A lung cancer auxiliary examination tool based on morphological analysis of the fibrous structure of biological tissues

Abstract

This invention discloses an auxiliary examination tool for lung cancer based on the morphological analysis of fibrous structures in biological tissues. It utilizes multiphoton microscopy to analyze the morphological characteristics of fibrous structures in the extracellular matrix, establishing an auxiliary examination model for lung cancer. The model performs multivariate parameter quantitative characterization of elastic and collagen fiber images, and then comprehensively characterizes the similarity of the two types of fibers across multiple morphological features, generating a normalized similarity index atlas. This invention can rapidly extract the distribution information of elastic and collagen fibers in the extracellular matrix of lung tissue. By assessing the difference in the morphological similarity of these two fibrous structures between normal individuals and lung cancer patients, it determines the level of lung tumor evolution, presenting a novel examination method. This method has significant clinical translational potential, enabling more accurate analysis of the morphological characteristics of fibrous structures in biological tissues, and has greater application potential compared to traditional methods.

Description

Technical Field

[0001] This invention belongs to the field of image analysis and disease detection technology of biological tissues, specifically relating to an auxiliary examination tool for lung cancer based on the morphological analysis of the fibrous structure of biological tissues. Background Technology

[0002] Lung cancer has become the most common cancer worldwide in recent years and continues to be the leading cause of cancer death. There are two main histological subtypes of lung cancer: small cell lung cancer and non-small cell lung cancer (NSCLC). The latter accounts for approximately 85% of all cases and is less aggressive and grows more slowly. However, the prognosis for NSCLC patients remains poor; more than half of patients die within one year of diagnosis, and the 5-year survival rate is less than 20%. Therefore, accurate staging is crucial for clinical treatment and scientific research. Although biopsy specimen staining and section examination has been the gold standard for NSCLC staging, procedures involving tumor boundary confirmation such as biopsy and surgical resection still have drawbacks, including complex sampling, time-consuming procedures, and a high dependence on the physician's experience.

[0003] Currently, it is known that many carcinogenesis processes are often accompanied by tissue sclerosis, including lung cancer. At the microscopic level, the remodeling of the extracellular matrix during cancer evolution is gradually becoming an important marker for cancer diagnosis, staging, and grading. Morphological changes in some extracellular matrix fibrous structures can also reflect changes in macroscopic mechanical properties, thus indirectly reflecting the evolution of lung tumors. Based on this characteristic, multiphoton microscopy is used to perform morphological analysis of fibrous structures (mainly elastic and collagen fibers) in the extracellular matrix. The degree of similarity between these two types of fibers can indirectly infer the lesion level of lung cancer.

[0004] Multiphoton microscopy (MPM) has been widely used due to its advantages of deep imaging penetration and high optical three-dimensional resolution. MPM technology can perform two-photon excitation fluorescence (TPEF) microscopy and second harmonic generation (SHG) imaging of elastic and collagen fibers in the extracellular matrix, offering advantages such as label-free and rapid imaging. Analysis of the imaging results can yield information related to lung cancer evolution.

[0005] The existing technology has the following technical problems:

[0006] 1) Traditional histological diagnostic methods rely on direct biopsy samples stained with hematoxylin and eosin. Besides requiring invasive in vivo sampling, the time required for staining and slide preparation is lengthy, around a week; and often, tumor spread necessitates resampling after the initial slide diagnosis. Therefore, there is a demand for faster and more accurate auxiliary diagnostic methods to determine the boundaries of lung tumors. Finally, traditional sampling requires open-chest surgery to remove a large section of lung tissue for examination, resulting in a relatively small examination window.

[0007] 2) Emerging in vivo imaging technologies are now available that can provide low-invasive, rapid and accurate auxiliary diagnosis of early-stage in situ lung cancer. However, these technologies usually involve the use of radioactive contrast agents and complex procedures. In addition to some laparoscopic techniques, most of these procedures require invasive examinations and will still leave wounds in the body.

[0008] 3) Other emerging technologies include endobronchial ultrasound-guided fine needle aspiration (EBUS-FNA) and endoesophageal ultrasound-guided fine needle aspiration (EUS-FNA), which, in addition to their relatively limited sensitivity, require additional training time.

[0009] 4) Furthermore, after the biopsy, it may be necessary to remove the early in situ tumor. Although thoracoscopic-assisted surgery (VATS) and robot-assisted thoracic surgery (RATS) can achieve minimally invasive surgery, there are significant potential risks during the operation, and the cost of consumables for the newly introduced robots is high if traditional equipment is abandoned.

[0010] 5) At present, the uses of endoscopes are still relatively limited. They are often highly specialized equipment applicable to a single disease or organ, and are expensive. There is a lack of general auxiliary diagnostic methods and equipment applicable to similar diagnostic and treatment principles such as tumor sclerosis, in order to reduce costs. Summary of the Invention

[0011] This invention addresses the shortcomings of existing technologies by proposing an auxiliary diagnostic tool for lung cancer based on morphological analysis of the fibrous structure of biological tissues. The method uses an MPM system to image the extracellular matrix in lung tissue. Through multi-parameter characterization of the morphological features of elastic and collagen fibers in the multiphoton microscopy results, and based on pixel-level precision characterization results, quantifies the similarity between the two types of fibers. Furthermore, a classification model is established based on statistical indicators from a normalized similarity index map, enabling auxiliary diagnosis of lung cancer. This method avoids the drawbacks of traditional methods for diagnosing lung cancer, such as lack of quantification, time-consuming and labor-intensive methods, and reliance on experience, providing a new approach for in vivo diagnosis and related research of human lung tumors.

[0012] The technical solution adopted in this invention is:

[0013] This invention discloses an auxiliary examination tool for lung cancer based on the morphological analysis of the fibrous structure of biological tissues. Its key feature is that it establishes an auxiliary examination model for lung cancer by analyzing the morphological characteristics of the fibrous structure of biological tissues in the extracellular matrix using multiphoton microscopy. The examination tool includes the following steps:

[0014] 1) By using multiphoton microscopy, two-photon excitation fluorescence imaging and second harmonic generation imaging were performed on elastic fibers and collagen fibers in the extracellular matrix of human lung tissue cells to obtain multiphoton microscopic images, including elastic fiber images and collagen fiber images.

