40. The information feedback unit 50 collects the real-time temperature of the treatment cooling device 20 and sends the real-time temperature to the main control module 40. The main control module 40 sets and adjusts parameters such as the power, pulse width, repetition frequency, working mode, and running time of the laser output of the laser module 10, and controls the shape and size of the output beam of the laser output device 30. The main control module 40 generates a cooling signal based on the real-time temperature and an intelligent temperature regulation model and sends it to the temperature control unit 60. The temperature control unit 60 controls the cooling temperature of the treatment cooling device 20 based on the cooling signal.

[0048] This invention creatively constructs an acne treatment neural network model and an intelligent temperature regulation model. The main control module 40 controls the laser module 10 to generate a personalized treatment beam based on the acne treatment neural network model running on the main control module 40. The main control module 40 sets and adjusts parameters such as the power, pulse width, repetition frequency, working mode, and running time of the laser output by the laser module 10, and controls the shape and size of the output beam of the laser output device 30. The main control module 40 generates a cooling signal based on the real-time temperature and the intelligent temperature regulation model, and sends it to the temperature control unit 60. The temperature control unit 60 controls the cooling temperature of the treatment cooling device 20 based on the cooling signal, thereby realizing the generation of laser that meets the treatment requirements and preventing the patient from being burned. This is one of the important inventive concepts of this invention.

[0049] In one embodiment, the acne treatment neural network model is a convolutional neural network model, and the intelligent temperature regulation model is an LSTM network model. Since temperature regulation is related to historical temperatures, the intelligent temperature regulation model uses an LSTM network model to make temperature adjustment more accurate. The treatment laser beam is generated based on the patient's relevant parameters, so a convolutional neural network model is used to make the generated treatment laser beam more suitable for the patient's needs. Both models need to be trained before use, which is another important inventive concept of this invention.

[0050] In one embodiment, the construction process of the acne treatment neural network model is as follows:

[0051] The training sample set for the acne treatment neural network model was obtained through simulation. The construction process of the training sample set included: constructing a human body with acne type, acne grade, skin color, gender, and age; using a geometric model of the human skin tissue structure containing sebaceous glands as the simulation domain; simplifying the skin model to a cube, and then representing the sebaceous gland model as a sphere located inside the skin; simulating the treatment beam incident on the target treatment area through a laser window; describing the light transmission and distribution within the skin tissue according to the diffuse approximation equation; and comparing the light distribution at different depths, different pulse widths, and different laser energies using the Pennes biological heat transfer equation; and simulating the photo-induced thermal effect using the Pennes biological heat transfer equation. Based on the cooling effect of the treatment cooling device 20, boundary conditions and relevant thermal parameters are set; light energy density, pulse width, and cooling temperature are input; the Arrhenius kinetic model is used to simulate tissue thermal damage caused by high temperature and to evaluate sebaceous gland damage; it is determined whether the epidermal temperature is below 50°C and whether the sebaceous gland has reached the thermal damage threshold. If "yes", the light energy density, pulse width, cooling temperature, treatment target area, acne type, acne grade, skin color, gender, and age are saved as a sample. If the answer is "no", the light energy density, pulse width, and cooling temperature are re-inputted. After repeating N times, a training sample set including N training samples is obtained, where N≥2, and the treatment target area includes the treatment location.

[0052] Each training sample in the training sample set, which includes N training samples, is used to treat patients with the same treatment target area, acne type, acne grade, skin color, gender, and age as those training samples. The light energy density, pulse width, and cooling temperature in the training sample are calibrated based on the actual treatment effect. After all training samples have been calibrated, a calibrated training sample set is generated.

[0053] The acne treatment neural network model is trained using the calibrated training sample set to obtain the trained acne treatment neural network model.

