CN117530659B - Early cervical cancer transformation auxiliary diagnostic tool based on optical coherence tomography - Google Patents
Early cervical cancer transformation auxiliary diagnostic tool based on optical coherence tomography Download PDFInfo
- Publication number
- CN117530659B CN117530659B CN202311460619.6A CN202311460619A CN117530659B CN 117530659 B CN117530659 B CN 117530659B CN 202311460619 A CN202311460619 A CN 202311460619A CN 117530659 B CN117530659 B CN 117530659B
- Authority
- CN
- China
- Prior art keywords
- collagen fibers
- cervical
- tissue
- spatial
- cervical cancer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 208000006105 Uterine Cervical Neoplasms Diseases 0.000 title claims abstract description 28
- 206010008342 Cervix carcinoma Diseases 0.000 title claims abstract description 27
- 201000010881 cervical cancer Diseases 0.000 title claims abstract description 27
- 238000012014 optical coherence tomography Methods 0.000 title claims abstract description 14
- 230000009466 transformation Effects 0.000 title claims abstract description 8
- 102000008186 Collagen Human genes 0.000 claims abstract description 86
- 108010035532 Collagen Proteins 0.000 claims abstract description 86
- 229920001436 collagen Polymers 0.000 claims abstract description 86
- 239000000835 fiber Substances 0.000 claims abstract description 86
- 238000003745 diagnosis Methods 0.000 claims abstract description 16
- 201000010099 disease Diseases 0.000 claims abstract description 13
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 13
- 238000005516 engineering process Methods 0.000 claims abstract description 11
- 238000001914 filtration Methods 0.000 claims abstract description 10
- 239000013598 vector Substances 0.000 claims description 28
- 238000003384 imaging method Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 19
- 206010028980 Neoplasm Diseases 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 10
- 230000008859 change Effects 0.000 claims description 6
- 238000012706 support-vector machine Methods 0.000 claims description 6
- 208000026062 Tissue disease Diseases 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000005452 bending Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims 1
- 238000000034 method Methods 0.000 abstract description 14
- 238000010586 diagram Methods 0.000 abstract description 6
- 238000011503 in vivo imaging Methods 0.000 abstract description 5
- 230000000007 visual effect Effects 0.000 abstract description 5
- 230000000877 morphologic effect Effects 0.000 abstract description 2
- 238000004445 quantitative analysis Methods 0.000 abstract description 2
- 210000001519 tissue Anatomy 0.000 description 54
- 238000012512 characterization method Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 6
- 230000003902 lesion Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 239000000523 sample Substances 0.000 description 3
- 230000002708 enhancing effect Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000001727 in vivo Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012634 optical imaging Methods 0.000 description 2
- 230000001575 pathological effect Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000003325 tomography Methods 0.000 description 2
- 208000032170 Congenital Abnormalities Diseases 0.000 description 1
- 201000009273 Endometriosis Diseases 0.000 description 1
- 102000010834 Extracellular Matrix Proteins Human genes 0.000 description 1
- 108010037362 Extracellular Matrix Proteins Proteins 0.000 description 1
- 206010061218 Inflammation Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000090 biomarker Substances 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000024245 cell differentiation Effects 0.000 description 1
- 230000032823 cell division Effects 0.000 description 1
- 230000008619 cell matrix interaction Effects 0.000 description 1
- 230000012292 cell migration Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000034994 death Effects 0.000 description 1
- 231100000517 death Toxicity 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000002405 diagnostic procedure Methods 0.000 description 1
- 238000004141 dimensional analysis Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 210000002744 extracellular matrix Anatomy 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000000338 in vitro Methods 0.000 description 1
- 230000004054 inflammatory process Effects 0.000 description 1
- 230000036210 malignancy Effects 0.000 description 1
- 230000004660 morphological change Effects 0.000 description 1
- 238000011369 optimal treatment Methods 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 235000004252 protein component Nutrition 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000007634 remodeling Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000001356 surgical procedure Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0062—Arrangements for scanning
- A61B5/0066—Optical coherence imaging
