Ultrasonic beauty knife and use method thereof

Abstract

The invention relates to an ultrasonic beauty knife and a using method thereof, and the ultrasonic beauty knife comprises a control device, a display device, a handle and a probe; the control device is connected with the display device, the control device is connected with the handle, and the probe is detachably arranged on the handle; the probe comprises a shell, an action assembly and an ultrasonic sheet; the action assembly and the ultrasonic sheet are both located in the shell; an opening is formed in the lower end of the shell; the action assembly comprises a transverse translation mechanism and a vertical lifting mechanism arranged on the transverse translation mechanism, and the ultrasonic sheet is arranged on the vertical lifting mechanism; an acoustic transparent component is arranged at the opening of the lower end of the shell.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic beauty technology, and particularly relates to an ultrasonic beauty knife and its usage method. Background Art

[0002] An ultrasonic beauty knife is an instrument that performs physical therapy on the human skin by generating focused ultrasonic waves, which can play roles such as lifting and firming or face slimming. During use, for the skin of different parts and skin conditions, it is usually necessary to adjust the focus of the ultrasonic waves so that it acts on the fascia layer of the human skin to achieve a better improvement effect. However, the existing ultrasonic beauty instruments do not have an adaptive focus depth adjustment function, and can only adjust the fixed height by replacing the probes with different ultrasonic chip heights, which will result in poor adaptability to the skin and an unclear improvement effect. Moreover, the existing ultrasonic beauty instruments can only perform physical therapy and cannot perform detection. Summary of the Invention

[0003] To solve the above problems, the object of the present invention is to provide an ultrasonic beauty knife that first emits a first ultrasonic wave to detect the skin and determine the position of the fascia layer, and then adaptively adjusts the focal depth, so as to perform physical therapy targeted according to the situation of the fascia layer.

[0004] The ultrasonic beauty knife of the present invention includes a control device, a display device, a handle, and a probe; the control device is connected to the display device, the control device is connected to the handle, and the probe is detachably arranged on the handle; the probe includes a housing, an action component, and an ultrasonic chip; both the action component and the ultrasonic chip are located inside the housing; the lower end of the housing is open; the action component includes a horizontal translation mechanism and a vertical lifting mechanism arranged on the horizontal translation mechanism, and the ultrasonic chip is arranged on the vertical lifting mechanism; an acoustic transparent member is arranged at the opening at the lower end of the housing.

[0005] A further improvement of the present invention lies in that the horizontal translation mechanism includes a horizontal motor, a horizontal screw rod, and a horizontal slider; the horizontal motor is arranged inside the housing; the horizontal screw rod is connected to the rotating shaft of the horizontal motor, the horizontal slider is arranged on the horizontal screw rod, and the vertical lifting mechanism is arranged on the horizontal slider.

[0006] A further improvement of the present invention lies in that the vertical lifting mechanism is a telescopic motor, and the ultrasonic chip is arranged at the telescopic end of the telescopic motor.

[0007] A further improvement of the present invention lies in that the ultrasonic chip includes a first ultrasonic chip for detection and a second ultrasonic chip for physical therapy; the first ultrasonic chip is sleeved outside the second ultrasonic chip.

[0008] The present invention also provides a method for using an ultrasonic beauty knife. The method is implemented based on the ultrasonic beauty knife and includes the following steps:

[0009] Step S1: Assemble the probe on the handle and preset the position and number of times through the display device;

[0010] Step S2: The probe contacts the skin, and the first ultrasonic sheet emits the first ultrasonic wave to perform a primary detection on the skin and feed the corresponding ultrasonic image back to the display device; the primary detection is used to identify the skin layers and determine the position of the fascia layer of the skin;

[0011] Step S3: Based on the position of the fascia layer, the control device determines the focal depth between the ultrasonic sheet and the fascia layer;

[0012] Step S4: Based on the focal depth determined in Step S3, the control device adjusts the height of the ultrasonic sheet through the vertical lifting mechanism;

[0013] Step S5: The horizontal translation mechanism drives the ultrasonic sheet to move to the preset position, and the second ultrasonic sheet emits the second ultrasonic wave to perform physical therapy on the skin according to the preset number of times.

[0014] A further improvement of the present invention is that it further includes Step S6; Step S6: The first ultrasonic sheet emits the first ultrasonic wave to perform a secondary detection on the beauty area.

