Prediction method and system for thyroid papillary carcinoma lymph node metastasis and electronic equipment

Abstract

The invention provides a papillary thyroid carcinoma lymph node metastasis prediction method, system and equipment. The method comprises the following steps: acquiring a staining collagen image of a first papillary thyroid carcinoma section stained by a hematoxylin eosin staining method; obtaining a multi-photon collagen image of the second papillary thyroid carcinoma slice through a multi-photon imaging method; setting an observation area on the multi-photon collagen image according to the dyed collagen image; extracting collagen characteristics in the observation area; constructing a collagen feature scoring model according to an LASSO algorithm and the collagen features; obtaining a collagen feature score according to the collagen feature scoring model; acquiring clinical risk factors; the papillary thyroid carcinoma metastasis risk information is output according to the clinical risk factors and the collagen feature scores, the collagen features in the papillary thyroid carcinoma microenvironment can be accurately and quantitatively extracted by combining the multi-photon imaging technology and collagen feature analysis, accurate prediction of the papillary thyroid carcinoma metastasis risk is achieved, and the accuracy of the papillary thyroid carcinoma metastasis risk is improved. And the prediction accuracy can be further improved in combination with clinical risk factors.

Description

Technical Field

[0001] The present application relates to the field of medical prediction technology, and in particular to a method, system and electronic device for predicting lymph node metastasis of papillary thyroid carcinoma. Background Art

[0002] Papillary thyroid carcinoma (PTC) is the most common thyroid malignancy, accounting for about 80% of all thyroid cancer cases. Although the survival rate of this disease is high, metastasis of papillary thyroid carcinoma is still an important factor affecting the patient's prognosis. Early detection of metastasis of papillary thyroid carcinoma is of great significance for formulating treatment strategies, evaluating prognosis, and optimizing surgical plans. Existing preoperative detection methods include ultrasound examination, CT / MRI, fine needle aspiration biopsy, etc., but these methods have certain limitations.

[0003] Currently, the methods provided in the related art are unable to detect the metastasis and micrometastasis of papillary thyroid carcinoma that cannot be palpated. Summary of the invention

[0004] In view of this, it is necessary to provide a method, system and electronic device for predicting lymph node metastasis of papillary thyroid carcinoma, which can overcome at least one of the above defects.

[0005] In a first aspect, an embodiment of the present application provides a method for predicting lymph node metastasis of papillary thyroid carcinoma, the method comprising: Acquire the stained collagen image of the first papillary thyroid carcinoma section after staining with hematoxylin and eosin; Multiphoton collagen images of the second thyroid papillary carcinoma section were obtained by multiphoton imaging; Setting an observation area on the multiphoton collagen image according to the stained collagen image; Extracting collagen features within the observation area; Constructing a collagen feature scoring model according to the LASSO algorithm and the collagen features; Obtaining a collagen characteristic score according to the collagen characteristic scoring model; Obtain clinical risk factors; The metastasis risk information of papillary thyroid carcinoma is output according to the clinical risk factors and the collagen characteristic score.

[0006] In one embodiment, the collagen characteristics include morphological parameters, intensity parameters and texture parameters; The morphological parameters include collagen area, collagen quantity, collagen length, collagen width, collagen stiffness, collagen cross-linking density, collagen cross-linking space, and collagen arrangement; The strength parameters include mean, variance, skewness, kurtosis, energy, and entropy.

[0007] In one embodiment, the texture parameters are extracted by gray level co-occurrence matrix or Gabor filter; When feature extraction is performed through the gray level co-occurrence matrix, the texture parameters include contrast, correlation, uniformity, gray level co-occurrence matrix, five distances and four directions; When feature extraction is performed through the Gabor filter, the texture parameters include mean, variance, four scales and six directions.

[0008] In one embodiment, the formula of the collagen characteristic scoring model is: 0.188×collagen cross-linking density-0.116×collagen distribution-0.077×collagen intensity mean-0.030×collagen grayscale co-occurrence matrix contrast in the second direction of 45 degrees+0.118×Gabor filter image convolution at the mean of the fifth direction of the second scale.

[0009] In one embodiment, the multi-photon imaging method uses an imaging magnification of 20 times, an imaging area size of 1 mm×1 mm, and the imaging area is 5 blocks.

[0010] In one embodiment, the multiphoton imaging method includes extracting the spatial structural characteristics and arrangement regularity of collagen through second harmonic generation images.

[0011] In one embodiment, the papillary thyroid carcinoma metastasis risk information is presented in the form of a nomogram.