[0015] 2) Perform morphological multi-parameter quantitative characterization on the elastic fiber images and collagen fiber images obtained in step 1), respectively. The morphological multi-parameters include local density, spatial orientation, directional variance, curvature, and diameter, thereby extracting the morphological features of elastic fibers and collagen fibers at the pixel level. For the elastic fiber image, generate a series of original morphological parameter maps of elastic fibers, including: elastic fiber local density parameter map, elastic fiber spatial orientation parameter map, elastic fiber directional variance parameter map, elastic fiber curvature parameter map, and elastic fiber diameter parameter map. For the collagen fiber image, generate a series of original morphological parameter maps of collagen fibers, including: collagen fiber local density parameter map, collagen fiber spatial orientation parameter map, collagen fiber directional variance parameter map, and collagen fiber curvature parameter map.

[0016] 3) Perform image segmentation on the elastic fiber image and collagen fiber image obtained in step 1), and record the image segmentation results as m1 and m2 respectively. In the segmentation results, the pixel representing the elastic fiber or collagen fiber is 1 and the pixel representing the background is 0.

[0017] 4) For m1, perform cross-channel nearest neighbor fiber pixel retrieval, that is, find the nearest non-zero pixel in m2 that is the nearest non-zero pixel to all non-zero pixels in m1, and calculate the distance between the corresponding pixels to generate an elastic fiber nearest neighbor distance parameter map. At the same time, assign the value of the morphological parameter at the corresponding nearest non-zero pixel in m2 to the non-zero pixel in m1, thereby generating an elastic fiber nearest neighbor pixel morphological parameter map involving the parameters in step 2), including: elastic fiber nearest neighbor pixel local density parameter map, elastic fiber nearest neighbor pixel spatial orientation parameter map, elastic fiber nearest neighbor pixel orientation variance parameter map, and elastic fiber nearest neighbor pixel curvature parameter map;

[0018] 5) For m2, perform cross-channel nearest neighbor fiber pixel retrieval, that is, find the nearest non-zero pixel in m1 that is the nearest neighbor of all non-zero pixels in m2, and calculate the distance between the corresponding pixels to generate a collagen fiber nearest neighbor distance parameter map. At the same time, assign the value of the morphological parameter at the nearest non-zero pixel in m1 to the non-zero pixel in m2, thereby generating a collagen fiber nearest neighbor pixel morphological parameter map involving the parameters in step 2), including: collagen fiber nearest neighbor pixel local density parameter map, collagen fiber nearest neighbor pixel spatial orientation parameter map, collagen fiber nearest neighbor pixel orientation variance parameter map, and collagen fiber nearest neighbor pixel curvature parameter map;

[0019] 6) Based on the nearest neighbor pixel morphological parameter map of the elastic fiber obtained in step 4), perform a differential processing on it and the corresponding original morphological parameter map of the elastic fiber obtained in step 2) to obtain the nearest neighbor pixel morphological parameter difference map of the elastic fiber.

[0020] 7) Based on the collagen fiber nearest neighbor pixel morphological parameter map obtained in step 5), perform a differential processing on it and the corresponding original collagen fiber morphological parameter map obtained in step 2) to obtain the collagen fiber nearest neighbor pixel morphological parameter difference map.

[0021] 8) Perform pairwise fusion of the nearest neighbor distance parameter map of elastic fibers obtained in step 4) and the nearest neighbor distance parameter map of collagen fibers obtained in step 5) with weighted image segmentation results to obtain a nearest neighbor pixel distance parameter difference fusion map. Furthermore, perform pairwise fusion of the nearest neighbor pixel morphological parameter difference map of elastic fibers obtained in step 6) and the nearest neighbor pixel morphological parameter difference map of collagen fibers obtained in step 7) with weighted image segmentation results based on the corresponding morphological parameters to obtain a series of nearest neighbor pixel morphological parameter difference fusion maps. The nearest neighbor pixel distance parameter difference fusion map and the series of nearest neighbor pixel morphological parameter difference fusion maps are collectively referred to as the nearest neighbor pixel parameter difference fusion map.

[0022] 9) The nearest neighbor pixel parameter difference fusion map obtained in step 8) is normalized and scored to obtain a series of normalized nearest neighbor pixel parameter difference fusion score maps, including: normalized nearest neighbor distance parameter difference fusion score map, normalized nearest neighbor pixel local density parameter difference fusion score map, normalized nearest neighbor pixel spatial orientation parameter difference fusion score map, normalized nearest neighbor pixel orientation variance parameter difference fusion score map, and normalized nearest neighbor pixel curvature parameter difference fusion score map. In the map, the closer the score is to 0, the greater the difference in morphological features between elastic fibers and collagen fibers at that position. The closer the score is to 1, the higher the similarity of morphological features between elastic fibers and collagen fibers at that position. Then, the corresponding pixels of all normalized nearest neighbor pixel parameter difference fusion score maps are multiplied together to finally obtain a normalized similarity index map that reflects the morphological similarity between elastic fibers and collagen fibers.

[0023] 10) Calculate the full stack mean, full stack standard deviation, and full stack depth direction variation of the non-background pixel portion in the normalized similarity index map. The combination of these three values ​​indirectly reflects the changes in lung tissue stiffness, thereby enabling the extraction of information related to lung cancer evolution.

[0024] As a further improvement, in step 6) of the present invention, the morphological parameter map of the nearest neighbor pixels of the elastic fiber obtained in step 4) is differentiated from the corresponding original morphological parameter map of the elastic fiber obtained in step 2). Specifically, the local density parameter map of the nearest neighbor pixels of the elastic fiber is differentiated from the local density parameter map of the elastic fiber to obtain a local density parameter difference map of the nearest neighbor pixels of the elastic fiber; the spatial orientation parameter map of the nearest neighbor pixels of the elastic fiber is differentiated from the spatial orientation parameter map of the elastic fiber to obtain a spatial orientation parameter difference map of the nearest neighbor pixels of the elastic fiber; and the orientation variance parameter map of the nearest neighbor pixels of the elastic fiber is differentiated from the orientation variance parameter map of the elastic fiber. Differentiation processing yields the following: a differential map of the directional variance parameter of the nearest neighbor pixel of the elastic fiber, a differential map of the bending parameter of the nearest neighbor pixel of the elastic fiber, and a differential map of the bending parameter of the nearest neighbor pixel of the elastic fiber. The differential maps of the local density parameter, spatial orientation parameter, directional variance parameter, and bending parameter of the nearest neighbor pixel of the elastic fiber are collectively referred to as the morphological parameter difference map of the nearest neighbor pixel of the elastic fiber. For the morphological parameter of spatial orientation, when the difference result is between 90° and 180°, the supplementary angle is taken as the difference result.