[0054] A key inventive concept of this invention is the process of generating the training sample set for training the acne treatment neural network model. This involves first generating multiple training samples through simulation, then using these training samples to treat actual patients. During the treatment process, the parameters in the training samples are calibrated to make the training samples more accurate. This avoids the problem of having to set parameters from 0 during actual patient treatment, improving the accuracy and generation speed of the training samples. This results in a neural network model that better meets the needs of patients, which is another important inventive concept of this invention.

[0055] In one embodiment, the acne treatment neural network model is self-updated during use, including: the acne treatment neural network model adjusts the weights based on the treatment effects of M[1] actual patients, and inputs the treatment target area, acne type, acne grade, skin color, gender and age of the M patients into the corresponding light energy density LE output by the acne treatment neural network model. i Pulse width FW i and cooling temperature CT i Based on the light energy density (FLE) used in actual treatment. i Pulse width FFW i and cooling temperature FCT i The light energy density LE corresponding to the output of the acne treatment neural network model i Pulse width FW i and cooling temperature CT i Update the output of the acne treatment neural network model;

[0056] The updated optical energy density ULE output by the acne treatment neural network model is:

[0057]

[0058] The updated pulse width UFW output by the acne treatment neural network model is:

[0059]

[0060] The updated cooling temperature UCT output by the acne treatment neural network model is:

[0061]

[0062] Among them, LE, FW, and CT are the patient's treatment target area, acne type, acne grade, skin color, gender, and age input to the acne treatment neural network model, and the output light energy density, pulse width, and cooling temperature.

[0063] To prevent the acne treatment neural network model from degrading during use, it needs to be updated. This invention updates the model based on the parameters of the historical laser beam and the actual treatment parameters. This update method has a small computational load, avoids retraining the neural network model, saves training time, and improves the output accuracy of the model. This is another important inventive concept of this invention.

[0064] Figure 2 The structure of the laser module 10 described in this invention is shown. The core of the laser module 10 is a multi-single-tube laser 12. The multiple single-tube laser chips 120 shape, collimate, and combine the output beam through an optical lens group 121 and an optical fiber combining device 13. The laser module 10 also includes a single indicator light source 14, which is output through an optical fiber. The optical fiber coupling device 13 combines the treatment light source and the indicator light, couples them to the central axis of the optical fiber connector 132, and then outputs them through a transmission optical fiber 19.

[0065] The multi-single-tube laser 12 is composed of multiple single-tube chips 120, which are arranged in a stepped manner and welded onto the substrate 70. The power can be superimposed to achieve high-power laser output of the multi-single-tube laser 12, with an adjustable power range of 0-100W.

[0066] The optical lens group 121 includes a fast-axis collimating lens 1211, a slow-axis collimating lens 1212, a reflecting mirror 1213, and a cemented polarizing beam combiner 1214.

[0067] The fiber optic bundle combining device 13 includes an optical focusing lens group 130, a replaceable protective window 133, and a fiber optic connector 132, wherein the fiber optic connector 132 may be of the SMA or FC interface type.

[0068] The laser driving power supply 11 is a chopper power supply that can output continuously or in pulses. Parameters such as current magnitude, pulse width, and repetition frequency can be controlled by input signals from the main control module 40. The adjustable range of parameters is: current 0-12A, pulse width 0-60ms, and repetition frequency 0-10Hz.

[0069] Figure 3A schematic diagram showing the connection between the single-tube chip 120 and the fast-axis collimating lens 1211 and slow-axis collimating lens 1212 is illustrated. The output laser beam is collimated, and the collimated beam is then reflected by the reflector 1213 into the cemented polarization combiner 1214 for beam combining. The number of fast-axis collimating lenses 1211 and slow-axis collimating lenses 1212 is the same as the number of single-tube chips 120, and they are fixed to the front end of the corresponding single-tube chips 120 on the substrate 70. Multiple single-tube chips 120 arranged in a stepped manner are divided into two groups, with the polarization states of the two groups of beams perpendicular to each other. Each group can contain N beams. The cemented polarization combiner 1214 is formed by cementing a half-wave plate 1214a and a polarization coupling prism 1214b. After the first group of beams passes through the half-wave plate 1214a, its polarization state rotates by 90°. The two groups of beams then undergo polarization coupling through the polarization coupling prism 1214b.