[0012] In one embodiment, the method further comprises: The metastasis risk score of papillary thyroid carcinoma is output according to the nomogram.

[0013] In a second aspect, an embodiment of the present application provides a papillary thyroid carcinoma metastasis prediction system, which is applied to the papillary thyroid carcinoma lymph node metastasis prediction method as described in the first aspect, and the system comprises: A first image acquisition module is used to acquire a stained collagen image of the first thyroid papillary carcinoma section stained by hematoxylin and eosin staining; A second image acquisition module is used to acquire a multiphoton collagen image of the second thyroid papillary carcinoma slice by a multiphoton imaging method; A feature extraction module, used to set an observation area on the multiphoton collagen image according to the stained collagen image, and extract collagen features in the observation area; A model building module, used to build a collagen feature scoring model based on the LASSO algorithm and the collagen features; The risk information output module is used to obtain the collagen characteristic score according to the collagen characteristic scoring model; obtain clinical risk factors; and output the thyroid papillary carcinoma metastasis risk information according to the clinical risk factors and the collagen characteristic score.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, including: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to implement the method for predicting lymph node metastasis of papillary thyroid carcinoma as described in the first aspect when executing the instructions.

[0015] The papillary thyroid carcinoma lymph node metastasis prediction method, system and electronic device provided in the embodiments of the present application can accurately and quantitatively extract the collagen characteristics in the papillary thyroid carcinoma microenvironment by combining multiphoton imaging technology with collagen feature analysis, thereby achieving accurate prediction of the metastasis risk of papillary thyroid carcinoma and further improving the accuracy of the prediction in combination with clinical risk factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic flow chart of a method for predicting lymph node metastasis of papillary thyroid carcinoma provided in one embodiment of the present application.

[0017] Figure 2 This is a stained collagen image after staining using the hematoxylin and eosin staining method provided in an embodiment of the present application.

[0018] Figure 3 A multi-photon collagen image obtained by a photon imaging method provided in one embodiment of the present application.

[0019] Figure 4 A risk matrix diagram is provided for an embodiment of the present application.

[0020] Figure 5 A schematic diagram of a papillary thyroid carcinoma metastasis prediction system module provided in one embodiment of the present application.

[0021] Figure 6 A schematic diagram of an electronic device provided in one embodiment of the present application.

[0022] Main component symbols DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0024] It should be noted that, in the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art in the present application. The terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application.

[0025] It should be noted that, in the embodiments of the present application, words such as "first" and "second" are only used for the purpose of distinguishing descriptions, and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying order. Features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way.

[0026] Based on the implementations in this application, all other implementations obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0027] Papillary thyroid carcinoma (PTC) is the most common thyroid malignancy, accounting for about 80% of all thyroid cancer cases. Although the survival rate of this disease is high, metastasis of PTC is still an important factor affecting the prognosis of patients. Early detection of PTC metastasis is of great significance for formulating treatment strategies, evaluating prognosis, and optimizing surgical plans. Existing preoperative detection methods include ultrasound, CT / MRI, and fine needle aspiration biopsy, but these methods have certain limitations. Although ultrasound examination is simple and easy to obtain, its sensitivity in detecting PTC metastasis is still insufficient, especially for lymph nodes that cannot be palpated, the detection rate of ultrasound is only 20%-33%. In addition, CT and MRI examinations are mainly used for cases with large tumors or those located behind the sternum, and the efficiency of detecting PTC metastasis is low. Although fine needle aspiration biopsy is effective in some cases, it depends on the accuracy of ultrasound and is only applicable to larger lymph nodes, which cannot effectively cover the needs of most patients.

[0028] At present, although the above methods provide some help for the early diagnosis of papillary thyroid carcinoma metastasis, their clinical effects are still limited, especially in the detection of papillary thyroid carcinoma metastasis and micrometastasis that cannot be palpated. However, the current research on collagen characteristics has not been directly combined with clinical diagnosis and treatment, and there is still a lack of effective quantitative prediction models to accurately assess papillary thyroid carcinoma metastasis and the prognosis risk of patients. Therefore, the development of a prognosis prediction model based on quantitative analysis of tumor microenvironment collagen has important clinical application value.