[0025] As a further improvement, in step 7) of the present invention, based on the collagen fiber nearest neighbor pixel morphological parameter map obtained in step 5), it is differentiated from the corresponding original collagen fiber morphological parameter map obtained in step 2). Specifically, the difference is as follows: the collagen fiber nearest neighbor pixel local density parameter map is differentiated from the collagen fiber local density parameter map to obtain a collagen fiber nearest neighbor pixel local density parameter difference map; the collagen fiber nearest neighbor pixel spatial orientation parameter map is differentiated from the collagen fiber spatial orientation parameter map to obtain a collagen fiber nearest neighbor pixel spatial orientation parameter difference map; and the collagen fiber nearest neighbor pixel directional variance parameter map is differentiated from the collagen fiber directional variance parameter map. Differentiation processing yields the following: a differential map of the directional variance parameter of the nearest neighbor pixels of collagen fibers, a differential map of the directional variance parameter of the nearest neighbor pixels of collagen fibers, and a differential map of the directional variance parameter of the nearest neighbor pixels of collagen fibers. These differential maps are collectively referred to as the morphological parameter difference maps of the nearest neighbor pixels of collagen fibers. For the morphological parameter of spatial orientation, when the difference result is between 90° and 180°, the supplementary angle is taken as the difference result.

[0026] As a further improvement, in step 8), the present invention performs pairwise fusion based on the morphological parameter difference map of the nearest neighbor pixels of elastic fibers obtained in step 6) and the morphological parameter difference map of the nearest neighbor pixels of collagen fibers obtained in step 7), weighted by the image segmentation results. Specifically, the local density parameter difference map of the nearest neighbor pixels of elastic fibers is fused with the local density parameter difference map of the nearest neighbor pixels of collagen fibers to obtain a fused map of the local density parameter difference of the nearest neighbor pixels; the spatial orientation parameter difference map of the nearest neighbor pixels of elastic fibers is fused with the spatial orientation parameter difference map of the nearest neighbor pixels of collagen fibers to obtain a fused map of the spatial orientation parameter difference of the nearest neighbor pixels; the directional variance parameter difference map of the nearest neighbor pixels of elastic fibers is fused with the directional variance parameter difference map of the nearest neighbor pixels of collagen fibers to obtain a fused map of the directional variance parameter difference of the nearest neighbor pixels; and the curvature parameter difference map of the nearest neighbor pixels of elastic fibers is fused with the curvature parameter difference map of the nearest neighbor pixels of collagen fibers to obtain a fused map of the curvature parameter difference of the nearest neighbor pixels.

[0027] As a further improvement, in step 8), when the present invention performs pairwise fusion of the morphological parameter difference maps of the nearest neighbor pixels of elastic fibers obtained in step 6) and the morphological parameter difference maps of the nearest neighbor pixels of collagen fibers obtained in step 7) based on the corresponding morphological parameters and weighted by the image segmentation results, the specific weighting method of the image segmentation results is as follows:

[0028]

[0029] Among them, M FP It is a fusion map of the morphological parameter differences between the nearest neighbor pixels of elastic fibers and collagen fibers, M SP1 M is the nearest neighbor pixel morphological parameter difference map of any morphological parameter of an elastic fiber. SP2 It is the nearest neighbor pixel morphological parameter difference map of the morphological parameters corresponding to collagen fibers, and m1 and m2 are the image segmentation results of elastic fiber image and collagen fiber image, respectively.

[0030] As a further improvement, in step 9) of the present invention, the nearest neighbor pixel parameter difference fusion map obtained in step 8) is normalized and scored, and the nearest neighbor distance parameter difference fusion map M is then scored. FD Use the following expression:

[0031]

[0032] Among them, S D It is a normalized nearest neighbor distance parameter difference fusion score map, D uplimitIt is a reference value related to the nearest neighbor pixel distance parameter fusion map or the elastic fiber diameter, taken as the maximum diameter value in the elastic fiber diameter parameter map, where e is the base of the natural logarithm; for the nearest neighbor pixel local density parameter difference fusion map M FL Use the following expression:

[0033]

[0034] Among them, S L It is a normalized nearest neighbor pixel local density parameter difference fusion score map; for the nearest neighbor pixel spatial orientation parameter difference fusion map M FO Use the following expression:

[0035] S O =cos(M FO );

[0036] Among them, S O It is a normalized nearest neighbor pixel spatial orientation parameter difference fusion score map; for the nearest neighbor distance direction variance parameter difference fusion map or the nearest neighbor distance curvature parameter difference fusion map M FCD Use the following expression:

[0037]

[0038] Among them, S CD It is either a normalized nearest neighbor pixel orientation variance parameter difference fusion score map or a normalized nearest neighbor pixel curvature parameter difference fusion score map, where C is an empirical coefficient, usually taken as 4.

[0039] As a further improvement, in step 9) of the present invention, the method for calculating the full-stack depth direction variation value is as follows: first calculate the average value of all non-background pixels in each layer of the normalized similarity index map, and then calculate the standard deviation of these average values.

[0040] As a further improvement, in step 9) of the present invention, the extraction of lung cancer evolution-related information is further performed using the support vector machine method as an auxiliary inspection tool for lung tumor boundaries.

[0041] This invention utilizes multiphoton microscopy to achieve high-resolution, rapid in vivo examination. By performing quantitative morphological characterization of images of elastic and collagen fibers in the extracellular matrix, the similarity between these two types of fibers can indirectly reveal evolutionary information of lung tumors by reflecting changes in matrix stiffness. The established auxiliary examination tool for lung cancer assists physicians in examinations, further enhancing our understanding of the pathology of human lung cancer.

[0042] Due to the application of the above technical solution, the present invention has the following advantages compared with the prior art:

[0043] This invention discloses a lung cancer auxiliary examination tool based on MPM (Multi-parameter Modeling) morphological analysis of the fibrous structure of biological tissues. It analyzes the morphological status of the fibrous structure in the extracellular matrix plexiform layer of biological tissues, diagnosing abnormalities in macroscopic stiffness and microscopic fibrous structure remodeling caused by lung cancer. This invention can perform multi-parameter quantitative morphological characterization of elastic and collagen fiber images, and then comprehensively characterize the similarity of the two fibers in multiple morphological features, thereby generating a normalized similarity index map. Based on the statistical characteristics related to the normalized similarity index map, a lung cancer auxiliary examination model is established, which can automatically perform lung cancer examination. This invention can rapidly extract the distribution information of elastic and collagen fibers in the extracellular matrix of lung tissue, and determine the level of lung tumor evolution by analyzing the difference in the morphological similarity of the two fibrous structures between normal individuals and lung cancer patients, serving as a novel examination method proposed in this invention. This method has great clinical translational potential, enabling more accurate analysis of the morphological status of the fibrous structure of biological tissues, and has greater application potential compared to traditional methods.