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0033—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0082—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence adapted for particular medical purposes
- A61B5/0084—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence adapted for particular medical purposes for introduction into the body, e.g. by catheters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/64—Analysis of geometric attributes of convexity or concavity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Theoretical Computer Science (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Radiology & Medical Imaging (AREA)
- Pathology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses an early cervical cancer transformation auxiliary diagnostic tool based on an optical coherence tomography technology. The method can obtain a large visual field structure diagram of cervical tissues in a label-free, endoscopic and non-invasive in-vivo imaging mode, further obtain the structure of the collagen fibers in a filtering mode, finally carry out multidimensional quantitative analysis on the structural characteristics of the collagen fibers to obtain multi-component parameters such as directional variance, spatial curvature and spatial fluctuation of the collagen fibers, and carry out auxiliary diagnosis on the disease condition of the cervical tissues by utilizing the parameters; the spatial characteristics of the collagen fibers can be quantitatively characterized in multiple parameters, so that the morphological characteristics of the collagen fibers which are mutually complemented are provided, and the classification precision of early cervical cancer variable tissues and normal cervical tissues is higher than 95%.
Description
Technical Field
The invention belongs to the technical field of image analysis and disease auxiliary diagnosis of biological tissues, and particularly relates to an early cervical cancer transformation auxiliary diagnosis tool based on an optical coherence tomography technology.
Background
Cervical lesions generally refer to various lesions occurring in the cervical region, including inflammation, tumors, deformities, endometriosis, and the like. Among them, cervical neoplasms are the second most common malignancy among women worldwide, with new cases of worldwide deaths of about 30 tens of thousands each year being nearly 60 tens of thousands. Currently, diagnosis of cervical cancer relies mainly on pathological examination after clinical findings of lesions. Since early cervical cancer changes (e.g., stage T1 a) are not clinically apparent, most patients are already at a later stage when they are diagnosed, and thus miss the optimal treatment time for the drug or surgery. Therefore, the innovation of an early cervical cancer transformation auxiliary diagnostic tool is a key to the improvement and prognosis of cervical cancer transformation.
The quantitative characterization of the spatial characteristics of extracellular matrix (the main component is collagen fibers) provides a brand new idea for the auxiliary diagnosis of early cervical cancer. Collagen fibers are constantly undergoing self-remodeling under the influence of cell-matrix interactions such as cell migration, division and differentiation. Recent researches show that the spatial structure change in the self-remodelling process of the collagen fiber has strong correlation with the mechanical characteristics and biochemical states, so that the collagen fiber can be used for revealing the history of tissue lesions. Collagen fibers are the most abundant protein component of cervical tissue, which makes the structural features of collagen fibers one of the most potential and valuable biomarkers in cervical tissue disease examination.
However, the current imaging and quantification methods of collagen fiber structures and related early cervical cancer change diagnosis tools are very deficient, and have the following defects and technical problems:
1) The second harmonic imaging/multiphoton imaging is taken as a traditional imaging method of a collagen fiber structure, the visual field and imaging depth are limited to hundreds of micrometers, and screening and detection of a large visual field cannot be realized at one time;
2) The collagen fiber structure is generally obtained by performing cervical tissue section and then performing in-vitro imaging, and is difficult to obtain in a non-invasive and nondestructive in-vivo imaging mode;
3) Most of the current technologies are only limited to two-dimensional analysis of collagen fibers, and few technologies can quantify the three-dimensional structure of the collagen fibers;
4) These techniques only quantify the single structural features of the collagen fibers, such as density, direction and order of arrangement, and the like, and currently lack techniques for analyzing tissue disease states by synthesizing multiple structural parameters of the cervical tissue collagen fibers.