[0029] In view of this, the method, system and electronic device for predicting lymph node metastasis of papillary thyroid carcinoma provided in an embodiment of the present application can accurately and quantitatively extract collagen features in the microenvironment of papillary thyroid carcinoma by combining multiphoton imaging technology with collagen feature analysis, and provide a new technical means for predicting the metastasis of papillary thyroid carcinoma. Compared with traditional imaging methods, the present invention can overcome the limitations of detection technologies such as ultrasound, CT, and MRI in the early diagnosis of papillary thyroid carcinoma metastasis, especially in the case where the lymph nodes are not obvious or difficult to palpate, and has higher sensitivity and specificity. In addition, by constructing a collagen feature scoring model based on LASSO regression, the present invention realizes the accurate prediction of the risk of papillary thyroid carcinoma metastasis, and can further improve the accuracy of prediction in combination with clinical risk factors. The collagen feature score finally output can not only provide clinicians with auxiliary decision-making basis and optimize treatment plans, but also help patients achieve personalized treatment and improve prognosis. Therefore, the implementation scheme of the present application has a high clinical application value and can significantly improve the level of early diagnosis and prognosis evaluation of papillary thyroid carcinoma.

[0030] Figure 1 FIG. 1 is a flow chart of a method for predicting lymph node metastasis of papillary thyroid carcinoma provided in an embodiment of the present application. Figure 1 As shown, the method for predicting lymph node metastasis of papillary thyroid carcinoma includes at least the following steps: S100: obtaining a stained collagen image of a first papillary thyroid carcinoma slice stained with hematoxylin and eosin staining; S200: obtaining a multiphoton collagen image of a second papillary thyroid carcinoma slice by a multiphoton imaging method; S300: setting an observation area on the multiphoton collagen image according to the stained collagen image; S400: extracting collagen features within the observation area; S500: constructing a collagen feature scoring model according to the LASSO algorithm and collagen features; S600: obtaining a collagen feature score according to the collagen feature scoring model; S700: obtaining clinical risk factors; S800: outputting papillary thyroid carcinoma metastasis risk information according to the clinical risk factors and the collagen feature score.

[0031] S100: Acquire a stained collagen image of the first thyroid papillary carcinoma section stained with hematoxylin and eosin staining.

[0032] In the embodiment of the present application, the method for predicting lymph node metastasis of papillary thyroid carcinoma includes, in step S100, obtaining a stained collagen image of a first papillary thyroid carcinoma section stained with hematoxylin and eosin.

[0033] Specifically, routine pathological processing was performed on surgical specimens of papillary thyroid carcinoma, and tissue sections were stained by hematoxylin and eosin staining to highlight extracellular matrix components such as collagen. Next, the stained image data was obtained, which can effectively show the distribution of collagen in the tissue and provide a basis for subsequent collagen feature extraction.

[0034] It is understandable that step S100 is to provide comparison and reference for subsequent multiphoton imaging, and to help identify potential metastatic areas by comparing the differences between the staining image and the multiphoton image.

[0035] S200: Multiphoton collagen image of the second thyroid papillary carcinoma slice obtained by multiphoton imaging method.

[0036] In the embodiment of the present application, the method for predicting lymph node metastasis of papillary thyroid carcinoma includes, in step S200, acquiring a multiphoton collagen image of the second papillary thyroid carcinoma slice by a multiphoton imaging method.

[0037] Specifically, multiphoton imaging technology is used to non-destructively image slices of papillary thyroid carcinoma and obtain the second harmonic signal of collagen in the tissue. This method can capture the spatial structural information of collagen in real time without the need for staining.

[0038] It is understandable that step S200 can provide high-resolution image data for subsequent collagen feature extraction, thereby enabling the morphological features of collagen (such as collagen cross-linking density, collagen arrangement, etc.) to be more accurately extracted and analyzed.

[0039] Please also read Figure 2 and Figure 3 . Figure 2 This is a stained collagen image after staining using the hematoxylin and eosin staining method provided in an embodiment of the present application. Figure 3 This is a multi-photon collagen image obtained by a photon imaging method provided in an embodiment of the present application.

[0040] It is understandable that the first and second thyroid papillary carcinoma sections are two continuous sections with a thickness of 5μm. During pathological analysis, in order to ensure the representativeness and integrity of the sample, two continuous tissue sections were cut from the tumor tissue of thyroid papillary carcinoma, each with a thickness of 5 microns (μm). In order to ensure that comprehensive information about the tumor tissue is obtained, the tumor specimen is usually divided into continuous thin slices through cutting technology. These sections will be arranged in a certain order to ensure that no important part of the tissue is missed. Continuous sections help to observe changes in different levels and regions of the tissue, especially when performing histological examinations, which can retain the structural characteristics of the lesions to the greatest extent.