[0070] The diagram showing the connection between the single-tube chip 120 and the fast-axis collimating mirror 1211 and the slow-axis collimating mirror 1212 indicates that the output laser beam is collimated. The collimated light is then reflected by the mirror 1213 and fed into the cemented polarization combiner 1214 for beam combining. The cemented polarization combiner 1214 is formed by cementing a half-wave plate 1214a and a polarization coupling prism 1214b. After the first beam passes through the half-wave plate 1214a, its polarization state is rotated by 90°. The two beams then undergo polarization coupling through the polarization coupling prism 1214b. This results in a more stable laser beam with higher beam accuracy, which is another important inventive concept of this invention.

[0071] Figure 4 The fiber optic combining device 13 couples the polarization-coupled beam and the indicator light into the transmission fiber 19 via the optical focusing lens group 130. The indicator light source 14 can be a red laser. As a further improvement, a replaceable protective window 133 is installed between the optical focusing lens group 130 and the fiber optic connector 132. This window can block dust and foreign objects brought into the transmission fiber 19 due to frequent connections, isolate the laser module 10 from the external environment, and ensure laser transmission efficiency. When the surface of the replaceable protective window 133 is dirty, it can be disassembled for cleaning or replacement.

[0072] As a further improvement, an optical fiber in-situ probe 18 and an optical power sensor 17 can be loaded at the rear end of the optical fiber combining device 13. The optical fiber in-situ probe 18 is used to detect whether the transmission optical fiber 19 is poorly connected or detached. The optical power sensor 17 can provide real-time power detection of the therapeutic laser.

[0073] As a further improvement, multiple temperature sensors 16 are provided on the substrate 70 to detect the operating temperature of the multi-single-tube laser 12 and provide over-temperature protection.

[0074] See Figure 5 This is a schematic diagram of the structure of the laser output device 30 and the treatment cooling device 20 described in this invention.

[0075] In this embodiment, the laser output device 30 includes a beam shaping mirror group 31, a scanning galvanometer 32, a beam shrinking assembly 33, and a contact pressure sensing element 34. The laser module 10 is connected to the laser output device 30 via a transmission optical fiber 19, the fiber having a core diameter of 400 μm and a numerical aperture of 0.22 NA. When the treatment light source is transmitted to the beam shaping mirror group 31 for beam expansion and collimation, the scanning galvanometer 32 forms a beam array of light spots by oscillating at specific angles in the X and Y axes. Finally, the beam is focused and collimated by the beam shrinking assembly 33 and output through the laser window. The contact pressure sensor at the front end of the laser window can detect whether the sapphire window area is in complete contact with the skin and feeds back the pressure signal to the main control module. When the pressure is sufficient and uniform, the laser emits normally; when the pressure is insufficient or there is no pressure at one end, the laser stops emitting.

[0076] In this invention, the laser module 10 is connected to the laser output device 30 via a transmission optical fiber 19, the fiber core of which is 400 μm and the numerical aperture is 0.22 NA. When the treatment light source is transmitted to the beam shaping mirror group 31 for beam expansion and collimation, the scanning galvanometer 32 swings at a specific angle in the X and Y axes to form a beam array after expansion and collimation. Finally, the beam is focused and collimated by the beam shrinking assembly 33 and output through the laser window. This generates a beam array that meets the treatment requirements, resulting in better treatment effects. This is another important inventive point of this invention.

[0077] The treatment cooling device 20 consists of a sapphire light guide crystal 21, a cooling plate 22, a TEC cooling plate 23, a heat sink 24, a cooling fan 25, and a temperature sensing element 26. The TEC cooling plate 23 is used to cool the sapphire light guide crystal 21. It has two end faces; one side is the cold side and the other side is the hot side during the cooling process. The heat from the sapphire light guide crystal 21 is transferred to the cold side of the TEC cooling plate 23 via the cooling plate 22, while the hot side is dissipated by the heat sink 24.