Thus, there is a need for an in-vivo, label-free, endoscopic, non-invasive, quantitative imaging and analysis tool that enables early detection and assisted diagnosis of major cervical tissue diseases.
Disclosure of Invention
In view of the above, the present invention provides an early cervical cancer transformation-aided diagnosis tool based on an optical coherence tomography technology. The method can obtain a large visual field structure diagram of cervical tissues in a label-free, endoscopic and non-invasive in-vivo imaging mode, further obtain the structure of the collagen fibers in a filtering mode, finally carry out multidimensional quantitative analysis on the structural characteristics of the collagen fibers to obtain multiple parameters such as directional variance, spatial curvature and spatial fluctuation of the collagen fibers, and carry out auxiliary diagnosis on the disease condition of the cervical tissues by utilizing the parameters. The invention is realized by the following technical scheme:
(1) Optical imaging; cervical tissue for which disease state information is known is imaged by the optical tomography system shown in fig. 1. Cervical tissue imaging can be divided into two types, in-vivo imaging which is realized by an endoscopic probe and ex-vivo imaging which is realized by a desk type measuring arm;
(2) Collagen fiber structure extraction: and (3) filtering and thresholding the cervical tissue imaging result obtained in the step (1) to obtain a preprocessed image. Preprocessing each frame of cervical tissue imaging results is processed frame by frame. The pretreatment process comprises the following steps: firstly, suppressing noise in an image through Gaussian filtering, then improving brightness non-uniformity in the image through homomorphic filtering, and finally enhancing contrast of collagen fibers through thresholding operation;
(3) Calculation of collagen fiber spatial orientation: calculating the collagen fiber spatial orientation based on the pre-processed image in step (2). The calculation process is as follows: firstly, carrying out binarization processing on a preprocessed image to obtain a binary template, and then taking the pixel point of the original image of the neighborhood of each effective pixel of the binary template according to a certain calculation window side length, wherein a vector is formed between a central pixel and any other pixels in the window. These vectors are multiplied by two weight factors: firstly, the reciprocal of the vector length; and secondly, the fluctuation of intensity along the vector direction. Finally, by the method ofAnd adding the vectors given with the weights to obtain the orientation of the center pixel. For three-dimensional images, two azimuth angles θ and β characterizing the vector direction and one polar angleCan be calculated by the following relation:
(4) Collagen fiber characterization parameter calculation: and (3) respectively calculating the direction variance, the space curvature and the space fluctuation of the collagen fibers according to the calculation result of the collagen fiber space orientation in the step (3). The direction variance represents the arrangement order degree of the collagen fibers, the value interval is 0-1, a value of 0 represents the arrangement of the collagen fibers which are completely parallel, and a value of 1 represents the arrangement of the collagen fibers which are completely unordered; the space curvature represents the bending degree of the collagen fiber, the value is 0-1, and the larger the value is, the more the collagen fiber is bent; the spatial fluctuation represents the unordered degree of the shallow collagen fibers and the change of the spatial curvature relative to the deep collagen fibers, the unordered degree and the spatial curvature of the shallow collagen fibers are smaller than those of the deep collagen fibers by positive values, and the negative values are opposite;
(5) Support vector machine decision plane normal vector and intercept calculation: and (3) inputting the multi-parameter calculation result obtained in the step (4) and the disease state label known by cervical tissues into a support vector machine model for training. Multi-parameter calculation result constitution vector Disease state y known to cervical tissue i The decision plane equation can be obtained through training:
thereby obtaining the normal vector of the decision planeAnd intercept b, training is specifically a conditional maximum problem:
wherein s.t. represents simultaneous satisfaction, N is the total number of samples;
through training of a large number of samples, cervical cancer becomes a normal vector of a diagnosis decision surface Intercept b= 1.728343.
