Melanoma surface imaging identification method
Through the combination of high-precision atomic force microscopy and terahertz near-field imaging system, the problem of insufficient identification and long diagnosis time in melanoma diagnosis is solved, high-resolution melanoma cell imaging and rapid diagnosis are achieved, and the staining steps are avoided. Improved diagnostic efficiency.
Patent Information
- Application Number
- CN202411166921.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-12-06
AI Technical Summary
The prior art has problems such as insufficient identification, difficulty in staining and long diagnosis in the diagnosis of melanoma, especially due to the low magnification of optical microscope and long immunohistochemistry time.
High-precision atomic force microscopy and terahertz near-field imaging system are used to achieve high-resolution imaging of melanoma cells through large-scale and small-scale scanning technology, identify cell characteristic values to identify melanomas, avoid staining steps, and shorten the diagnosis time.
It achieves more than 1,000 times the imaging accuracy of melanoma cells, quickly recognizes melanoma, eliminates staining steps, significantly shortens the diagnosis time and improves diagnostic efficiency.
Smart Images

Figure CN119091438A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of biological imaging, and in particular relates to a melanoma surface imaging and identification method, which can effectively image and identify the melanoma surface, and becomes a diagnostic means. Background Art
[0002] Melanoma, usually refers to malignant melanoma, is a highly malignant tumor derived from melanocytes, referred to as melanoma. It mostly occurs in the skin, and can also be seen in mucous membranes and internal organs, accounting for about 3% of all tumors. Cutaneous malignant melanoma ranks third among malignant skin tumors (accounting for about 6.8%~20%). It is more common in adults, with a high incidence in white people with fair skin, and a lower incidence in Asians and Africans with dark skin, and it is rarely seen in children. Some patients have familial multiple phenomena. Malignant melanoma can evolve from congenital or acquired benign melanocytic nevi, or from the malignant transformation of dysplastic nevi, or it can be a new occurrence.
[0003] The current diagnosis of melanoma has some defects. The initial diagnosis of melanoma in hospitals is mainly based on morphological observation and cell counting under a microscope. Since the conventional optical microscope has a low magnification of 100-500 times, the morphological observation is not detailed enough and the cell counting is inaccurate, which leads to inaccurate identification and diagnosis of melanoma.
[0004] Melanoma is difficult to stain or the staining is not obvious. In the pathological diagnosis of melanoma, the characteristics of melanoma cells themselves may affect the staining effect. Melanoma cells contain a large number of melanin particles, which may mask or interfere with the combination of dyes and cell structures, resulting in unclear staining. At the same time, the heterogeneity of tumor tissue may also affect the staining results. The morphology and tissue structure of melanoma may vary greatly, which may make some areas difficult to be marked by dyes, thereby affecting the identification and diagnosis of melanoma.
[0005] Immunohistochemistry of melanoma is an indispensable auxiliary pathological diagnosis technology, which can further confirm melanoma based on the preliminary diagnosis. It uses specific antibodies to mark specific proteins in tumor cells, thereby helping to identify the type and source of the tumor. In the diagnosis of melanoma, we use specific antibodies to identify melanocyte markers, such as S-100 protein, HMB-45, and Melan-A. Based on the positive expression of antibodies, it can be determined whether the tumor is melanoma. Normally, melanoma cells will show positive expression of markers such as S-100 protein, HMB-45, and Melan-A. However, this process generally consumes a lot of time, generally 5-7 days. Only a few large hospitals in China can perform this, and general tertiary hospitals do not have this means. Summary of the invention
[0006] In view of the technical problems of melanoma identification in the prior art mentioned in the background technology, the present invention proposes a melanoma surface imaging identification method. The present invention uses a new imaging system to firstly use a large-scale scanning method to find the location of the melanoma in the entire tissue, and then narrow the range of scanning to image the melanoma surface. The imaging accuracy exceeds the micron-level accuracy limit of the optical microscope, and the imaging magnification can reach more than a thousand times. Whether it is a melanoma can be identified by the cell characteristics of the surface imaging image.
[0007] The present invention is achieved through the following technical solutions: A melanoma surface imaging recognition method comprises the following steps: Step 1: After placing the cell sample on the stage, fix the stage, turn on the laser, automatically tune the needle tip, adjust the magnification of the atomic force microscope, and find the area to be tested.