[0044] 1. This invention differs from traditional histological diagnostic methods that rely on direct sampling and hematoxylin-eosin staining of biopsy samples. Instead, it utilizes multiphoton microscopy (MPM) imaging of human lung tissue for rapid, label-free, high-resolution imaging of the extracellular matrix. By detecting the microscopic remodeling of the extracellular matrix, it diagnoses lung cancer and enables discriminative analysis of the lung cancer evolution process. Due to the characteristics of MPM imaging, it can achieve non-destructive three-dimensional imaging of in-situ organs with a certain depth of penetration, thus obtaining more detailed information about tissue changes during lung cancer evolution.

[0045] 2. This invention utilizes MPM imaging results of two fibrous structures in the extracellular matrix to extract and analyze the morphological features of elastic fibers and collagen fibers, thereby revealing the tissue distribution of these two fibrous structures in terms of morphological characteristics. The similarity between the two types of fibers can indirectly reflect the degree of tumor sclerosis, thus further revealing the development status of lung cancer. Using MPM images for morphological analysis breaks away from the complex operations of previous techniques such as radioactive contrast agents in medical imaging, enabling traditional techniques such as thoracoscopic biopsy to achieve low-invasive, rapid, accurate, and simple auxiliary diagnosis of early-stage in situ lung cancer.

[0046] 3. This invention innovatively proposes a quantitative similarity index method, which indirectly reflects the evolution of lung cancer by reflecting changes in lung tissue stiffness. This parameter integrates multiple morphological features, including the nearest fiber distance and local fiber density related to the pixel intensity and distribution of fibrous structures, the spatial orientation of fibrous structures, and the directional variance (orderliness) and curvature related to the tissue and distribution characteristics of fibrous structures. By integrating these multiple morphological features, the obtained normalized parameter can intuitively reflect the overall similarity between two fibrous structures, has strong interpretability, and can provide doctors with histological information more conveniently and directly, without the need for complex training and extensive experience accumulation.

[0047] 4. This invention is based on three-dimensional label-free in vivo imaging analysis. Compared with single-image-based analysis and video guidance, the utilization rate of spatial information in the depth direction of MPM images is greatly improved, and the extraction of extracellular matrix microscopic information also provides updated information for diagnosis and treatment. This method provides a new approach to boundary identification in clinical in vivo sampling and in situ tumor resection surgery, effectively improving the accuracy of examination tools in probing lung tumor boundaries. Furthermore, this method can be adapted to traditional thoracoscopic instruments, demonstrating significant clinical translational value.

[0048] 5. This invention enables morphological distribution similarity analysis of two fibrous structures in MPM images and applies it to the microscopic detection of lung cancer evolution. Furthermore, this invention can not only perform similarity analysis on two or more fibrous structures at the microscopic level, but also, because it is based on the reflection of macroscopic stiffness by the microscopic remodeling of extracellular matrix fibrous structures, it can be extended to more cancer and disease detection applications that conform to this pattern, which is of great significance for revealing information about changes in fibrous structures in the microscopic biological environment. Attached Figure Description

[0049] Figure 1 A flowchart for constructing an auxiliary examination tool for lung cancer based on morphological similarity analysis of extracellular matrix biological tissue fibrous structures;

[0050] Figure 2 Comparison of imaging results and morphological multi-parameter quantitative characterization results of two fibrous structures in the extracellular matrix of isolated healthy human lung and NSCLC lung cancer cells under MPM.

[0051] Figure 3 Flowchart for quantitative characterization of morphological similarity index using simulated fibers;

[0052] Figure 4 This is a schematic diagram illustrating the application of a classification model based on similarity index analysis of in vivo MPM imaging results in mice. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0054] This invention relates to an auxiliary diagnostic tool for lung cancer based on the morphological analysis of the fibrous structure of extracellular matrix biological tissues. The specific steps include: Figure 1 As shown:

[0055] 1) Extracellular matrix multiphoton microscopy: Using a clinical MPM instrument, the extracellular matrix of suspected cancerous tissue in human lung was imaged. Two-photon excitation fluorescence imaging and second harmonic generation imaging were performed on elastic fibers and collagen fibers in the extracellular matrix of human lung tissue, each occupying one channel, thus obtaining multiphoton microscopic images, including images of elastic fibers and collagen fibers. Three-dimensional fusion reconstruction of elastic fiber and collagen fiber images from healthy human lungs and lungs of patients with NSCLC is shown below. Figure 2 As shown in (a).

[0056] 2) Perform morphological multi-parameter quantitative characterization on the elastic fiber images and collagen fiber images obtained in step 1), respectively. The morphological multi-parameters include local density, spatial orientation, directional variance, curvature, and diameter, thereby extracting the morphological features of elastic fibers and collagen fibers at the pixel-level precision. For the elastic fiber image, generate a series of original morphological parameter maps of elastic fibers, including: elastic fiber local density parameter map, elastic fiber spatial orientation parameter map, elastic fiber directional variance parameter map, elastic fiber curvature parameter map, and elastic fiber diameter parameter map. For the collagen fiber image, generate a series of original morphological parameter maps of collagen fibers, including: collagen fiber local density parameter map, collagen fiber spatial orientation parameter map, collagen fiber directional variance parameter map, and collagen fiber curvature parameter map, such as... Figure 2 As shown in (b). For a more intuitive representation, the fusion image of the morphological multi-parameter quantitative characterization results of dual-channel two-dimensional simulated fibers, consisting of elastic fiber and collagen fiber images, is shown in [image format]. Figure 3 As shown in (a), the color-coded legends are all of type JET except for the spatial orientation, which uses the hsv type.

[0057] 3) Perform image segmentation on the elastic fiber image and collagen fiber image obtained in step 1), and denote the resulting image segmentation as m1 and m2, respectively. In the segmentation results, pixels representing elastic fibers or collagen fibers are 1, and pixels representing the background are 0. Figure 3 (b) As shown in the leftmost column.