(6) The tumor trend factors are constructed to assist in diagnosing whether the cervical tissue to be tested is at risk of canceration. Based on the normal vector of the decision surface obtained in the step (5)And intercept b, the tumor trend factor is calculated by:
wherein TPI represents a tumor trend factor,and a vector formed by multi-parameter calculation results of the tissue to be measured. When the TPI is greater than 0, the tissue to be tested is indicated to have tumor risk, and the larger the value is, the larger the risk is, and otherwise, the larger the value is, the no tumor risk is.
Based on the technical scheme, the invention has the following beneficial effects compared with the prior art:
1. compared with the traditional pathological examination diagnosis technology, the invention is an in-vivo, label-free, endoscopic, non-invasive, nondestructive and quantitative auxiliary diagnosis technology, does not need tissue sections, and has higher patient acceptance.
2. Compared with the traditional second harmonic/multiphoton imaging technology for obtaining the structure diagram of the collagen fiber, the invention can obtain the three-dimensional imaging result with the visual field larger than 2X 2mm and the depth larger than 1mm at one time, and the imaging time is smaller than 5s, so that the detection of cervical tissue can be realized clinically.
3. Compared with the traditional technology for analyzing the two-dimensional image, the method can carry out complete quantitative characterization on the three-dimensional structure information of the collagen fiber. The proposed spatial fluctuation parameters characterize the spatial morphological changes of the collagen fibers in the imaging depth direction and can only be obtained from three-dimensional images of tissues.
4. Compared with the traditional single-structure parameter analysis of the spatial characteristics of the collagen fibers, the method can quantitatively characterize the spatial characteristics of the collagen fibers in a multi-parameter way, so that the morphological characteristics of the collagen fibers which are mutually complemented are provided, and the classification precision of early cervical cancer variable tissues and normal cervical tissues is higher than 95%.
Drawings
FIG. 1 is a schematic diagram of the multi-parameter calculation and output results based on an optical coherence tomography. FIG. 1 (a) is a schematic diagram showing the result of filtering and contrast enhancement from an optical coherence tomography three-dimensional image; FIG. 1 (b) is a schematic view showing the results of calculating the three-dimensional spatial orientation of collagen fibers from the pre-processed image of FIG. (a); FIG. 1 (c) is a schematic representation of the results of multi-parameter quantitative characterization of the results of the spatial orientation of collagen fibers obtained from FIG. (b);
FIG. 2 is an image output result of the present invention for use in an early cervical cancer assisted diagnosis application. The first row is the preprocessed image of normal tissue and the calculated spatial orientation theta and spatial orientation of the collagen fiberThe second row is a map of the cancerous tissue, respectively, of the directional variance, spatial curvature, and spatial fluctuation. Wherein the mean of the direction variance, the spatial curvature and the spatial fluctuation is marked in the lower left corner of the graph. Comparison shows that cancerous tissueThe directional variance and spatial curvature of collagen fibers are smaller than those of normal tissue, while the spatial fluctuations are larger than those of normal tissue;
FIG. 3 is a classification output result of the present invention for use in an early cervical cancer-aiding diagnostic application. Fig. 3 (a-c) shows the results of classifying cancerous and normal samples of cervical tissue using the directional variance, spatial curvature and spatial fluctuation of collagen fibers, respectively; fig. 3 (d-f) shows the cervical tissue as a result of classifying the cancerous and normal samples using two parameters, respectively, with a classification accuracy significantly higher than that using a single parameter; FIG. 3 (g) shows the result of classifying the cancerous and normal samples using three parameters, with further improved classification accuracy;
FIG. 4 shows the classification and characterization results of two groups of samples, namely early cervical cancer (T1 a stage) tissue and normal cervical tissue, by tumor trend factor (TPI) according to the present invention. Fig. 4 (a) is a subject characteristic curve (ROC) generated using tumor trend factors. Fig. 4 (b) shows a comparison of TPI spectra of two representative samples, and it can be intuitively seen that the color of the cancerous tissue mass region is darker, indicating that the TPI value is large, and the corresponding high risk of cancerous.