[0008] Step 2: First, perform a first large-scale scan of the cell sample, and after the scan is completed, an image with a range of 20μm is obtained; secondly, the image with a range of 20μm is used to confirm the area for further scanning, and a second fine scan with a small range is performed, and after the scan is completed, an image with a range of 10μm can be obtained.
[0009] Step 3: For the imaging image in the range of 10 μm obtained in step 2, cell characteristic values are identified. When the number of the following cell characteristic values in the imaging image is greater than 5, the cells in the imaging image are identified as melanoma cells.
[0010] 1) The cell edges are unclear; 2) Poor cell symmetry; 3) The cells are heteromorphic; 4) The cell surface is uneven; 5) Uneven color and irregular pigment distribution.
[0011] In the above technical solution, in step 2, the first large-range scan selects a scanning range of 20 μm, a scanning frequency of 0.4 Hz, and a resolution of 512.
[0012] In the above technical solution, in step 2, the second fine scan selects a scanning range of 10 μm, a scanning frequency of 0.2 Hz, and a resolution of 256.
[0013] In the above technical solution, the identification method adopts a terahertz near-field imaging system with an imaging accuracy of nanometer level.
[0014] Beneficial effects: 1. Melanoma cell surface imaging mainly utilizes the high precision of the imaging system, and adopts large-scale measurement and reduced-range measurement to ensure the high resolution of the position and surface imaging of melanoma cells. The resulting image can reach an imaging level of more than a thousand times, and the details of the cell surface can be observed, solving the problem of difficult melanoma identification caused by the low magnification of optical microscope imaging.
[0015] 2. In the present invention, melanoma cells are imaged with high precision in the imaging system. The imaging level of more than a thousand times can clearly show the details of the cell surface. The melanoma that meets the characteristics can be identified through the morphological characteristics of the surface imaging image, thereby eliminating the step of melanoma staining, thereby solving the problem of difficulty in identifying melanoma caused by the difficulty in staining the surface of conventional pathological melanoma.
[0016] 3. To solve the problem of long diagnosis time for melanoma, the melanoma imaging technology of the present invention can achieve imaging within 1 hour, solving the problem of difficulty in identifying melanoma caused by long immunohistochemistry time and slow diagnosis speed in the further diagnosis process. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The present invention is further described below in conjunction with the accompanying drawings and embodiments.
[0018] Figure 1 This is a picture of a melanoma taken under an optical microscope; Figure 2 This is a cell surface image of melanoma at 20 μm. Figure 3 This is a cell surface image of melanoma at a range of 10μm. DETAILED DESCRIPTION
[0019] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.
[0020] Example like Figure 1 As shown in the figure, this is an image of melanoma under an ordinary optical microscope. The magnification of the eyepiece is 10 times, and the magnification of the objective lens is 50 times. The image with a magnification of 500 times is obtained. It can be seen that at this magnification, the details of the cells cannot be distinguished, and only some general trends of the overall cells can be seen, making it difficult to determine whether it is melanoma. Figure 2 and Figure 3As shown, this is an image of the surface of melanoma cells by the imaging system, which adopts the needle tip point-by-point scanning method, with a scanning range of 10μm and 20μm respectively, using 512 and 256 scanning points, and dividing the scanning range by the number of scanning points, it can be obtained that the accuracy of each scanning point is 19 and 39 nanometers, so the imaging accuracy can reach the nanometer level.