[0058] 4) For m1, perform cross-channel nearest neighbor fiber pixel retrieval, i.e., find the nearest non-zero pixel in m2 that is the closest to all non-zero pixels in m1, calculate the distance between the corresponding pixels, and generate an elastic fiber nearest neighbor distance parameter map. Simultaneously, assign the morphological parameter value of the corresponding nearest non-zero pixel in m2 to the non-zero pixel in m1, thereby generating an elastic fiber nearest neighbor pixel morphological parameter map involving the parameters in step 2), including: elastic fiber nearest neighbor pixel local density parameter map, elastic fiber nearest neighbor pixel spatial orientation parameter map, elastic fiber nearest neighbor pixel orientation variance parameter map, and elastic fiber nearest neighbor pixel curvature parameter map, thus obtaining the following... Figure 3 (b) The result shown to the right of the arrow in the upper part.

[0059] 5) For m2, perform cross-channel nearest neighbor fiber pixel retrieval, i.e., find the nearest non-zero pixel in m1 that is at a distance of all non-zero pixels therefrom, and calculate the distance between the corresponding pixels to generate a collagen fiber nearest neighbor distance parameter map. Simultaneously, assign the corresponding morphological parameter value at the nearest non-zero pixel in m1 to the non-zero pixel in m2, thereby generating a collagen fiber nearest neighbor pixel morphological parameter map involving the parameters in step 2), including: collagen fiber nearest neighbor pixel local density parameter map, collagen fiber nearest neighbor pixel spatial orientation parameter map, collagen fiber nearest neighbor pixel orientation variance parameter map, and collagen fiber nearest neighbor pixel curvature parameter map, thus obtaining the following... Figure 3 (b) The result shown to the right of the arrow in the lower half.

[0060] 6) Based on the nearest neighbor pixel morphological parameter map of the elastic fiber obtained in step 4), perform a differential processing on it and the corresponding original morphological parameter map of the elastic fiber obtained in step 2) to obtain the nearest neighbor pixel morphological parameter difference map of the elastic fiber. Specifically, a subtraction-then-absolute-value conversion operation is adopted. That is, the local density parameter map of the nearest neighbor pixel of the elastic fiber is differentiated from the local density parameter map of the elastic fiber to obtain the local density parameter difference map of the nearest neighbor pixel of the elastic fiber, the spatial orientation parameter map of the nearest neighbor pixel of the elastic fiber is differentiated from the spatial orientation parameter map of the elastic fiber to obtain the spatial orientation parameter difference map of the nearest neighbor pixel of the elastic fiber, the directional variance parameter map of the nearest neighbor pixel of the elastic fiber is differentiated from the directional variance parameter map of the elastic fiber to obtain the directional variance parameter difference map of the nearest neighbor pixel of the elastic fiber, and the curvature parameter map of the nearest neighbor pixel of the elastic fiber is differentiated from the curvature parameter map of the elastic fiber to obtain the curvature parameter difference map of the nearest neighbor pixel of the elastic fiber. The local density parameter difference map of the nearest neighbor pixel of the elastic fiber, the spatial orientation parameter difference map of the nearest neighbor pixel of the elastic fiber, the directional variance parameter difference map of the nearest neighbor pixel of the elastic fiber, and the curvature parameter difference map of the nearest neighbor pixel of the elastic fiber are collectively referred to as the morphological parameter difference map of the nearest neighbor pixel of the elastic fiber. Specifically, for the parameter of spatial orientation, when the difference result is between 90° and 180°, the supplementary angle is directly used as the difference result.

[0061] 7) Based on the collagen fiber nearest neighbor pixel morphological parameter map obtained in step 5), differentiate it from the corresponding original collagen fiber morphological parameter map obtained in step 2) to obtain a collagen fiber nearest neighbor pixel morphological parameter difference map. Specifically, a difference-then-absolute-value operation is adopted, that is, the local density parameter map of the collagen fiber nearest neighbor pixel is differentiated from the local density parameter map of the collagen fiber to obtain a collagen fiber nearest neighbor pixel local density parameter difference map, the spatial orientation parameter map of the collagen fiber nearest neighbor pixel is differentiated from the spatial orientation parameter map of the collagen fiber to obtain a collagen fiber nearest neighbor pixel spatial orientation parameter difference map, and the directional variance parameter map of the collagen fiber nearest neighbor pixel is differentiated from the directional variance parameter map of the collagen fiber to obtain a collagen fiber nearest neighbor pixel spatial orientation parameter difference map. The collagen fiber nearest neighbor pixel directional variance parameter difference map, the collagen fiber nearest neighbor pixel curvature parameter difference map, and the collagen fiber curvature parameter difference map are differentiated to obtain the collagen fiber nearest neighbor pixel curvature parameter difference map. The collagen fiber nearest neighbor pixel local density parameter difference map, collagen fiber nearest neighbor pixel spatial orientation parameter difference map, collagen fiber nearest neighbor pixel directional variance parameter difference map, and collagen fiber nearest neighbor pixel curvature parameter difference map obtained by the differentiation process are collectively referred to as the collagen fiber nearest neighbor pixel morphological parameter difference map. In particular, for the spatial orientation parameter, when the difference result is between 90° and 180°, its supplementary angle is directly used as the difference result.

[0062] 8) Perform pairwise fusion of the nearest neighbor distance parameter map of elastic fibers obtained in step 4) and the nearest neighbor distance parameter map of collagen fibers obtained in step 5), weighted by the image segmentation results, to obtain a nearest neighbor pixel distance parameter difference fusion map. Furthermore, perform pairwise fusion of the nearest neighbor pixel morphological parameter difference map of elastic fibers obtained in step 6) and the nearest neighbor pixel morphological parameter difference map of collagen fibers obtained in step 7), weighted by the image segmentation results, based on the corresponding morphological parameters, to obtain a series of nearest neighbor pixel morphological parameter difference fusion maps. The nearest neighbor pixel distance parameter difference fusion map and the series of nearest neighbor pixel morphological parameter difference fusion maps are collectively referred to as the nearest neighbor pixel parameter difference fusion map. When performing pairwise fusion based on the corresponding morphological parameters and weighted by the image segmentation results, the specific weighting method is as follows:

[0063]

[0064] Among them, M FP It is a fusion map of the nearest neighbor pixel parameter differences of the corresponding morphological parameters of elastic fibers and collagen fibers, M SP1 M is the nearest neighbor pixel morphological parameter difference map of any morphological parameter of an elastic fiber. SP2This is a nearest-neighbor pixel morphological parameter difference map of collagen fibers, where m1 and m2 are the image segmentation results of the elastic fiber image and the collagen fiber image, respectively. The nearest-neighbor pixel parameter difference fusion map is shown below. Figure 3 As shown in (c), all the color-coded legends are of type jet.