[0015] A further improvement of the present invention is that in Step S2, after the ultrasonic image is fed back to the control device, the control device inputs the ultrasonic image into the improved image segmentation model and performs data post-processing on the output of the improved image segmentation model to obtain the depth of the fascia layer; the skin includes the skin layer, fat layer, fascia layer, and muscle layer.

[0016] The construction steps of the improved image segmentation model are as follows:

[0017] Data preparation step: Collect ultrasonic images of the skin as data images in the data set and label them layer by layer; each layer of the skin corresponds to a category, and each category corresponds to a category number;

[0018] Data preprocessing step: Perform grayscale value normalization, brightness adjustment, and image scaling operations on the data images in the data set;

[0019] Model training step: Input the data images in the data set after data preprocessing into the image segmentation model for training, and obtain the improved image segmentation model after training; the output of the improved image segmentation model is the segmentation probability map of each category.

[0020] A further improvement of the present invention is that the steps of performing data post-processing on the output of the improved image segmentation model are as follows:

[0021] Step A01: Threshold the segmentation probability maps of each category to obtain the binary maps of each category;

[0022] Step A02: Perform morphological operations on the binary maps of each category and scale them to obtain the segmentation results of each category;

[0023] Step A03: Determine the segmentation results of the skin layer and the fascia layer based on the segmentation results of each category;

[0024] Step A04: Calculate the depth of the fascia layer based on the segmentation results of the skin layer and the fascia layer.

[0025] A further improvement of the present invention lies in that in the model training step, in the improved image segmentation model, the loss function is a mixed loss function, and the mixed loss function is a cross-entropy loss function and a Dice loss function.

[0026] A further improvement of the present invention lies in that in Step A04, the pixel height of the skin layer and the pixel height of the fascia layer are calculated based on the segmentation results of the skin layer and the fascia layer, and the difference between the pixel height of the skin layer and the pixel height of the fascia layer is multiplied by a scale factor to obtain the depth of the fascia layer; the scale factor is the ratio between the true length and the pixel length of the data image.

[0027] Advantages of the present invention:

[0028] Through the settings of the horizontal translation mechanism and the vertical lifting mechanism, the ultrasonic beauty knife of the present invention can adjust the horizontal position and the vertical position of the ultrasonic sheet, so as to perform physical therapy on the skin at appropriate positions and focal depths.

[0029] The ultrasonic beauty knife of the present invention integrates detection and physical therapy. The ultrasonic sheet includes a first ultrasonic sheet and a second ultrasonic sheet. The first ultrasonic sheet can first detect the skin to determine the position of the fascia layer, and then adjust the ultrasonic sheet to an appropriate position, and the second ultrasonic sheet performs physical therapy on the skin.

[0030] The ultrasonic image obtained by the first ultrasonic sheet of the present invention is input into the improved image segmentation model, and the output is post-processed to determine the depth of the fascia layer. The present invention first detects the skin of different parts of different people, and then adjusts the ultrasonic sheet based on the detection results in a targeted manner, so as to ensure that the ultrasonic beauty knife can effectively act on the fascia layer of the skin and improve the physical therapy effect. Description of the Drawings

[0031] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 is a schematic diagram of an ultrasonic cosmetic knife of the present invention;

[0033] Figure 2 Schematic diagram of the probe of the present invention.

[0034] Description of Reference Numerals

[0035] 1. Control device, 2. Display device, 3. Handle, 4. Probe, 41. Shell, 42. Horizontal translation mechanism, 43. Vertical lifting mechanism, 44. Ultrasonic plate, 5. Skin. DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present invention are described clearly and completely below. In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0037] The purpose of the present invention is to provide an ultrasonic beauty knife, which first emits a first ultrasonic wave to detect the skin and determine the position of the fascia layer, and then adaptively adjusts the focal depth, so as to perform physical therapy in a targeted manner according to the situation of the fascia layer.

[0038] The ultrasonic beauty knife of the present invention comprises a control device 1, a display device 2, a handle 3 and a probe 4. Figure 1 shown.

[0039] The display device 2 is connected to the control device 1, and the display device 2 can be a display, a tablet, a touch screen or other display device. The display device 2 has a relevant graphical interface, and the control device 1 can be configured through the display device 2 with a graphical interface. The control device 1 is connected to the handle 3 through a connecting line. The handle 3 includes a power supply and a switch. The probe 4 is detachably arranged on the handle 3.