[0078] The sapphire light guide crystal 21 can directly contact the skin surface and actively control the cooling temperature between 0-5℃, ensuring that the temperature of the treatment area remains below 50℃, thus avoiding discomfort caused by overheating. This achieves better treatment results and prevents patients from being burned.

[0079] The temperature sensing element 26 is located at the rear end of the sapphire light guide crystal 21. It measures the cooling temperature in real time and transmits the temperature signal to the information feedback system unit 50. When the information feedback unit 50 receives an abnormal signal, it prompts the main control module 40 to guide the adjustment of the cooling level, and the temperature control unit 60 maintains a constant cooling temperature.

[0080] The cooling fan 25 is fixedly connected to one side of the radiator 24, transferring the heat from the radiator 24 to the air.

[0081] Furthermore, the laser output device 30 and the treatment cooling device 20 can be encapsulated in a convenient treatment handpiece.

[0082] The heat dissipation module 15 is a water-cooled plate with embedded copper pipes. The inlet 15a and outlet 15b can be connected to an external water chiller to dissipate heat for the laser module 10.

[0083] In one embodiment, the system also includes a touch screen for human-computer interaction. This allows doctors or patients [j2] to easily adjust the treatment laser parameters; that is, the output of the acne treatment neural network model is also provided for reference by doctors or patients. The laser parameters generated accordingly can be adjusted by doctors or patients to better suit their needs. This is another important aspect of the invention.

[0084] The acne treatment neural network model and intelligent temperature regulation model of the present invention are implemented by computer programs and can be stored in a computer storage medium. When the computer program on the computer storage medium is executed by a processor, the above-described method is implemented. The computer storage medium can be a hard disk, DVD, CD, flash memory, or other memory.

[0085] For ease of description, the above-described apparatus is divided into various functional units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware components.

[0086] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the apparatus described in various embodiments or some parts of the embodiments of this application.

[0087] Finally, it should be noted that the above embodiments are for illustration only and not for limiting the technical solutions of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention without departing from the spirit and scope of the present invention. Any modifications or partial substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An intelligent laser acne treatment system, characterized in that, The system includes a laser module, a main control module, a laser output device, a treatment cooling device, an information feedback unit, and a temperature control unit. The main control module controls the laser module to generate a personalized treatment beam based on an acne treatment neural network model running on the main control module. The laser module is connected to the laser output device, and the personalized treatment beam generated by the laser module is output through the laser window of the laser output device; The information feedback unit is connected to the main control module. The information feedback unit collects the real-time temperature of the treatment cooling device and sends the real-time temperature to the main control module. The main control module sets and adjusts the power, pulse width, repetition frequency, working mode, and running time of the laser output by the laser module, and controls the shape and size of the output beam of the laser output device. The main control module generates a cooling signal based on the real-time temperature and an intelligent temperature regulation model, and sends it to the temperature control unit. The temperature control unit controls the cooling temperature of the treatment cooling device based on the cooling signal.

2. The system according to claim 1, characterized in that, The acne treatment neural network model is a convolutional neural network model, and the intelligent temperature regulation model is an LSTM network model.