[0065] 9) The nearest neighbor pixel parametric difference fusion map obtained in step 8) is normalized and scored to obtain a series of normalized nearest neighbor pixel parametric difference fusion score maps. Specifically, different scoring strategies are used according to different morphological features, such as... Figure 3 As shown in (d), a series of normalized nearest neighbor pixel parameter difference fusion score maps include: normalized nearest neighbor distance parameter difference fusion score map, normalized nearest neighbor pixel local density parameter difference fusion score map, normalized nearest neighbor pixel spatial orientation parameter difference fusion score map, normalized nearest neighbor pixel orientation variance parameter difference fusion score map, and normalized nearest neighbor pixel curvature parameter difference fusion score map. Among these, pixels with scores closer to 0 indicate a greater difference in morphological features between elastic fibers and collagen fibers at that location, while pixels with scores closer to 1 indicate a high degree of similarity in morphological features between elastic fibers and collagen fibers at that location. Then, the corresponding pixels in all normalized nearest neighbor pixel parameter difference fusion score maps are multiplied together to obtain a normalized similarity index map reflecting the morphological similarity between elastic fibers and collagen fibers, as shown in [reference needed]. Figure 3 (e), where the color-coded legend is of jet type. During normalized scoring, the nearest neighbor distance parameter difference fusion map M is used. FD Use the following expression:

[0066]

[0067] Among them, S D It is a normalized nearest neighbor distance parameter difference fusion score map, D uplimit It is a reference value related to the nearest neighbor pixel distance parameter fusion map or the elastic fiber diameter, taken as the maximum diameter value in the elastic fiber diameter parameter map, where e is the base of the natural logarithm; for the nearest neighbor pixel local density parameter difference fusion map M FL Use the following expression:

[0068]

[0069] Among them, S L It is a normalized nearest neighbor pixel local density parameter difference fusion score map; for the nearest neighbor pixel spatial orientation parameter difference fusion map M FO Use the following expression:

[0070] S O=cos(M FO )

[0071] Among them, S O This is a normalized nearest neighbor pixel spatial orientation parameter difference fusion score map; for the nearest neighbor distance direction variance parameter difference fusion map or the nearest neighbor distance curvature parameter difference fusion map, the following expression is used:

[0072]

[0073] Among them, S CD It is either a normalized nearest neighbor pixel orientation variance parameter difference fusion score map or a normalized nearest neighbor pixel curvature parameter difference fusion score map, where C is an empirical coefficient, usually taken as 4.

[0074] 10) Calculate the full-stack mean, full-stack standard deviation, and full-stack depth-direction variability for the non-background pixel portion of the normalized similarity index map. The full-stack depth-direction variability is calculated by first calculating the mean of all non-background pixels in each layer of the normalized similarity index map, and then calculating the standard deviation of these means. The combination of these three values ​​indirectly reflects changes in lung tissue stiffness, thereby extracting information related to lung cancer evolution. Furthermore, a support vector machine method is used as an auxiliary tool for examining lung tumor boundaries. Figure 4 As shown in (a), the stiffness of lung cancer tissue is significantly higher than that of healthy tissue, and Figure 4 (b) Linear regression results show that the similarity index parameter can reflect the changes in lung tissue stiffness during cancer development. Based on this, the full-stack mean, full-stack standard deviation, and full-stack depth-direction variability of the non-background pixels in the normalized similarity index map were calculated. The full-stack depth-direction variability was calculated as follows: first, the mean value of all non-background pixels in each layer of the normalized similarity index map was calculated, and then the standard deviation of these means was calculated. Figure 4 As shown in (c), the three statistical values ​​of the similarity index have strong distinguishability among the three groups. Therefore, the combination of the three values ​​can indirectly reflect the changes in lung tissue stiffness, thereby enabling the extraction of information related to lung cancer evolution. Figure 4 As shown in (d), a scatter plot of three values ​​can basically distinguish between healthy, boundary, and cancerous tissues in mouse in vivo lung imaging results.

[0075] In training a multi-parameter lung cancer auxiliary examination model, mouse lung in vivo imaging results were morphologically characterized to construct a similarity index dataset, which was then used to train the model using a support vector machine algorithm. This dataset was divided into a training set and a test set using a leave-one-out method. The training set was used for model construction and training, while the test set was used to test the model's ability to detect lung cancer and to optimize the detection model. Figure 4 As shown in (e), the AUC value obtained by using the support vector machine algorithm for classification is better. Figure 4 (f) records that the classification model achieved high classification accuracy in both the original classification and cross-validation classification.

[0076] It is worth mentioning that this method uses in situ in vivo MPM images for morphological analysis of the extracellular matrix fibrous structure, which is a novel approach for the auxiliary diagnosis of lung cancer. We used a nanoindentation testing device to measure the stiffness of ex vivo decellularized lung tissue and correlated it with the morphological similarity index of the imaging results, demonstrating the feasibility of using the morphological similarity index to reflect tissue hardening and identify the evolutionary process of lung tumors. Figure 4 As shown in (a), the stiffness of isolated human lung cancer tissue is significantly higher than that of healthy tissue, which preliminarily demonstrates the effectiveness of using stiffness to distinguish lung cancer tissue from healthy lung tissue. Furthermore, Figure 4 (b) Linear regression results show that the mean and standard deviation statistical indicators derived from the similarity index parameter can linearly reflect the changes in lung tissue stiffness during cancer development, providing a basis for quantitatively identifying lung cancer tissue. Furthermore, as... Figure 4 As shown in (c), the combination of the full-stack mean, full-stack standard deviation and full-stack depth direction variation value of the similarity index of in vivo MPM images of mouse lung cancer has a high degree of distinguishability of different evolutionary regions of lung cancer, and can indirectly reflect the changes in the stiffness of lung tissue, thereby realizing the extraction of information such as boundaries during the evolution of lung cancer.

[0077] Based on this approach, this invention, through steps such as optical imaging, image segmentation, quantitative morphological characterization, and construction of a similarity index classification model, ultimately yields a lung cancer auxiliary examination tool based on the morphological analysis of the fibrous structure of biological tissues. This method enables rapid processing and analysis of in vivo imaging results in clinical practice, accurately analyzing the morphological condition of the fibrous structure of biological tissues, and achieving accurate classification of healthy, borderline, and cancerous samples, which is of great significance in the auxiliary examination of diseases.

[0078] The above description of the embodiments is provided to enable those skilled in the art to understand and apply the present invention. Those skilled in the art can readily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without creative effort. Therefore, the present invention is not limited to the above embodiments, and any improvements and modifications made to the present invention by those skilled in the art based on the disclosure thereof should be within the scope of protection of the present invention.