[0040] The probe 4 includes a housing 41, an action component and an ultrasonic sheet 44. The lower end of the housing 41 is open, and an acoustically transparent member is arranged at the opening of the lower end of the housing 41, such as Figure 2 shown.

[0041] The motion assembly is arranged inside the housing 41, and the ultrasonic sheet 44 is arranged on the motion assembly and driven by the motion assembly to move. The motion assembly includes a horizontal translation mechanism 42 and a vertical lifting mechanism 43. Preferably, the horizontal translation mechanism 42 includes a horizontal motor, a horizontal screw rod, and a horizontal slider. The horizontal motor is arranged on a bracket inside the housing 41. One end of the horizontal screw rod is connected to the rotating shaft of the horizontal motor, and the horizontal screw rod is driven by the horizontal motor to rotate. The horizontal slider is arranged on the horizontal screw rod to enable the horizontal slider to move along the horizontal screw rod.

[0042] The vertical lifting mechanism 43 is arranged on the horizontal slider. In this embodiment, the vertical lifting mechanism 43 is a telescopic motor, and the ultrasonic sheet 44 is arranged at the telescopic end of the telescopic motor.

[0043] The ultrasonic sheet 44 includes a first ultrasonic sheet and a second ultrasonic sheet. The first ultrasonic sheet is used to detect the skin 5, and the second ultrasonic sheet is used to perform physical therapy on the skin 5. Specifically, the first ultrasonic sheet is provided with an opening, and the second ultrasonic sheet is bonded to the opening of the first ultrasonic sheet.

[0044] It should be particularly noted that the power required for the devices in the probe 4 is provided by the handle 3. Specifically, the handle 3 is provided with a slot, and the probe 4 can be inserted into the slot. A power docking connector is arranged on one side of the slot, and a power interface is arranged on the side of the housing 41 corresponding to the power docking connector of the handle 3. The power interface is connected to each device through a connecting wire. The probe 4 is inserted on the handle 3, and the power docking connector is docked on the power interface, so that the handle 3 can provide the required power for the probe 4. The switch on the handle 3 is a power switch. The way of inserting the probe 4 on the handle 3 is not limited to this, and other existing ways can also be adopted as long as the functions can be realized and satisfied.

[0045] The horizontal translation mechanism 42 can drive the ultrasonic sheet 44 to move horizontally, so that the ultrasonic sheet 44 can move and stop at the corresponding points (positions) according to the points preset on the display device 2. The vertical lifting mechanism 43 can adjust the height of the ultrasonic sheet 44, so as to realize the adjustment of the focal depth.

[0046] The present invention also provides a method for using the ultrasonic beauty knife, including the following steps:

[0047] Step S1: Assemble the probe 4 on the handle 3, and preset the points and times through the display device 2.

[0048] Step S2: The bottom of the probe 4 contacts the skin coated with gel. The first ultrasonic sheet emits the first ultrasonic wave to detect the skin 5 once, and feeds back the corresponding ultrasonic image to the display device 2, and the display device 2 displays the ultrasonic image. The primary detection is used to identify the skin layers and determine the depth of the fascia layer of the skin 5.

[0049] Step S3: Based on the depth of the fascia layer, the control device 1 determines the focal depth between the ultrasonic sheet 44 and the fascia layer.

[0050] Step S4: Based on the focal depth determined in Step S3, the control device 1 adjusts the height of the ultrasonic sheet 44 through the vertical lifting mechanism 43.

[0051] Step S5: The lateral translation mechanism 42 drives the ultrasonic sheet 44 to move to a preset position, and the second ultrasonic sheet emits the second ultrasonic wave to perform physical therapy on the skin 5 according to a preset number of times.

[0052] Step S6: The first ultrasonic sheet emits the first ultrasonic wave to perform a secondary detection on the skin 5.

[0053] The secondary detection can be used to detect the state of the skin 5 after physical therapy, and accordingly, the skin 5 can be evaluated specifically to facilitate the subsequent physical therapy.

[0054] A further improvement of this application lies in: in Step S2, after the ultrasonic image is fed back to the control device 1, the control device 1 inputs the ultrasonic image into the improved image segmentation model, and then performs post-processing of the data to obtain the depth of the fascia layer.

[0055] The output of the improved image segmentation model is the segmentation probability map of each layer of the skin 5. The skin 5 includes the skin layer, the fat layer, the fascia layer, and the muscle layer.