[0041] The first slice is used for routine hematoxylin-eosin (HE) staining, which is a common pathological staining method used to observe the morphological structure of tissues and help doctors perform basic pathological analysis.

[0042] The second slice is used for multiphoton imaging. Multiphoton microscopy can obtain higher-resolution images, which is particularly suitable for analyzing collagen structure and other microscopic features in tissues, and can provide more in-depth pathological information.

[0043] The thickness of tissue sections is generally controlled at 5 microns (μm), which is the standard section thickness commonly used in pathology. The section thickness of 5μm is thin enough to effectively show the structure of the tissue while retaining enough tissue details for detailed analysis. When observed under a microscope, this section thickness can ensure clear images while avoiding the impact of section thickness on analysis results.

[0044] S300: setting an observation area on the multiphoton collagen image according to the stained collagen image.

[0045] In the embodiment of the present application, the method for predicting lymph node metastasis of papillary thyroid carcinoma includes, in step S300 , setting an observation area on the multiphoton collagen image according to the stained collagen image.

[0046] Specifically, by comparing the stained collagen image and the multiphoton image, the region of interest (ROI), i.e. the observation area, can be selectively marked on the multiphoton image. These areas are usually located at the junction of the tumor and normal tissue, and extend to the tumor tissue and normal tissue, forming the invasion edge area for further collagen feature analysis.

[0047] It is understandable that by combining two different imaging methods, the microenvironmental regions associated with papillary thyroid carcinoma metastasis can be more accurately defined, ensuring that the extracted collagen features can reflect the aggressive characteristics of the tumor.

[0048] S400: Extract collagen features within the observation area.

[0049] In the embodiment of the present application, the method for predicting lymph node metastasis of papillary thyroid carcinoma includes extracting collagen features in the observation area in step S400.

[0050] Specifically, within the observation area determined in S300, various morphological features (such as density, length, distribution, etc. of collagen), intensity features (such as energy, skewness, kurtosis, etc.) and texture features (such as gray-level co-occurrence matrix, Gabor filter features, etc.) of collagen are extracted from the multiphoton imaging image through computer algorithms.

[0051] It is understandable that these characteristic parameters will be used to construct the subsequent collagen feature scoring model, so their accurate extraction is crucial to the predictive performance of the model.

[0052] S500: Construct a collagen feature scoring model based on the LASSO algorithm and collagen features.

[0053] In the embodiment of the present application, the method for predicting lymph node metastasis of papillary thyroid carcinoma includes, in step S500, constructing a collagen feature scoring model according to the LASSO algorithm and collagen features.

[0054] Specifically, the extracted collagen features were used as input variables through the LASSO regression algorithm, and the features most relevant to the risk of papillary thyroid carcinoma metastasis were selected to establish a mathematical model. The model obtained an overall collagen feature score by weighting the importance of each feature.

[0055] It is understandable that the LASSO algorithm effectively avoids overfitting through feature selection, and can screen out the collagen features that truly contribute to the prediction, ensuring the simplicity and efficiency of the model.

[0056] S600: Obtaining a collagen feature score according to a collagen feature scoring model.

[0057] In the embodiment of the present application, the method for predicting lymph node metastasis of papillary thyroid carcinoma includes, in step S600, obtaining a collagen feature score according to a collagen feature scoring model.

[0058] Specifically, the collagen features extracted in step S400 are input into the collagen feature scoring model constructed in step S500 to calculate the collagen feature score of each slice.

[0059] It can be understood that the collagen signature score provides a quantitative indicator to help clinical judgment whether a patient is at risk of metastasis of papillary thyroid carcinoma.

[0060] S700: Obtain clinical risk factors.

[0061] In the embodiment of the present application, the method for predicting lymph node metastasis of papillary thyroid carcinoma includes obtaining clinical risk factors in step S700.

[0062] Specifically, clinical risk factors include, but are not limited to, the patient's age, tumor stage, tumor size, lymphadenopathy, serum markers, etc. These factors are usually obtained through routine examinations in clinical practice.

[0063] It is understandable that clinical risk factors can provide further reference for collagen signature scoring, comprehensively consider individual differences among patients, and improve the accuracy of model predictions.

[0064] S800: Output metastasis risk information of papillary thyroid carcinoma based on clinical risk factors and collagen signature score.

[0065] In the embodiment of the present application, the method for predicting lymph node metastasis of papillary thyroid carcinoma includes, in step S800, outputting papillary thyroid carcinoma metastasis risk information according to clinical risk factors and collagen feature scores.