Detailed Description
In order to more particularly describe the present invention, the following detailed description of the technical scheme of the present invention is provided with reference to the accompanying drawings and the specific embodiments.
FIG. 1 is a process flow diagram of a quantitative characterization algorithm for the spatial orientation of collagen fibers based on optical coherence tomography. The invention relates to an auxiliary diagnostic tool for major cervical tissue diseases based on an optical coherence tomography technology, which comprises the following steps:
(1) Optical imaging; cervical tissue for which disease state information is known is imaged by the optical tomography system shown in fig. 1. Cervical tissue imaging can be divided into two types, in-vivo imaging which is realized by an endoscopic probe and ex-vivo imaging which is realized by a desk type measuring arm;
(2) Collagen fiber structure extraction: and (3) filtering and thresholding the cervical tissue imaging result obtained in the step (1) to obtain a preprocessed image. Preprocessing each frame of cervical tissue imaging results is processed frame by frame. The pretreatment process comprises the following steps: firstly, suppressing noise in an image through Gaussian filtering, then improving brightness non-uniformity in the image through homomorphic filtering, and finally enhancing contrast of collagen fibers through thresholding operation;
(3) Calculation of collagen fiber spatial orientation: calculating the collagen fiber spatial orientation based on the pre-processed image in step (2). The calculation process is as follows: firstly, carrying out binarization processing on a preprocessed image to obtain a binary template, and then taking the pixel point of the original image of the neighborhood of each effective pixel of the binary template according to a certain calculation window side length, wherein a vector is formed between a central pixel and any other pixels in the window. These vectors are multiplied by two weight factors: firstly, the reciprocal of the vector length; and secondly, the fluctuation of intensity along the vector direction. Finally, the orientation of the center pixel is obtained by adding the vectors given with all weights. For three-dimensional images, two azimuth angles θ and β characterizing the vector direction and one polar angleCan be calculated by the following relation:
(4) Collagen fiber characterization parameter calculation: and (3) respectively calculating the direction variance, the space curvature and the space fluctuation of the collagen fibers according to the calculation result of the collagen fiber space orientation in the step (3). The direction variance represents the arrangement order degree of the collagen fibers, the value interval is 0-1, a value of 0 represents the arrangement of the collagen fibers which are completely parallel, and a value of 1 represents the arrangement of the collagen fibers which are completely unordered; the space curvature represents the bending degree of the collagen fiber, the value is 0-1, and the larger the value is, the more the collagen fiber is bent; the spatial fluctuation represents the unordered degree of the shallow collagen fibers and the change of the spatial curvature relative to the deep collagen fibers, the unordered degree and the spatial curvature of the shallow collagen fibers are smaller than those of the deep collagen fibers by positive values, and the negative values are opposite;
(5) Support vector machine decision plane normalCalculation of quantity and intercept: and (3) inputting the multi-parameter calculation result obtained in the step (4) and the disease state label known by cervical tissues into a support vector machine model for training. Multi-parameter calculation result constitution vector Disease state y known to cervical tissue i The decision plane equation can be obtained through training:
thereby obtaining the normal vector of the decision planeAnd intercept b, training is specifically a conditional maximum problem:
wherein s.t. represents simultaneous satisfaction, N is the total number of samples;
through training of a large number of samples, cervical cancer becomes a normal vector of a diagnosis decision surface Intercept b= 1.728343.
(6) The tumor trend factors are constructed to assist in diagnosing whether the cervical tissue to be tested is at risk of canceration. Based on the normal vector of the decision surface obtained in the step (5)And intercept b, the tumor trend factor is calculated by:
wherein TPI represents a tumor trend factor,and a vector formed by multi-parameter calculation results of the tissue to be measured. When the TPI is greater than 0, the greater the value is, the greater the risk of canceration exists in the tissue to be detected, otherwise, the greater the risk is.