[0056] The construction process of the improved image segmentation model is as follows:

[0057] (I) Data preparation

[0058] Construct a data set: Collect ultrasonic images of the skin 5 of different parts of different human bodies as data images in the data set, and perform hierarchical annotation.

[0059] Specifically, use a professional medical image annotation tool (ITK-SNAP or Labelbox) to perform hierarchical annotation on the data images to generate a pixel-level segmentation mask. Each layer of the skin 5 corresponds to a category and is assigned a unique category number. For example: 0 is the background, 1 is the skin layer, 2 is the fat layer, 3 is the fascia layer, and 4 is the muscle layer. After the hierarchical annotation is completed, the data set is expanded by rotating, flipping, adding noise, adjusting contrast, etc. to the data images in the data set, so as to enhance the robustness of the image segmentation model.

[0060] (II) Data preprocessing

[0061] Normalize the gray value of the data images in the data set to the range of [0, 1], and the formula is: I norm =(I - I min ) / (I max - I min), where I norm is the normalized value, I is the value before normalization, and I min is the minimum value of the grayscale of the original data image, and I max is the maximum value of the grayscale of the original data image.

[0062] In this step, the brightness distribution of the data image can also be corrected and the data image can be scaled to the same size. Specifically, Gamma correction is used to adjust the brightness distribution of the normalized data image, and the too dark or too bright parts in the normalized data image are enhanced.

[0063] (III) Model training

[0064] The data images of the preprocessed data set are input into the image segmentation model for training. After the training is completed, an improved image segmentation model is obtained. The improved image segmentation model outputs the segmentation probability maps of each category. In this embodiment, the image segmentation model can be selected from common medical image segmentation models such as U-Net, Attention U-net, DeepLabv3+, etc.

[0065] In the improved image segmentation model, the loss function is a mixed loss function to address the problem of class imbalance. In this embodiment, the mixed loss function is the cross-entropy loss function and the Dice loss function.

[0066] The formula of the mixed loss function is:

[0067] L = (1 - β)L ce + βL dice

[0068] L ce = -∑ k∈K V k logU k

[0069]

[0070] where L ce and L dice represent the cross-entropy loss function and the Dice loss function respectively, β is used to adjust the weight balance of the two loss functions, |X∩Y| represents the number of pixels in the overlapping part of the predicted region and the true region, |X| and |Y| represent the total number of pixels in the predicted region and the true region respectively, U K is the actual output, V K is the expected output, and L represents the mixed loss function.

[0071] A further improvement of this application is that the steps for post-processing the output of the improved image segmentation model are as follows:

[0072] Step A01: Threshold the segmentation probability maps of each category to obtain the binary maps of each category.

[0073] Step A02: Perform morphological operations on the binary maps of each category, and then scale them to obtain the segmentation results of each category.

[0074] Step A03: Determine the segmentation result of the skin layer and the segmentation result of the fascia layer based on the segmentation results of each category.

[0075] Step A04: Calculate the depth of the fascia layer based on the segmentation result of the skin layer and the segmentation result of the fascia layer.

[0076] Specifically, the improved image segmentation model outputs a tensor with a shape of (H, W, C), where each channel C represents the segmentation probability map P of each category. C , H represents the pixel length of each category, and W represents the pixel height of each category. Then, the segmentation probability map P of each category C is thresholded to generate a binary map B C , and morphological operations (such as opening operation and closing operation) are used to remove the noise in it, and it is scaled and restored to the original size to obtain the final segmentation result B of each category Ci , i = 0, 1, 2, 3, 4, i is the category number. From B C1 (the segmentation result of the skin layer), B C3 (the segmentation result of the fascia layer), calculate the pixel heights W1 and W3 of the skin layer and the fascia layer, take the difference between the two and multiply by the scale factor f from the real length to the pixel length to obtain the depth D of the fascia layer. Based on the depth D of the fascia layer, subsequent adjustments and physiotherapy can be carried out.

[0077] Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

Claims

1. An ultrasonic beauty knife, characterized in that: The invention comprises a control device, a display device, a handle and a probe; the control device is connected to the display device, the control device is connected to the handle, and the probe is detachably arranged on the handle; the probe comprises a shell, an action component and an ultrasonic sheet; the action component and the ultrasonic sheet are both located in the shell; the lower end of the shell is open; the action component comprises a lateral translation mechanism and a vertical lifting mechanism arranged on the lateral translation mechanism, and the ultrasonic sheet is arranged on the vertical lifting mechanism; an acoustically transparent component is arranged at the opening at the lower end of the shell.