Claims

1. A method for auxiliary examination of lung cancer based on morphological analysis of the fibrous structure of biological tissue, characterized in that, A lung cancer auxiliary examination model was established by analyzing the morphological characteristics of fibrous structures in the extracellular matrix using multiphoton microscopy. The examination method includes the following steps: 1) By using multiphoton microscopy, two-photon excitation fluorescence imaging and second harmonic generation imaging were performed on elastic fibers and collagen fibers in the extracellular matrix of human lung tissue cells to obtain multiphoton microscopic images, including elastic fiber images and collagen fiber images. 2) Perform morphological multi-parameter quantitative characterization on the elastic fiber images and collagen fiber images obtained in step 1), respectively. The morphological multi-parameters include local density, spatial orientation, directional variance, curvature, and diameter, thereby extracting the morphological features of elastic fibers and collagen fibers at the pixel level. For the elastic fiber image, generate a series of original morphological parameter maps of elastic fibers, including: elastic fiber local density parameter map, elastic fiber spatial orientation parameter map, elastic fiber directional variance parameter map, elastic fiber curvature parameter map, and elastic fiber diameter parameter map. For the collagen fiber image, generate a series of original morphological parameter maps of collagen fibers, including: collagen fiber local density parameter map, collagen fiber spatial orientation parameter map, collagen fiber directional variance parameter map, and collagen fiber curvature parameter map. 3) Perform image segmentation on the elastic fiber image and collagen fiber image obtained in step 1), and denote the resulting image segmentation as follows: and In the segmentation results, pixels representing elastic fibers or collagen fibers are 1, and pixels representing the background are 0. 4) For Perform cross-channel nearest neighbor fiber pixel search, that is, find the fiber pixels that are within a distance of all non-zero pixels. The nearest non-zero pixel in the graph is identified, and the distance between corresponding pixels is calculated to generate a nearest neighbor distance parameter map for the elastic fiber. Assigning values ​​to non-zero pixels The values ​​of morphological parameters at the nearest non-zero pixel are used to generate a morphological parameter map of the nearest neighbor pixels of the elastic fiber involving the parameters in step 2), including: local density parameter map of the nearest neighbor pixels of the elastic fiber, spatial orientation parameter map of the nearest neighbor pixels of the elastic fiber, orientation variance parameter map of the nearest neighbor pixels of the elastic fiber, and curvature parameter map of the nearest neighbor pixels of the elastic fiber. 5) For Perform cross-channel nearest neighbor fiber pixel search, that is, find the fiber pixels that are within a distance of all non-zero pixels. The nearest non-zero pixels in the matrix are identified, and the distance between corresponding pixels is calculated to generate a nearest neighbor distance parameter map of collagen fibers. Assigning values ​​to non-zero pixels The values ​​of morphological parameters at the nearest non-zero pixel are used to generate a morphological parameter map of the nearest neighbor pixels of collagen fibers involving the parameters in step 2), including: local density parameter map of the nearest neighbor pixels of collagen fibers, spatial orientation parameter map of the nearest neighbor pixels of collagen fibers, orientation variance parameter map of the nearest neighbor pixels of collagen fibers, and curvature parameter map of the nearest neighbor pixels of collagen fibers. 6) Based on the nearest neighbor pixel morphological parameter map of the elastic fiber obtained in step 4), perform a differential processing on it and the corresponding original morphological parameter map of the elastic fiber obtained in step 2) to obtain the nearest neighbor pixel morphological parameter difference map of the elastic fiber. 7) Based on the collagen fiber nearest neighbor pixel morphological parameter map obtained in step 5), perform differential processing on it and the corresponding original collagen fiber morphological parameter map obtained in step 2) to obtain the collagen fiber nearest neighbor pixel morphological parameter difference map. 8) Perform pairwise fusion of the nearest neighbor distance parameter map of elastic fibers obtained in step 4) and the nearest neighbor distance parameter map of collagen fibers obtained in step 5) with weighted image segmentation results to obtain a nearest neighbor pixel distance parameter difference fusion map. Furthermore, perform pairwise fusion of the nearest neighbor pixel morphological parameter difference map of elastic fibers obtained in step 6) and the nearest neighbor pixel morphological parameter difference map of collagen fibers obtained in step 7) with weighted image segmentation results based on the corresponding morphological parameters to obtain a series of nearest neighbor pixel morphological parameter difference fusion maps. The nearest neighbor pixel distance parameter difference fusion map and the series of nearest neighbor pixel morphological parameter difference fusion maps are collectively referred to as the nearest neighbor pixel parameter difference fusion map. 9) The nearest neighbor pixel parameter difference fusion map obtained in step 8) is normalized and scored to obtain a series of normalized nearest neighbor pixel parameter difference fusion score maps, including: normalized nearest neighbor distance parameter difference fusion score map, normalized nearest neighbor pixel local density parameter difference fusion score map, normalized nearest neighbor pixel spatial orientation parameter difference fusion score map, normalized nearest neighbor pixel orientation variance parameter difference fusion score map, and normalized nearest neighbor pixel curvature parameter difference fusion score map. In the map, the closer the score is to 0, the greater the difference in morphological features between elastic fibers and collagen fibers at that position. The closer the score is to 1, the higher the similarity of morphological features between elastic fibers and collagen fibers at that position. Then, the corresponding pixels of all normalized nearest neighbor pixel parameter difference fusion score maps are multiplied together to finally obtain a normalized similarity index map that reflects the morphological similarity between elastic fibers and collagen fibers. 10) Calculate the full stack mean, full stack standard deviation, and full stack depth direction variation of the non-background pixel portion in the normalized similarity index map. The combination of these three values ​​indirectly reflects the changes in lung tissue stiffness, thereby enabling the extraction of information related to lung cancer evolution.