[0066] Specifically, the collagen characteristic score in step S600 and the clinical risk factors in step S700 are combined to output the final papillary thyroid carcinoma metastasis risk assessment result through the constructed comprehensive model. The result can be presented through visualization tools such as nomograms to help clinicians make more accurate diagnosis and treatment decisions.

[0067] It can be understood that the output of this step not only provides the risk level of papillary thyroid carcinoma metastasis, but also assists clinical judgment of the patient's prognosis and provides a reference for subsequent treatment plans.

[0068] In the embodiments of the present application, collagen characteristics include morphological parameters, strength parameters and texture parameters. Morphological parameters include collagen area, collagen quantity, collagen length, collagen width, collagen stiffness, collagen cross-linking density, collagen cross-linking space, and collagen arrangement. Strength parameters include mean, variance, skewness, kurtosis, energy, and entropy.

[0069] Specifically, by extracting and quantifying these parameters through computer algorithms, the structural and functional status of collagen can be deeply analyzed. These parameters can reflect the quantity, distribution, arrangement and physical and chemical properties of collagen in the tumor microenvironment, thus providing an effective basis for predicting the metastasis of papillary thyroid carcinoma.

[0070] Understandably, these collagen signatures can not only reveal the basic structure of the tumor microenvironment, but also capture the microscopic changes that affect tumor aggressiveness through multidimensional analysis, thereby improving the accuracy of metastasis prediction for papillary thyroid carcinoma.

[0071] In the embodiment of the present application, the texture parameters are extracted by gray level co-occurrence matrix or Gabor filter. When feature extraction is performed by gray level co-occurrence matrix, the texture parameters include contrast, correlation, uniformity, gray level co-occurrence matrix, five distances and four directions. When feature extraction is performed by Gabor filter, the texture parameters include mean, variance, four scales and six directions.

[0072] Specifically, the gray-level co-occurrence matrix is ​​used to describe the relationship between the gray levels of pixels in an image. By calculating its contrast, correlation, uniformity, etc., it can reveal the microstructural changes of collagen in the tumor microenvironment, such as the density and arrangement of collagen fibers.

[0073] Understandably, these texture parameters can provide additional information for the collagen feature model, help evaluate the complexity and heterogeneity of tumor tissue, and thus provide a more accurate basis for the risk assessment of papillary thyroid carcinoma metastasis.

[0074] Specifically, Gabor filters are used to extract texture features of images. By filtering at different scales and directions, the changes of collagen in different directions and frequencies can be effectively captured.

[0075] Understandably, this method helps to identify the directionality and spatial frequency information of collagen fibers, thereby providing richer details for the microstructural analysis of collagen and further improving the sensitivity of metastasis prediction of papillary thyroid carcinoma.

[0076] In an embodiment of the present application, the formula of the collagen feature scoring model is: 0.188×collagen cross-linking density-0.116×collagen distribution-0.077×collagen intensity mean-0.030×collagen grayscale co-occurrence matrix contrast in the second direction of 45 degrees + 0.118×Gabor filter image mean convolution in the fifth direction of the second scale.

[0077] Specifically, the scoring model calculates a comprehensive score by weighting the contribution of each collagen characteristic parameter, reflecting the potential impact of the collagen microenvironment on the metastasis of papillary thyroid carcinoma.

[0078] It is understandable that the coefficients in this formula are optimized by the LASSO regression algorithm to ensure the simplicity and efficiency of the model. Through this scoring model, the collagen characteristics of the tumor microenvironment can be quantitatively evaluated, providing a practical metastasis risk prediction tool for clinical practice.

[0079] In the embodiment of the present application, the imaging magnification adopted by the multi-photon imaging method is 20 times, the imaging area size is 1mm×1mm, and a total of five imaging areas are generated.

[0080] Specifically, multiphoton imaging technology uses a higher magnification and a smaller imaging area to accurately obtain the spatial structural information of collagen in the tumor microenvironment.

[0081] It can be understood that this high magnification and appropriate imaging area size can effectively capture subtle changes in collagen, ensuring that the extracted collagen features have high resolution, further improving the accuracy of model predictions.

[0082] In an embodiment of the present application, the multiphoton imaging method includes extracting the spatial structural characteristics and arrangement rules of collagen through second harmonic imaging.

[0083] Specifically, multiphoton imaging can non-destructively capture the microstructure of collagen using second harmonic signals, providing information about spatial features such as collagen fiber cross-linking and arrangement.