2. The ultrasonic beauty knife according to claim 1, characterized in that: The lateral translation mechanism includes a horizontal motor, a horizontal screw and a horizontal slider; the horizontal motor is arranged in a housing; the horizontal screw is connected to the rotating shaft of the horizontal motor, the horizontal slider is arranged on the horizontal screw, and the vertical lifting mechanism is arranged on the horizontal slider.

3. The ultrasonic beauty knife according to claim 1, characterized in that: The vertical lifting mechanism is a telescopic motor, and the ultrasonic sheet is arranged at the telescopic end of the telescopic motor.

4. The ultrasonic beauty knife according to claim 1, characterized in that: The ultrasonic sheet comprises a first ultrasonic sheet for detection and a second ultrasonic sheet for physical therapy; the first ultrasonic sheet is sleeved on the outside of the second ultrasonic sheet.

5. A method for using an ultrasonic beauty knife, characterized in that: The method of use is implemented based on the ultrasonic beauty knife according to claim 4, and the method of use comprises the following steps: Step S1: The probe is assembled on the handle, and the point position and the number of times are preset through the display device; Step S2: the probe contacts the skin, and the first ultrasonic sheet emits a first ultrasonic wave to perform a primary detection on the skin and feeds back a corresponding ultrasonic image to the control device and the display device; the primary detection is used to identify the skin layers and determine the depth of the fascia layer of the skin; Step S3: Based on the depth of the fascia layer, the control device determines the focal depth between the ultrasonic blade and the fascia layer; Step S4: Based on the focal depth determined in step S3, the control device adjusts the height of the ultrasonic blade through the vertical lifting mechanism; Step S5: The transverse translation mechanism drives the ultrasonic plate to move to a preset point, and the second ultrasonic plate emits a second ultrasonic wave for a preset number of times to perform physical therapy on the skin.

6. The method for using the ultrasonic beauty knife according to claim 5, characterized in that: The step also includes step S6; Step S6: the first ultrasonic sheet emits a first ultrasonic wave to perform a secondary detection on the beauty area.

7. The method for using the ultrasonic cosmetic knife according to claim 5, characterized in that: In step S2, after the ultrasound image is fed back to the control device, the control device inputs the ultrasound image into the improved image segmentation model, and performs data post-processing on the output of the improved image segmentation model to obtain the depth of the fascia layer; the skin includes a skin layer, a fat layer, a fascia layer and a muscle layer; The steps to build the improved image segmentation model are as follows: Data preparation steps: Collect ultrasound images of the skin as data images in the data set and annotate them in layers; each layer of the skin corresponds to a category, and each category corresponds to a category number; Data preprocessing steps: grayscale normalization, brightness adjustment and image scaling operations are performed on the data images in the data set; Model training steps: input the data images in the data set after data preprocessing into the image segmentation model for training, and obtain an improved image segmentation model after the training; the improved image segmentation model outputs the segmentation probability map of each category.

8. The method for using the ultrasonic cosmetic knife according to claim 7, characterized in that: The steps for data post-processing the output of the improved image segmentation model are as follows: Step A01: Thresholding the segmentation probability map of each category to obtain a binary map of each category; Step A02: performing morphological operations on the binary images of each category and scaling them to obtain the segmentation results of each category; Step A03: determining the segmentation result of the skin layer and the segmentation result of the fascia layer based on the segmentation results of each category; Step A04: Based on the segmentation results of the skin layer and the segmentation results of the fascia layer, the depth of the fascia layer is calculated.

9. The method for using an ultrasonic cosmetic knife according to claim 7, characterized in that: In the model training step, in the improved image segmentation model, the loss function is a mixed loss function, and the mixed loss function is a cross entropy loss function and a Dice loss function.

10. The method for using an ultrasonic cosmetic knife according to claim 8, characterized in that: In step A04, the pixel height of the skin layer and the pixel height of the fascia layer are calculated based on the segmentation results of the skin layer and the fascia layer. The pixel height of the skin layer and the pixel height of the fascia layer are subtracted and multiplied by a scale factor to obtain the depth of the fascia layer; the scale factor is the ratio of the actual length of the data image to the pixel length.

Images