2. The method for auxiliary examination of lung cancer based on morphological analysis of fibrous structures of biological tissues according to claim 1, characterized in that: In step 6), based on the morphological parameter map of the nearest neighbor pixels of the elastic fiber obtained in step 4), it is differentiated from the corresponding original morphological parameter map of the elastic fiber obtained in step 2). Specifically, the local density parameter map of the nearest neighbor pixels of the elastic fiber is differentiated from the local density parameter map of the elastic fiber to obtain a local density parameter difference map of the nearest neighbor pixels of the elastic fiber; the spatial orientation parameter map of the nearest neighbor pixels of the elastic fiber is differentiated from the spatial orientation parameter map of the elastic fiber to obtain a spatial orientation parameter difference map of the nearest neighbor pixels of the elastic fiber; and the directional variance parameter map of the nearest neighbor pixels of the elastic fiber is differentiated from the directional variance parameter map of the elastic fiber to obtain a local density parameter difference map of the nearest neighbor pixels of the elastic fiber. The elastic fiber nearest neighbor pixel directional variance parameter difference map, the elastic fiber nearest neighbor pixel curvature parameter difference map, and the elastic fiber curvature parameter difference map are differentiated to obtain the elastic fiber nearest neighbor pixel curvature parameter difference map. The elastic fiber nearest neighbor pixel local density parameter difference map, elastic fiber nearest neighbor pixel spatial orientation parameter difference map, elastic fiber nearest neighbor pixel directional variance parameter difference map, and elastic fiber nearest neighbor pixel curvature parameter difference map obtained by the differentiation process are collectively referred to as the elastic fiber nearest neighbor pixel morphological parameter difference map. Among them, for the spatial orientation morphological parameter, when the difference result is between 90° and 180°, its supplementary angle is taken as the difference result.

3. The method for auxiliary examination of lung cancer based on morphological analysis of fibrous structures of biological tissues according to claim 1, characterized in that: In step 7), based on the collagen fiber nearest neighbor pixel morphological parameter map obtained in step 5), it is differentiated from the corresponding original collagen fiber morphological parameter map obtained in step 2). Specifically, the difference is as follows: the local density parameter map of the collagen fiber nearest neighbor pixel is differentiated from the collagen fiber local density parameter map to obtain a collagen fiber nearest neighbor pixel local density parameter difference map; the spatial orientation parameter map of the collagen fiber nearest neighbor pixel is differentiated from the collagen fiber spatial orientation parameter map to obtain a collagen fiber nearest neighbor pixel spatial orientation parameter difference map; and the orientation variance parameter map of the collagen fiber nearest neighbor pixel is differentiated from the collagen fiber orientation variance parameter map to obtain a collagen fiber spatial orientation parameter difference map. The collagen fiber nearest neighbor pixel directional variance parameter difference map, the collagen fiber nearest neighbor pixel curvature parameter difference map, and the collagen fiber curvature parameter difference map are differentiated to obtain the collagen fiber nearest neighbor pixel curvature parameter difference map. The collagen fiber nearest neighbor pixel local density parameter difference map, collagen fiber nearest neighbor pixel spatial orientation parameter difference map, collagen fiber nearest neighbor pixel directional variance parameter difference map, and collagen fiber nearest neighbor pixel curvature parameter difference map obtained by the differentiation process are collectively referred to as the collagen fiber nearest neighbor pixel morphological parameter difference map. Among them, for the morphological parameter of spatial orientation, when the difference result is between 90° and 180°, its supplementary angle is taken as the difference result.

4. The method for auxiliary examination of lung cancer based on morphological analysis of fibrous structures of biological tissues according to claim 1, characterized in that: In step 8), the morphological parameter difference maps of the nearest neighbor pixels of elastic fibers obtained in step 6) and the morphological parameter difference maps of the nearest neighbor pixels of collagen fibers obtained in step 7) are fused pairwise based on the corresponding morphological parameters and weighted by the image segmentation results. Specifically, the local density parameter difference maps of the nearest neighbor pixels of elastic fibers and collagen fibers are fused to obtain a fused map of the local density parameter difference of the nearest neighbor pixels; the spatial orientation parameter difference maps of the nearest neighbor pixels of elastic fibers and collagen fibers are fused to obtain a fused map of the spatial orientation parameter difference of the nearest neighbor pixels; the directional variance parameter difference maps of the nearest neighbor pixels of elastic fibers and collagen fibers are fused to obtain a fused map of the directional variance parameter difference of the nearest neighbor pixels; and the curvature parameter difference maps of the nearest neighbor pixels of elastic fibers and collagen fibers are fused to obtain a fused map of the curvature parameter difference of the nearest neighbor pixels.

5. The method for auxiliary examination of lung cancer based on morphological analysis of fibrous structures of biological tissues according to claim 1, characterized in that: In step 8), when performing pairwise fusion of the morphological parameter difference maps of the nearest neighbor pixels of elastic fibers obtained in step 6) and collagen fibers obtained in step 7) based on the corresponding morphological parameters and weighted by the image segmentation results, the specific weighting method based on the image segmentation results is as follows: ； in, It is a fusion map of the morphological parameter differences between the nearest neighbor pixels of elastic fibers and collagen fibers. It is a morphological parameter difference map of the nearest neighbor pixels for any morphological parameter of an elastic fiber. It is a morphological parameter difference map of the nearest neighbor pixels corresponding to the morphological parameters of collagen fibers. and The images are segmentation results for elastic fiber images and collagen fiber images, respectively.

6. The method for auxiliary examination of lung cancer based on morphological analysis of fibrous structures of biological tissues according to claim 1, characterized in that: In step 9), the nearest neighbor pixel parameter difference fusion map obtained in step 8) is normalized and scored, and the nearest neighbor distance parameter difference fusion map is then used as the basis for the score. Use the following expression: ； in, It is a fusion score map of the difference in the normalized nearest neighbor distance parameter. It is a reference value related to the nearest neighbor pixel distance parameter fusion map or the elastic fiber diameter, and is taken as the maximum diameter value in the elastic fiber diameter parameter map. It is the base of the natural logarithm; it is used to fuse maps based on the differences in local density parameters of nearest neighbor pixels. Use the following expression: ； in, It is a normalized nearest neighbor pixel local density parameter difference fusion score map; and a nearest neighbor pixel spatial orientation parameter difference fusion map. Use the following expression: ； in, It is a normalized nearest neighbor pixel spatial orientation parameter difference fusion score map; and a nearest neighbor distance direction variance parameter difference fusion map or a nearest neighbor distance curvature parameter difference fusion map. Use the following expression: ； in, It is either a normalized nearest neighbor pixel orientation variance parameter difference fusion score map or a normalized nearest neighbor pixel curvature parameter difference fusion score map. The empirical coefficient is set to 4.

7. The method for auxiliary examination of lung cancer based on morphological analysis of fibrous structures of biological tissues according to claim 1, characterized in that: In step 9), the method for calculating the full-stack depth direction variation value is as follows: first, calculate the average value of all non-background pixels in each layer of the normalized similarity index map, and then calculate the standard deviation of these average values.

8. The method for auxiliary examination of lung cancer based on morphological analysis of fibrous structures of biological tissues according to claim 1, characterized in that: In step 9), the information related to lung cancer evolution is extracted and further supported by a support vector machine method as an auxiliary method for examining the boundaries of lung tumors.

Images