In order to demonstrate the effectiveness and accuracy of the present invention, the present invention will be described below with reference to the results of the auxiliary diagnostic tests for cervical cancer, respectively.
FIG. 2 is a graph showing the results of the multiparameter characterization of collagen fibers used in cervical tissue (T1 a phase) for normal and early stage cancerous changes in accordance with the present invention. The first row is the preprocessed image of normal tissue and the calculated spatial orientation theta and spatial orientation of the collagen fiberDirectional variance, spatial curvature, and spatial fluctuation, the second row corresponds to various maps of cancerous tissue. Wherein the mean of the direction variance, the spatial curvature and the spatial fluctuation is marked in the lower left corner of the graph. The comparison shows that the direction variance and the space curvature of the collagen fibers of the cancerous tissue are smaller than those of the normal tissue, which means that the collagen fibers are more orderly and straighter; and a larger spatial fluctuation than normal tissue means a larger change in the spatial morphology of the superficial collagen fibers than in the deeper layers. Therefore, the multi-parameter characterization provided by the invention can be used for distinguishing the early cancerous cervical tissues from the normal tissues obviously.
FIG. 3 shows the results of the test classification of two sets of samples according to the invention for early cervical cancer tissue (stage T1 a) and normal cervical tissue, wherein the sample sizes of the two sets of samples are 66 and 36, respectively. FIGS. 3 (a-c) are results of classification of two groups of samples using directional variance, spatial curvature and spatial fluctuation of collagen fibers, respectively, with classification accuracies of 85.3%, 87.3% and 88.2%, respectively; FIG. 3 (d-f) shows the classification results of two groups of samples using two parameters of directional variance and spatial curvature, directional variance and spatial fluctuation, spatial curvature and spatial fluctuation of collagen fibers, respectively, with classification accuracies of 90.2%, 95.1% and 94.1%, respectively; FIG. 3 (g) shows the classification results of two groups of samples using three parameters of collagen fibers, and the classification accuracy reached 97.1%. It can be seen that the individual parameters provide spatially complementary information to the collagen fibers, so that the classification accuracy using two parameters is higher than the classification accuracy using a single parameter, while the classification accuracy using three parameters is higher than the classification accuracy using two parameters. After three characterization parameters are used, the classification accuracy of the early cervical cancer variable tissues and the normal cervical tissues is up to 97.1%, which shows that the invention has high sensitivity and high accuracy.
FIG. 4 shows the classification and characterization results of two groups of samples, namely early cervical cancer (T1 a stage) tissue and normal cervical tissue, by tumor trend factor (TPI) according to the present invention. Fig. 4 (a) is a subject characteristic curve (ROC) generated using tumor trend factors. After selecting the appropriate threshold, it can be seen that the specificity and sensitivity of the TPI proposed in the present invention for the two groups of samples was 94.4% and 97.0%, respectively. Whereas the area under the ROC curve (AUC) reached 0.998, indicating that the classification accuracy of TPI for both groups of samples was very high. Fig. 4 (b) shows a comparison of TPI spectra of two representative samples, and it can be intuitively seen that the color of the cancerous tissue mass region is darker, indicating that the TPI value is large, and the corresponding high risk of cancerous.
The embodiments described above are described in order to facilitate the understanding and application of the present invention to those skilled in the art, and it will be apparent to those skilled in the art that various modifications may be made to the embodiments described above and that the general principles described herein may be applied to other embodiments without the need for inventive faculty. Therefore, the present invention is not limited to the above-described embodiments, and those skilled in the art, based on the present disclosure, should make improvements and modifications within the scope of the present invention.