[0084] Understandably, this technology can deeply analyze the collagen status of the tumor microenvironment, reveal the role of collagen in tumor invasion, and provide basic data for subsequent collagen feature extraction and scoring.

[0085] In an embodiment of the present application, the metastasis risk information of papillary thyroid carcinoma is presented in the form of a nomogram.

[0086] Specifically, the nomogram can be used to present the prediction results of the model in a visual way, helping clinicians understand the impact of various factors on the metastasis of papillary thyroid carcinoma.

[0087] It is understandable that the output in the form of a nomogram can enable doctors to more intuitively assess the patient's metastasis risk, further enhancing the practicality and acceptance of the model in actual clinical applications.

[0088] In an embodiment of the present application, the method further includes outputting a papillary thyroid carcinoma metastasis risk score according to a nomogram.

[0089] Please also read Figure 4 , Figure 4 This is a risk array diagram provided by an embodiment of the present application.

[0090] like Figure 4 As shown, the risk array includes risk score, collagen characteristics, location, maximum tumor size, thyroid capsule invasion, total score, and risk of metastasis of papillary thyroid carcinoma.

[0091] Understandably, Figure 4The risk array diagram shown is a visualization tool used in the embodiments of the present application to display and evaluate the risk of papillary thyroid cancer metastasis in patients with papillary thyroid cancer. The diagram presents different clinical and pathological features in the form of an array to quickly assess the patient's risk situation. Specifically, the risk array diagram includes the following key components: Risk score: This score is obtained by comprehensively evaluating the patient's different clinical and pathological characteristics to quantify the patient's risk of papillary thyroid cancer metastasis. The risk score directly affects the possibility of papillary thyroid cancer metastasis, thus providing a reference for clinical decision-making.

[0092] Collagen features: This refers to the collagen features associated with the tumor microenvironment of papillary thyroid carcinoma extracted by multiphoton imaging technology. These features may include the cross-linking density, distribution pattern, cross-linking space, etc. of collagen, which can reflect the changes in collagen in the tumor microenvironment and the invasiveness of the tumor. The changes in collagen are closely related to the metastasis of the tumor, so it is one of the important indicators for predicting the metastasis of papillary thyroid carcinoma.

[0093] Location: Location refers to the location of the tumor and its papillary thyroid cancer metastasis, which may involve different areas of the thyroid gland or adjacent lymph nodes. Tumors in different parts may show different metastasis risks, and understanding the specific location of the tumor is very important for accurately assessing the risk of metastasis.

[0094] Maximum tumor size: The maximum tumor size refers to the maximum diameter of the tumor in the slice image. Generally speaking, the larger the tumor size, the higher the risk of papillary thyroid cancer metastasis. This parameter helps to assess the extent of tumor growth and its possible invasiveness.

[0095] Thyroid capsule invasion: Thyroid capsule invasion refers to whether the tumor breaks through the thyroid capsule and extends to the surrounding tissues. Capsule invasion usually means that the tumor is more malignant and the possibility of metastasis of papillary thyroid carcinoma is increased.

[0096] Total score: The total score is a score calculated by comprehensively calculating the above parameters. This score is usually combined with the influence of each parameter in a weighted manner to finally obtain a score that reflects the patient's overall risk, helping doctors make more accurate diagnosis and treatment decisions.

[0097] Risk of papillary thyroid cancer metastasis: This is the final prediction result, indicating the patient's risk of papillary thyroid cancer metastasis. Based on the analysis of the total score and various parameters, the risk of papillary thyroid cancer metastasis will be quantified as a specific ratio.

[0098] Specifically, based on the nomogram, a comprehensive papillary thyroid carcinoma metastasis risk score was output in combination with the patient's collagen signature score and clinical risk factors.

[0099] Understandably, this risk score provides clinicians with a quantitative reference to help them more accurately determine whether a patient is at risk of papillary thyroid cancer metastasis and provide a basis for subsequent treatment decisions.

[0100] Figure 5 This is a schematic diagram of a papillary thyroid carcinoma metastasis prediction system module provided in one embodiment of the present application. Figure 5 The papillary thyroid carcinoma metastasis prediction system 10 shown includes at least the following parts: a first image acquisition module 11 , a second image acquisition module 12 , a feature extraction module 13 , a model construction module 14 , and a risk information output module 15 .

[0101] In the embodiment of the present application, the first image acquisition module 11 is used to acquire a stained collagen image of the first thyroid papillary carcinoma section stained by hematoxylin and eosin staining. Figures 1 to 4 And the corresponding description will not be repeated here.