[0021] 1. Terahertz near-field imaging system Figure 2 and Figure 3 The imaging diagram uses a terahertz near-field imaging system. For the terahertz band, due to its relatively long wavelength (30μm~3mm), the terahertz far-field imaging technology has limitations in measuring relevant information of samples in a small area. The terahertz s-SNOM near-field imaging technology based on metal tips performs well. It can provide nanometer-level spatial resolution, which is basically not related to the wavelength of the terahertz radiation used. When the incident terahertz wave is focused on the metal tip, due to the lightning rod effect, a strongly constrained near field similar to the radius of curvature of the tip is formed at the top of the tip. If the tip is close enough to the sample being measured, it can be observed that the enhanced near field at the tip can be equivalent to scattered light containing sample information. Therefore, by recording the scattered light, the terahertz s-SNOM system can obtain a high-resolution optical scanning image with an imaging accuracy of about tens of nanometers. The terahertz near-field imaging system is generally composed of a terahertz source, a mirror module, a probe, etc., and works in reflection mode. The terahertz radiation emitted from the terahertz source is focused and transmitted by the mirror module and then converged to the probe tip, and then hits the test sample through the tip. The interaction between the tip and the sample to be tested realizes the modulation of the terahertz wave, and then the modulated terahertz wave is reflected back by the tip and converged to the detector through the mirror module. Finally, the received terahertz wave is processed to obtain an imaging image. The terahertz near-field imaging system is based on the atomic force microscope system, which consists of a terahertz source, a terahertz optical path, a parabolic mirror, a probe platform, etc. The terahertz wave is emitted by the terahertz source, passes through the terahertz optical path, reaches the parabolic mirror and is focused on the front end needle tip of the probe in the atomic force microscope. The probe needle tip is above the sample. The terahertz wave reflected by the probe carries the near-field information of the sample and returns to the terahertz optical path through the parabolic mirror, and reaches the signal processing system of the atomic force microscope, realizing the high-resolution imaging function of the terahertz near field. Its imaging accuracy is generally nanometer-level accuracy, and the accuracy of tens of nanometers or even a few nanometers can be obtained through this system.
[0022] 2. Test method of imaging image After placing the sample on the stage, fix the vibration stage. Open the test software, turn on the laser, automatically adjust the frequency of the needle tip, adjust the magnification of the atomic force microscope, and find the area to be tested. First, scan the sample over a large range. Generally, a 20μm scanning range is selected, a scanning frequency of 0.4hz is selected, and a resolution of 512 is selected. After the scan is completed, an image of a 20μm range can be obtained; secondly, the 20μm range scanning image is used to confirm the area that needs further scanning, and a small range of fine scanning is performed. Generally, a 10μm scanning range is selected, a scanning frequency of 0.2hz is selected, and a resolution of 256 is selected. After the scan is completed, an image of a 10μm range can be obtained.
[0023] Through Figure 2 and Figure 3 The size can be measured, Figure 2 The geometric size is 9.3cm, Figure 3 The geometric size is 8.9cm, and Figure 2 The actual measurement size of the image is 20μm. Figure 3 The actual measured size of the image is 11.5μm. Dividing the geometric size by the actual measured size of the image can give the magnification of the imaging system, 9.3cm / 20μm=4650 times, 8.9cm / 10.5μm=8476 times. Therefore, the magnification of the imaging system is more than 4600 times, which is about 9 times more than the 500 times of the optical microscope. Figure 2 and Figure 3 Irregular characteristic details of the melanoma surface can be seen in the figure. Multiple irregular cell features can be used to determine that the cell sample is melanoma, which can be a means of diagnosing and identifying melanoma.
[0024] By comparing the images of ordinary optical microscopes and the images of the new imaging system, the difference between the imaging at magnifications of 500 times and 4600 times can be seen. The images produced by the present invention can clearly show the details of the cell surface, providing a means for the identification of melanoma and solving the problem of the difficulty in identifying melanoma due to the low magnification of optical microscope imaging.
[0025] Through the example of melanoma diagnosis, it can be seen that the current diagnosis time of melanoma is generally 5-7 days, which is mainly because the immunohistochemical diagnosis time in the diagnosis link is relatively long. Usually, immunohistochemical diagnosis requires multiple rounds of staining of tissue samples, and then the diagnosis and judgment of the positive and negative results of multiple rounds of staining are made, which leads to a long diagnosis time. The imaging system used in this patent generally has a scanning frequency of 0.4 Hz when scanning in a large range, that is, a scanning cycle is 1 / 0.4=2.5 seconds, and a scanning frequency of 0.2 Hz when scanning in a small range, that is, a scanning cycle is 1 / 0.2=5 seconds. The imaging system adopts a needle-point scanning method. Generally, the number of scanning points in a small range is 256 or 512, and one scanning point is 1 cycle, so one imaging requires 256 or 512 cycles. This patent uses one large-range scan and one small-range scan, that is, a complete imaging requires 256×5×2=2560 seconds=43 minutes. Therefore, the imaging time required for this patent is generally 43 minutes, and imaging can be achieved within one hour, which is much faster than the diagnosis time required for immunohistochemistry. This greatly improves the doctor's diagnostic efficiency and solves the problem of difficulty in identifying melanoma due to the long time of immunohistochemistry and the fact that only a few hospitals have immunohistochemistry diagnostic means.