Claims (4)
1. An early cervical cancer transformation auxiliary diagnostic tool based on an optical coherence tomography technology is characterized in that the diagnostic tool extracts, calculates and characterizes the multi-element structural characteristics of cervical tissue collagen fibers according to the result of optical coherence tomography of cervical tissues, establishes an analysis model for auxiliary diagnosis of cervical tissue disease states, and comprises the following steps:
1) Obtaining a whole three-dimensional image of cervical tissue through an optical coherence tomography system by using the cervical tissue with known disease state information;
2) Filtering and contrast enhancement are carried out on the whole three-dimensional imaging data of cervical tissues, and a three-dimensional space structure image of the collagen fibers is extracted;
3) Respectively calculating the direction variance, the space curvature and the space fluctuation based on the three-dimensional space structure image of the collagen fiber to obtain a multi-parameter numerical result;
4) Constructing a support vector machine model according to the multi-parameter numerical value result obtained in the step 3) and the known disease state information in the step 1);
5) In the constructed support vector machine model, the multi-parameter numerical results calculated in the step 3) of the cervical tissue sample form a vector And the disease state information y known in step 1) i The decision plane equation can be obtained through training:
thereby obtaining the normal vector of the decision planeAnd intercept b, training is specifically a conditional maximum problem:
wherein s.t. represents simultaneous satisfaction, N is the total number of samples;
6) According to the normal vector of the decision plane obtained in step 5)And intercept b, calculating tumor trend factors of cervical tissues to be measured:
wherein TPI represents a tumor trend factor,and a vector formed by multi-parameter calculation results of the tissue to be measured.
2. The tool for assisting in diagnosing early cervical cancer based on optical coherence tomography according to claim 1, wherein in the step 3), the direction variance represents the arrangement order of collagen fibers, the value interval is 0-1,0 represents the arrangement of collagen fibers completely parallel, and 1 represents the arrangement of collagen fibers completely unordered; the space curvature represents the bending degree of the collagen fiber, the value is 0-1, and the larger the value is, the more the collagen fiber is bent; the spatial fluctuation represents the unordered degree of the shallow collagen fibers and the change of the spatial curvature relative to the deep collagen fibers, the positive value indicates that the unordered degree and the spatial curvature of the shallow collagen fibers are smaller than those of the deep collagen fibers, and the negative value is opposite.
3. The early cervical cancer transformation auxiliary diagnostic tool based on the optical coherence tomography according to claim 1 or 2, wherein the cervical cancer diagnostic decision surface normal vector trained in the step 5) isIntercept b= 1.728343.
4. The tool for assisting diagnosis of early cervical cancer according to claim 1 or 2, wherein in step 6), the TPI is indicative of the disease state of the cervical tissue to be tested, and when the TPI is greater than 0, it is indicative of the risk of tumor in the tissue to be tested, and a larger value indicates a larger risk, and vice versa.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311460619.6A CN117530659B (en) | 2023-11-06 | 2023-11-06 | Early cervical cancer transformation auxiliary diagnostic tool based on optical coherence tomography |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311460619.6A CN117530659B (en) | 2023-11-06 | 2023-11-06 | Early cervical cancer transformation auxiliary diagnostic tool based on optical coherence tomography |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117530659A CN117530659A (en) | 2024-02-09 |
| CN117530659B true CN117530659B (en) | 2024-04-09 |
Family
ID=89791020
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311460619.6A Active CN117530659B (en) | 2023-11-06 | 2023-11-06 | Early cervical cancer transformation auxiliary diagnostic tool based on optical coherence tomography |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117530659B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118436308A (en) * | 2024-04-23 | 2024-08-06 | 浙江大学 | Parkinson disease auxiliary inspection tool based on mitochondrial metabolism analysis |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107907395A (en) * | 2017-11-16 | 2018-04-13 | 南方医科大学南方医院 | Collagen tissue evaluation method by the cancer of early carcinoma of stomach Operated Specimens |