[0102] In the embodiment of the present application, the second image acquisition module 12 is used to acquire a multiphoton collagen image of the second thyroid papillary carcinoma slice by a multiphoton imaging method. Figures 1 to 4 And the corresponding description will not be repeated here.

[0103] In the embodiment of the present application, the feature extraction module 13 is used to set an observation area on the multiphoton collagen image according to the stained collagen image, and extract collagen features in the observation area. Figures 1 to 4 And the corresponding description will not be repeated here.

[0104] In the embodiment of the present application, the model building module 14 is used to build a collagen feature scoring model based on the LASSO algorithm and collagen features. Figures 1 to 4 And the corresponding description will not be repeated here.

[0105] In the embodiment of the present application, the risk information output module 15 is used to obtain the collagen feature score according to the collagen feature score model. The risk information output module 15 is also used to obtain clinical risk factors; and output the thyroid papillary carcinoma metastasis risk information according to the clinical risk factors and the collagen feature score. For specific acquisition methods, please refer to Figures 1 to 4 And the corresponding description will not be repeated here.

[0106] Figure 6 This is an electronic device 20 provided in an embodiment of the present application. Figure 6 As shown, the electronic device 20 includes at least the following parts: a processor 21 and a memory 22 .

[0107] In the embodiment of the present application, the memory 22 is used to store instructions executable by the processor 21, and the processor 21 is configured to implement the following when executing the instructions: Figure 1 The method for predicting lymph node metastasis in papillary thyroid carcinoma is shown.

[0108] In an embodiment of the present application, a computer-readable storage medium includes instructions, and the instructions instruct a device to execute the method of the first aspect. For example, the instructions instruct the device to execute Figure 1 The method for predicting lymph node metastasis of papillary thyroid carcinoma is shown in steps S100 to S800.

[0109] The program operating in the electronic device 20 involved in one embodiment of the present application may be a program (a program that enables a computer to function) that controls a central processing unit (CPU) and the like to realize the functions of the above-mentioned embodiment involved in one scheme of the present invention. Then, the information processed by these devices is temporarily stored in a random access memory (RAM) during its processing, and then stored in various ROMs such as a read-only memory (Flash ROM) and a hard disk drive (HDD), and is read, corrected, and written by the CPU as needed.

[0110] It should be noted that part of the electronic device 20 of the above embodiment may be realized by a computer. In this case, a program for realizing the control function may be recorded in a computer-readable recording medium, and the program recorded in the recording medium may be read into a computer and executed.

[0111] It should be noted that the "computer" mentioned here refers to a computer built into the electronic device 20, and uses a computer including hardware such as an OS and peripheral devices. In addition, "computer-readable recording medium" refers to removable media such as floppy disks, magneto-optical disks, ROMs, CD-ROMs, and storage devices such as hard disks built into the computer.

[0112] Furthermore, "computer-readable recording media" may include: media that dynamically store programs for a short period of time, such as communication lines when sending programs via networks such as the Internet or communication lines such as telephone lines; and media that store programs for a fixed period of time, such as volatile memories inside computers that serve as servers or clients in this case. In addition, the above-mentioned program may be a program for realizing a part of the above-mentioned functions, or a program that can realize the above-mentioned functions by combining with a program already recorded in a computer.

[0113] In addition, the electronic device 20 in the above-mentioned embodiment can also be implemented as a collection (device group) composed of multiple devices. Each device constituting the device group can have a part or all of the functions or functional blocks of the electronic device 20 in the above-mentioned embodiment. As a device group, it is sufficient to have all the functions or functional blocks of the electronic device 20.

[0114] It is understandable that the method, system 10 and electronic device 20 for predicting lymph node metastasis of papillary thyroid carcinoma provided in the embodiment of the present application can accurately and quantitatively extract the collagen features in the microenvironment of papillary thyroid carcinoma by combining multiphoton imaging technology with collagen feature analysis, and provide a new technical means for predicting the metastasis of papillary thyroid carcinoma. Compared with traditional imaging methods, the present invention can overcome the limitations of detection technologies such as ultrasound, CT, and MRI in the early diagnosis of papillary thyroid carcinoma metastasis, especially in the case where the lymph nodes are not obvious or difficult to palpate, and has higher sensitivity and specificity. In addition, by constructing a collagen feature scoring model based on LASSO regression, the present invention realizes an accurate prediction of the risk of papillary thyroid carcinoma metastasis, and can further improve the accuracy of prediction in combination with clinical risk factors. The collagen feature score finally output can not only provide clinicians with auxiliary decision-making basis and optimize treatment plans, but also help patients achieve personalized treatment and improve prognosis. Therefore, the implementation scheme of the present application has a high clinical application value and can significantly improve the level of early diagnosis and prognosis evaluation of papillary thyroid carcinoma.