[0026] The conventional melanoma diagnosis process requires not only observation under an optical microscope, but also staining of the tissue to help identify the type and source of tumor cells, especially for the diagnosis of melanoma. However, the large number of melanin particles in melanoma cells themselves and the heterogeneity of the tissue themselves may affect the staining effect. In the present invention, melanoma cells can be tested in an imaging system to obtain imaging images with a nanometer-level imaging accuracy of more than a thousand times. The imaging details of the cell surface can be observed through high-precision imaging images. The morphological characteristic values of melanoma cells obtained by a large number of tests can be used to identify some melanoma cells that meet the characteristic values, thereby helping doctors to identify and diagnose, and there is no need for subsequent staining diagnosis work. This solves the problem of difficulty in identifying melanoma due to the difficulty in staining of conventional pathological melanoma.
[0027] 3. Identification of cell feature values in imaging images By analyzing and comparing a large number of melanoma surface images, some irregular shape feature values of melanoma cells can be obtained. If the number of irregular shape feature values of tumor cells in a picture is greater than 5, it can be considered as melanoma, thus achieving the purpose of identifying melanoma. Figure 2 and 3 As shown, these are images of melanoma cells at 20μm and 10μm respectively. Careful observation shows irregular feature values, such as blurred boundaries, poor symmetry, anisotropy, unevenness, and unequal surface feature values.
[0028] 1) The cell edges are blurred and it is difficult to see the clear boundaries of the cell edges; 2) The symmetry of the cells is poor. It can be seen that the cells are not symmetrical left and right, and the differences are large; 3) The cells are atypical. You can see that the entire cell is not a regular shape like normal cells. Melanoma cells have irregular shapes. 4) The cell surface is uneven. It can be observed that the height of the cells in the lower right corner is significantly higher than that of the cells in the left half. The surface height of the cells varies and fluctuates. 5) The color is uneven and the pigment is irregularly distributed. It is obvious that the melanin is concentrated in the lower right corner of the cell and the color is very dark, while the left half of the cell is lighter in color and has almost no pigment.
[0029] By identifying these cell feature values, when the number is greater than 5, melanoma cells can be identified. When a tumor surface image is obtained by scanning, we can use it to determine whether it is melanoma.
[0030] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A melanoma surface imaging and recognition method, It is characterized in that The steps include: Step 1: After placing the cell sample on the stage, fix the stage, turn on the laser, automatically tune the needle tip, adjust the magnification of the atomic force microscope, and find the area to be tested; Step 2: First, perform a first large-scale scan of the cell sample, and after the scan is completed, an image of a 20μm range is obtained; secondly, the image of a 20μm range is used to confirm the area for further scanning, and a second fine scan of a small range is performed, and after the scan is completed, an image of a 10μm range can be obtained; Step 3: Identify the cell characteristic values for the imaging image in the range of 10 μm obtained in step 2. When the number of the following cell characteristic values in the imaging image is greater than 5, identify the cells in the imaging image as melanoma cells; 1) The cell edges are unclear; 2) Poor cell symmetry; 3) The cells are heteromorphic; 4) The cell surface is uneven; 5) Uneven color and irregular pigment distribution.
2. A melanoma surface imaging and identification method as claimed in claim 1, Features: In step 2, the first large-scale scan selects a scanning range of 20 μm, a scanning frequency of 0.4 Hz, and a resolution of 512.
3. A melanoma surface imaging and identification method as claimed in claim 1, Features: In step 2, the second fine scan selects a scanning range of 10 μm, a scanning frequency of 0.2 Hz, and a resolution of 256.
4. A melanoma surface imaging and identification method as claimed in claim 1, Features: The identification method adopts a terahertz near-field imaging system, and the imaging accuracy is at the nanometer level.
Citation Information
Patent Citations
Reflection type real-time tumor imaging method and system
CN107095673A
Near-field microscopy system and construction method thereof
CN111044481A
Terahertz lesion detection method and system based on artificial intelligence
CN111311574A
Melanoma identification method and device based on deep learning
CN115082691A