3. The system according to claim 2, characterized in that, The construction process of the neural network model for acne treatment is as follows: The training sample set for the acne treatment neural network model was obtained through simulation. The construction process of the training sample set included: constructing a human body with acne type, acne grade, skin color, gender, and age; using a geometric model of the human skin tissue structure containing sebaceous glands as the simulation domain; simplifying the skin model to a cube, and then representing the sebaceous gland model as a sphere located inside the skin; simulating the treatment beam incident on the target treatment area through a laser window; describing the light transmission and distribution within the skin tissue based on the diffuse approximation equation; and comparing the light distribution at different depths, different pulse widths, and different laser energies using the Pennes biological heat transfer equation; and simulating the photo-induced thermal effect using the Pennes biological heat transfer equation. Based on the cooling effect of the treatment cooling device, boundary conditions and relevant thermal parameters are set; light energy density, pulse width, and cooling temperature are input; the Arrhenius kinetic model is used to simulate tissue thermal damage caused by high temperature and to evaluate sebaceous gland damage; it is determined whether the epidermal temperature is below 50°C and whether the sebaceous gland has reached the thermal damage threshold. If "yes", the light energy density, pulse width, cooling temperature, treatment target area, acne type, acne grade, skin color, gender, and age are saved as a sample. If the answer is "no", the light energy density, pulse width, and cooling temperature are re-inputted. After repeating N times, a training sample set including N training samples is obtained, where N≥2, and the treatment target area includes the treatment location. Each training sample in the training sample set, which includes N training samples, is used to treat patients with the same treatment target area, acne type, acne grade, skin color, gender, and age as those training samples. The light energy density, pulse width, and cooling temperature in the training sample are calibrated based on the actual treatment effect. After all training samples have been calibrated, a calibrated training sample set is generated. The acne treatment neural network model is trained using the calibrated training sample set to obtain the trained acne treatment neural network model.

4. The system according to claim 3, characterized in that, The acne treatment neural network model undergoes self-updating during use, including: adjusting the weights of the acne treatment neural network model based on the treatment effects of M actual patients; and inputting the treatment target area, acne type, acne grade, skin color, gender, and age of the M patients into the acne treatment neural network model to output the corresponding light energy density (LE). i Pulse width FW i and cooling temperature CT i Based on the light energy density (FLE) used in actual treatment. i Pulse width FFW i and cooling temperature FCT i The light energy density LE corresponding to the output of the acne treatment neural network model i Pulse width FW i and cooling temperature CT i Update the output of the acne treatment neural network model; The updated optical energy density ULE output by the acne treatment neural network model is: The updated pulse width UFW output by the acne treatment neural network model is: The updated cooling temperature UCT output by the acne treatment neural network model is: Among them, LE, FW, and CT are the patient's treatment target area, acne type, acne grade, skin color, gender, and age input to the acne treatment neural network model, and the output light energy density, pulse width, and cooling temperature.

5. The system according to claim 4, characterized in that, The treatment cooling device includes a sapphire light guide crystal, a cooling block, a TEC cooling chip, a heat sink, a cooling fan, and a temperature measuring element; the temperature measuring element transmits the measured temperature signal to the information feedback unit, and the temperature control unit adjusts the cooling temperature; The laser module includes: multiple single-tube lasers, an optical fiber combiner, an indicator light source, a heat dissipation module, and a laser driver power supply. The multiple single-tube lasers are coupled and output through optical fibers. The laser module also includes a red light indicator light source for displaying the shape, size, and position of the treatment beam. The optical fiber combiner combines multiple laser beams into a transmission optical fiber, couples the treatment light source to the laser output device through the transmission optical fiber, and outputs the treatment beam through the laser output device.

6. The system according to claim 5, characterized in that, The fiber optic bundle combining device includes an optical focusing lens group and a fiber optic coupling device. The optical focusing lens group focuses the laser and couples it with the indicator light through the fiber optic coupling device, which is connected to a fiber optic connector.

7. The system according to claim 6, characterized in that, The laser output device integrates a beam shaping mirror group, a scanning galvanometer, and a beam shrinking component to shape, focus, and form a beam array of therapeutic light sources in the transmission optical fiber, and then output the light through the laser window.

8. The system according to claim 7, characterized in that, A contact pressure sensor is embedded in the laser window to sense whether the laser output window is in contact with the skin and transmit the pressure information to the main control module to directly control whether the laser is output.

9. The system according to claim 8, characterized in that, The output wavelength range of the single-tube laser is 1710-1730nm, the average power of the single-tube laser is 0-100W, and the pulse width is modulated from 0-60ms.

10. The system according to claim 9, characterized in that, The system also includes a touch screen for human-computer interaction.

Images