[0115] Those skilled in the art should recognize that the above embodiments are only used to illustrate the present application and are not intended to be limiting of the present application. As long as they are within the spirit and scope of the present application, appropriate changes and modifications to the above embodiments are within the scope of protection claimed in the present application.

Claims

1. A method for predicting lymph node metastasis of papillary thyroid carcinoma, characterized in that: The method comprises: Acquire the stained collagen image of the first papillary thyroid carcinoma section after staining with hematoxylin and eosin; Multiphoton collagen images of the second thyroid papillary carcinoma section were obtained by multiphoton imaging; Setting an observation area on the multiphoton collagen image according to the stained collagen image; Extracting collagen features within the observation area; Constructing a collagen feature scoring model according to the LASSO algorithm and the collagen features; Obtaining a collagen characteristic score according to the collagen characteristic scoring model; Obtain clinical risk factors; The metastasis risk information of papillary thyroid carcinoma is output according to the clinical risk factors and the collagen characteristic score.

2. The method for predicting lymph node metastasis of papillary thyroid carcinoma according to claim 1, characterized in that: The collagen characteristics include morphological parameters, strength parameters and texture parameters; The morphological parameters include collagen area, collagen quantity, collagen length, collagen width, collagen stiffness, collagen cross-linking density, collagen cross-linking space, and collagen arrangement; The strength parameters include mean, variance, skewness, kurtosis, energy, and entropy.

3. The method for predicting lymph node metastasis of papillary thyroid carcinoma according to claim 2, characterized in that: The texture parameters are extracted by gray level co-occurrence matrix or Gabor filter; When feature extraction is performed through the gray level co-occurrence matrix, the texture parameters include contrast, correlation, uniformity, gray level co-occurrence matrix, five distances and four directions; When feature extraction is performed through the Gabor filter, the texture parameters include mean, variance, four scales and six directions.

4. The method for predicting lymph node metastasis of papillary thyroid carcinoma according to claim 3, characterized in that: The formula of the collagen characteristic scoring model is: 0.188×collagen cross-linking density-0.116×collagen distribution-0.077×collagen intensity mean-0.030×collagen grayscale co-occurrence matrix contrast in the second direction of 45 degrees+0.118×Gabor filter image convolution at the mean of the fifth direction of the second scale.

5. The method for predicting lymph node metastasis of papillary thyroid carcinoma according to claim 1, characterized in that: The multi-photon imaging method adopts an imaging magnification of 20 times, an imaging area size of 1 mm×1 mm, and the imaging area is 5 blocks.

6. The method for predicting lymph node metastasis of papillary thyroid carcinoma according to claim 1, characterized in that: The multiphoton imaging method includes extracting the spatial structural characteristics and arrangement regularity of collagen through second harmonic images.

7. The method for predicting lymph node metastasis of papillary thyroid carcinoma according to claim 1, characterized in that: The metastasis risk information of papillary thyroid carcinoma is presented in the form of a nomogram.

8. The method for predicting lymph node metastasis of papillary thyroid carcinoma according to claim 7, characterized in that: The method further comprises: The metastasis risk score of papillary thyroid carcinoma is output according to the nomogram.

9. A papillary thyroid carcinoma metastasis prediction system, applied to the papillary thyroid carcinoma lymph node metastasis prediction method according to any one of claims 1 to 8, characterized in that: The system comprises: A first image acquisition module is used to acquire a stained collagen image of the first thyroid papillary carcinoma section stained by hematoxylin and eosin staining; A second image acquisition module is used to acquire a multiphoton collagen image of the second thyroid papillary carcinoma slice by a multiphoton imaging method; A feature extraction module, used to set an observation area on the multiphoton collagen image according to the stained collagen image, and extract collagen features in the observation area; A model building module, used to build a collagen feature scoring model based on the LASSO algorithm and the collagen features; The risk information output module is used to obtain the collagen characteristic score according to the collagen characteristic scoring model; obtain clinical risk factors; and output the thyroid papillary carcinoma metastasis risk information according to the clinical risk factors and the collagen characteristic score.

10. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to implement the method for predicting lymph node metastasis of papillary thyroid carcinoma according to any one of claims 1 to 8 when executing the instructions.

Images