| CN110946552A (en) * | 2019-10-30 | 2020-04-03 | 南京航空航天大学 | Cervical cancer pre-lesion screening method combining spectrum and image |
| CN114596256A (en) * | 2022-01-20 | 2022-06-07 | 浙江大学 | Image analysis method for extracting space curvature characteristics of fibrous structure based on space orientation change in neighborhood |
| CN115602311A (en) * | 2022-07-20 | 2023-01-13 | 浙江大学（Cn） | Pancreatic cancer auxiliary inspection tool based on collagen fiber multivariate parameter analysis |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5953035B2 (en) * | 2011-11-28 | 2016-07-13 | 学校法人慶應義塾 | Pathological diagnosis support apparatus, pathological diagnosis support method, and pathological diagnosis support program |
-
2023
- 2023-11-06 CN CN202311460619.6A patent/CN117530659B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107907395A (en) * | 2017-11-16 | 2018-04-13 | 南方医科大学南方医院 | Collagen tissue evaluation method by the cancer of early carcinoma of stomach Operated Specimens |
| CN110946552A (en) * | 2019-10-30 | 2020-04-03 | 南京航空航天大学 | Cervical cancer pre-lesion screening method combining spectrum and image |
| CN114596256A (en) * | 2022-01-20 | 2022-06-07 | 浙江大学 | Image analysis method for extracting space curvature characteristics of fibrous structure based on space orientation change in neighborhood |
| CN115602311A (en) * | 2022-07-20 | 2023-01-13 | 浙江大学（Cn） | Pancreatic cancer auxiliary inspection tool based on collagen fiber multivariate parameter analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117530659A (en) | 2024-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rajan et al. | Brain tumor detection and segmentation by intensity adjustment | |
| Abuared et al. | Skin cancer classification model based on VGG 19 and transfer learning | |
| Hoshyar et al. | Review on automatic early skin cancer detection | |
| CN117530659B (en) | Early cervical cancer transformation auxiliary diagnostic tool based on optical coherence tomography | |
| CN110991536A (en) | Training method of early warning model of primary liver cancer | |
| CN112927187A (en) | Method for automatically identifying and positioning focal cortical dysplasia epileptic focus | |
| Cao et al. | A deep learning approach for detecting colorectal cancer via Raman spectra | |
| Florez et al. | Emergence of radiomics: novel methodology identifying imaging biomarkers of disease in diagnosis, response, and progression | |
| JP2021519920A (en) | A method for detecting cells with at least one abnormality in a cytological sample | |
| CN109191422A (en) | Cerebral arterial thrombosis early detection system and method based on conventional CT image | |
| CN116030032A (en) | Breast cancer analysis equipment, system and storage medium based on Raman spectrum data | |
| Glorindal et al. | A simplified approach for melanoma skin disease identification | |
| CN117541864A (en) | Deep learning-based pathological feature recognition and classification method for mammary gland pathological data | |
| Wei et al. | Multi-feature fusion for ultrasound breast image classification of benign and malignant | |
| Tian et al. | Combining hyperspectral imaging techniques with deep learning to aid in early pathological diagnosis of melanoma | |
| Li | Hyperspectral imaging technology used in tongue diagnosis | |
| CN115602311A (en) | Pancreatic cancer auxiliary inspection tool based on collagen fiber multivariate parameter analysis | |
| EP3791361B1 (en) | Tumoral mass detection system based on magnetic resonance imaging | |
| CN114529554A (en) | Intelligent auxiliary interpretation method for gastric cancer HER2 digital pathological section | |
| Ballerini et al. | Fractal analysis of microscopic images of breast tissue | |
| WO2024051015A1 (en) | Image feature extraction and classification method based on muscle ultrasound | |
| Bhagat et al. | Computational Intelligence approach to improve the Classification accuracy of Brain Tumor Detection | |
| TWI845365B (en) | Method for establishing model for identifying hormone receptor status, and method and system for identifying hormone receptor status | |
| Divya et al. | A hybrid CNN-FC approach for automatic grading of brain tumors from non-invasive MRIs | |
| CN118436308A (en) | Parkinson disease auxiliary inspection tool based on mitochondrial metabolism analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |