WO2023168981A1 - 一种结直肠癌胶原评价和预后预测装置和存储介质 - Google Patents

一种结直肠癌胶原评价和预后预测装置和存储介质 Download PDF

Info

Publication number
WO2023168981A1
WO2023168981A1 PCT/CN2022/131939 CN2022131939W WO2023168981A1 WO 2023168981 A1 WO2023168981 A1 WO 2023168981A1 CN 2022131939 W CN2022131939 W CN 2022131939W WO 2023168981 A1 WO2023168981 A1 WO 2023168981A1
Authority
WO
WIPO (PCT)
Prior art keywords
collagen
area
feature
value
coef
Prior art date
Application number
PCT/CN2022/131939
Other languages
English (en)
French (fr)
Inventor
严俊
董淑敏
严博韬
陈伟生
董小玉
刘秀敏
赵书涵
程家鑫
董彦凤
蒋伟
陈德鑫
李国新
Original Assignee
南方医科大学南方医院
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 南方医科大学南方医院 filed Critical 南方医科大学南方医院
Priority to US18/286,475 priority Critical patent/US11990241B2/en
Publication of WO2023168981A1 publication Critical patent/WO2023168981A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • 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
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
    • G06V10/449—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30028—Colon; Small intestine
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30096—Tumor; Lesion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/03—Recognition of patterns in medical or anatomical images

Definitions

  • the present application relates to the field of medical imaging technology, and in particular to a device and storage medium for colorectal cancer collagen evaluation and prognosis prediction.
  • Colorectal cancer is one of the most common malignant tumors worldwide, and its incidence has increased significantly in recent years.
  • the TNM staging system is currently the most widely used collagen evaluation and prognosis prediction method in clinical practice. It evaluates tumors by evaluating the primary tumor (Tumor), lymph node involvement (Nodes) and distant metastasis (Metastasis) assessment and staging to predict prognosis.
  • the TNM staging system still has certain limitations. On the one hand, patients with the same stage have very different clinical outcomes after receiving similar treatment regimens. This difference is even more prominent in stage II and III colorectal cancer. On the other hand, the TNM staging device focuses on the characteristics of the tumor itself, but lacks the assessment of the "soil" of the tumor microenvironment. Therefore, prognostic strategies based on the tumor microenvironment need to be further supplemented.
  • the purpose of this application is to provide a device and storage medium for colorectal cancer collagen evaluation and prognosis prediction, which can improve the above problems.
  • the present application provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, the computer program includes program instructions, and the program instructions when executed by a processor cause The processor performs the following steps:
  • the scoring parameters include the average value of each collagen characteristic parameter in each area, the inter-regional difference value and the degree of variation value. ; Wherein, the difference value between regions is the difference between the average values of the collagen characteristic parameters between any two regions, and the degree of variation value is the ratio of the difference values between two of the regions;
  • Prognostic prediction is performed by combining the patient's TNM staging information and the collagen signature score.
  • this application discloses a computer-readable storage medium.
  • multi-photon imaging can be performed on pathological tissue slices containing tumor environment information, and further pathological division of the tumor microenvironment can be performed through image processing; from tumor tissue, invasion edge, and normal tissue respectively Extract various collagen characteristic parameter values from three regions, such as morphology, energy, and texture characteristic parameters; calculate the difference and degree of variation between regions based on the regional characteristic parameters; establish a collagen characteristic scoring model, and calculate the collagen characteristics through the collagen characteristic parameters included in the model score; this collagen characteristic score can well complement the TNM staging system's lack of assessment of the "soil" of the tumor microenvironment, and provide a more effective reference for predicting the prognosis of colorectal cancer.
  • image processing is performed on the multi-photon imaging image, and the multi-photon imaging image is divided into a tumor center area, an invasion edge area and a normal tissue area, including:
  • Extract the boundary line between the tumor and the normal tissue determine the area containing the tumor as the initial tumor center area, and determine the area containing the normal tissue as the initial normal tissue area;
  • the area in the initial tumor center area except the invasion edge area is determined to be the tumor center area, and the area in the initial normal tissue area except the invasion edge area is determined to be the normal tissue area.
  • each collagen characteristic parameter value in each area is extracted, and each scoring parameter is calculated based on the collagen characteristic parameter value, including:
  • Randomly select three normal tissue sub-regions ROI7, ROI8, and ROI9 within the normal tissue region calculate the average value of each collagen characteristic parameter value of the three normal tissue sub-regions as the normal tissue region.
  • the average value of the collagen characteristic parameters described in each item is the average value of the collagen characteristic parameters described in each item.
  • the step of extracting various collagen characteristic parameter values in each area and calculating each scoring parameter respectively based on the collagen characteristic parameter values also includes:
  • the ratio of the first inter-regional difference value to the third inter-regional difference value is used as the inter-regional variation degree value.
  • the difference between regions and the degree of variation between regions is conducive to multi-dimensional analysis of the collagen characteristics of the tumor microenvironment.
  • the inter-regional variation value when the inter-regional variation value is less than or equal to 1, it means that the tumor microenvironment changes in the same trend from the tumor center area to the invasion edge area to the normal tissue area, and the inter-regional variation degree is low; when the inter-regional variation value is greater than 1, This shows that the tumor microenvironment changes in a non-uniform trend from the tumor center area to the invasion edge area to the normal tissue area, and the degree of variation between regions is high.
  • the scoring parameters corresponding to the collagen features of the selected target item are selected to calculate the collagen feature score, including:
  • the collagen feature scoring model is constructed through the LASSO regression algorithm as follows:
  • Cf score represents the collagen feature score
  • collagen_feature i represents the scoring parameter corresponding to the collagen feature of the i-th target item
  • coef i represents the calculation coefficient corresponding to the parameter collagen_feature i
  • k represents the collagen feature of the target item. quantity.
  • the present application provides a colorectal cancer collagen evaluation and prognosis prediction device, including a processor, an input device, an output device and a memory, and the processor, input device, output device and memory are connected to each other, wherein, the The memory may be a computer-readable storage medium as described in the first aspect of the invention, the processor being configured to invoke the program instructions.
  • This application discloses a computer-readable storage medium.
  • multi-photon imaging can be performed on pathological tissue slices containing tumor environment information, and further pathological division of the tumor microenvironment can be performed through image processing; from tumor tissue, invasion edge, and normal tissue respectively Extract various collagen characteristic parameter values from three regions, such as morphology, energy, and texture characteristic parameters; calculate the difference and degree of variation between regions based on regional characteristics; establish a collagen characteristic scoring model, and calculate the collagen characteristic score through the collagen characteristic parameters included in the model ;
  • This collagen characteristic score can well complement the TNM staging system's lack of assessment of the "soil" of the tumor microenvironment, and provide a more effective reference for predicting the prognosis of colorectal cancer.
  • Calculating the difference between regions and the degree of variation between regions is beneficial to multidimensional analysis of the collagen characteristics of the tumor microenvironment.
  • the inter-regional variation value when the inter-regional variation value is less than or equal to 1, it means that the tumor microenvironment changes in the same trend from the tumor center area to the invasion edge area to the normal tissue area, and the inter-regional variation degree is low; when the inter-regional variation value is greater than 1, This shows that the tumor microenvironment changes in a non-uniform trend from the tumor center area to the invasion edge area to the normal tissue area, and the degree of variation between regions is high.
  • Figure 1 is a schematic flowchart of program instructions stored in a computer-readable storage medium provided by this application;
  • Figure 2 is a schematic diagram of the division of multi-photon imaging images in this application.
  • Figure 3 is a schematic diagram of the calculation principle of the average value of collagen characteristic parameters in this application.
  • Figure 4 is a survival nomogram established by this application in conjunction with the TNM staging project and the collagen characteristic scoring project;
  • Figure 5 is a schematic diagram of the ROC curve of the collagen evaluation and prognosis prediction method for colorectal cancer and the TNM staging system collagen evaluation and prognosis prediction method disclosed in this application;
  • Figure 6 is a schematic structural diagram of a colorectal cancer collagen evaluation and prognosis prediction device provided by this application.
  • the TNM staging system is the most common tumor staging system in the world.
  • the TNM staging system has become a standard method for clinicians and medical scientists to stage malignant tumors. It evaluates and stages tumors by evaluating the primary tumor (Tumor), lymph node involvement (Nodes), and distant metastasis (Metastasis). , and then predict the prognosis.
  • the TNM staging system still has certain limitations. On the one hand, patients with the same stage have very different clinical outcomes after receiving similar treatment regimens. This difference is even more prominent in stage II and III colorectal cancer.
  • tumors are not just a collection of single tumor cells, but also include a variety of stromal cells, immune cells, extracellular matrix, microvessels, etc., which form a tumor microenvironment. Malignant tumor cells are equivalent to "seeds" and the tumor microenvironment Equivalent to "soil", the TNM staging system focuses on the characteristics of the tumor itself, but lacks the assessment of the "soil" of the tumor microenvironment. Therefore, prognostic strategies based on the tumor microenvironment need to be further supplemented.
  • the present application provides a computer-readable storage medium.
  • the computer-readable storage medium stores a computer program.
  • the computer program includes program instructions. When executed by a processor, the program instructions causes the processor to perform the following steps:
  • the target imaging area on the above-mentioned pathological tissue sections needs to be the area at the tumor junction.
  • the method for obtaining the above-mentioned pathological tissue sections includes: preparing colorectal cancer surgical specimens with routine pathological wax blocks, selecting the wax blocks at the tumor junction, and taking 2 consecutive slices with a thickness of 4 ⁇ m. The first slice is stained with HE, and the second slice is stained with HE. Zhang prepared the above pathological tissue sections for multiphoton imaging.
  • the method for determining the above target imaging area includes: scanning the first slice 20 times with the digital slice scanning system PRECICE 600, and selecting the tumor junction area with a deeper infiltration depth as the above-mentioned junction area; The corresponding area is determined as the target imaging area, and the size of the target imaging area is generally selected as 5mm*5mm.
  • the above multi-photon imaging images can be obtained by scanning the target imaging area 20 times using the Zeiss LSM 880 with Airyscan multi-photon confocal laser scanning system.
  • Multiphoton imaging is a multiphoton microscopy imaging technology based on nonlinear optics and femtosecond laser. It uses the inherent autofluorescence of living tissue cells and the second harmonic generated by collagen to perform imaging. It can be used directly without staining. Obtain collagen histological characteristics of tissues and cells. Matrix collagen is an important component of the structure of the tumor microenvironment. It participates in multiple processes of tumor occurrence and development and affects prognosis. High collagen content in the tumor microenvironment is associated with poor prognosis.
  • ZEN software Zen 2.3 lite, Carl Zeiss Microscopy GmbH, 2011 completes the above-mentioned Center of Tumor (CT) area, normal tissue (Normal tissue, N) area and invasive margin (Invasive Margin, IM) ) division of areas.
  • the scoring parameters include the average value of each collagen characteristic parameter in each region, the difference value between regions, and the degree of variation value.
  • the difference value between regions is the difference between the average values of collagen characteristic parameters between any two regions, and the degree of variation value is the ratio of the difference values between two regions.
  • the collagen morphological characteristic parameters include: collagen area (Collagen area), collagen fiber number (Collagen fiber number), collagen fiber length (Collagen fiber length), collagen fiber width (Collagen fiber width), collagen fiber straightness (Collagen fiber straightness), collagen crosslink density, collagen crosslink space and collagen orientation.
  • the characteristic parameters of collagen strength are analyzed by histograms.
  • the characteristic parameters of collagen strength include: Histogram mean, Histogram Variance, Histogram Skewness, and Histogram Kurtosis. ), histogram energy (Histogram Energy) and histogram entropy (Histogram Entropy).
  • the collagen structure characteristic parameters include 80 gray-level co-occurrence matrix (GLCM) texture parameters and 48 Gabor wavelet transform parameters.
  • the above 80 are calculated by calculating the contrast, correlation, energy and uniformity of 5 pixels at positions 1, 2, 3, 4 and 5 and 4 different directions of 0°, 45°, 90° and 135° GLCM. Then through the mean and variance of 4 different scales and 5 different directions, 48 Gabor wavelet transform parameters are obtained.
  • GLCM gray-level co-occurrence matrix
  • the selection of the Region of Interest can be completed by ZEN software (Zen 2.3lite, Carl Zeiss Microscopy GmbH, 2011). Three ROIs were randomly selected in the tumor center area, invasion edge area, and normal tissue area, resulting in a total of 9 ROIs.
  • the ROI size was 844 pixels ⁇ 844 pixels (approximately 700 ⁇ m ⁇ 700 ⁇ m).
  • the average value of the 142 collagen characteristic parameter values extracted from the ROI in the three areas of tumor center, invasion edge and normal tissue was calculated as the average value of each collagen characteristic parameter in this area.
  • the inter-regional differences and degree of variation of collagen parameters were calculated based on regional characteristics.
  • FIG. 5 shows a schematic diagram of the receiver operating characteristic curve (ROC) of the computer-readable storage medium and the TNM staging system collagen evaluation and prognosis prediction method disclosed in this application.
  • the ROC curve is also called the sensitivity curve.
  • the reason for this name is that each point on the curve reflects the same sensitivity. They are all responses to the same signal stimulus, but under several different criteria. Just the results obtained.
  • the receiver operating characteristic curve is a coordinate graph with the probability of false alarm as the horizontal axis and the probability of hit as the vertical axis. It is also a curve drawn by the different results obtained by the subject under specific stimulation conditions due to the use of different judgment criteria.
  • AUC Absolute Under Curve
  • the value of this area will not be greater than 1.
  • the value range of AUC is between 0.5 and 1. The closer the AUC is to 1.0, the higher the authenticity of the detection method; when it is equal to 0.5, the authenticity is the lowest and has no application value.
  • the AUC of the patient's 5-year disease-free survival (DFS) calculated by the TNM staging system collagen evaluation and prognosis prediction method is only 0.674 (shown by the TNM line in Figure 5); through this The AUC of the patient's 5-year disease-free survival rate derived from the computer-readable storage medium disclosed in the application is significantly improved to 0.742 (shown by the TNM+CS TME line in Figure 5). It can be seen from this that the computer-readable storage medium disclosed in this application has higher prediction performance.
  • this application discloses a computer-readable storage medium.
  • This method performs multiphoton imaging on pathological tissue sections containing tumor environment information, and further performs pathological division of the tumor microenvironment through image processing; extracts various collagen characteristic parameter values from three areas: tumor tissue, invasion edge, and normal tissue. , such as morphological, energy, and texture feature parameters; calculate the difference and degree of variation between regions based on regional features; establish a collagen feature score model, and calculate the collagen feature score through the collagen feature parameters included in the model; this collagen feature score can be a good supplement
  • the TNM staging system lacks an assessment of the "soil" of the tumor microenvironment and can provide a more effective reference for predicting the prognosis of colorectal cancer.
  • the step "Perform image processing on the multi-photon imaging image, and divide the multi-photon imaging image into the tumor center area, the invasion edge area and the normal tissue area” can be performed by ZEN software (Zen 2.3 lite, Carl Zeiss Microscopy GmbH, 2011) completed, this step specifically includes:
  • Extract the boundary line between the tumor and the normal tissue determine the area containing the tumor as the initial tumor center area, and determine the area containing the normal tissue as the initial normal tissue area.
  • the boundary line between the tumor and the normal tissue can be extracted through the image algorithm, as shown by the dotted line in Figure 2.
  • the left side of the dotted line is the above-mentioned initial tumor center area, and the right side of the dotted line is the above-mentioned initial normal tissue area.
  • a first boundary line is formed, and after translating the boundary line a second distance toward the direction of the initial normal tissue area, a second boundary line is formed.
  • the boundary line can be translated 500 ⁇ m toward the direction of the initial tumor center area to form the first boundary line L1 (as shown in Figure 2), and the boundary line can be translated 500 ⁇ m toward the direction of the initial normal tissue area to form the second boundary line L2 ( as shown in picture 2).
  • CT center of tumor
  • N normal tissue
  • step 120 specifically includes:
  • the difference between regions and the degree of variation between regions is conducive to multi-dimensional analysis of the collagen characteristics of the tumor microenvironment.
  • the inter-regional variation value when the inter-regional variation value is less than or equal to 1, it means that the tumor microenvironment changes in the same trend from the tumor center area to the invasion edge area to the normal tissue area, and the inter-regional variation degree is low; when the inter-regional variation value is greater than 1, This shows that the tumor microenvironment changes in a non-uniform trend from the tumor center area to the invasion edge area to the normal tissue area, and the degree of variation between regions is high.
  • step 130 specifically includes:
  • the collagen feature scoring model is constructed through the LASSO regression algorithm as follows:
  • Cf score represents the collagen feature score
  • collagen_feature i represents the scoring parameter corresponding to the collagen feature of the i-th target item
  • coef i represents the calculation coefficient corresponding to the parameter collagen_feature i
  • k represents the collagen feature of the target item. quantity.
  • collagen_feature 1 represents the collagen area value of the invasion edge area, and coef 1 is 0.181;
  • collagen_feature 2 represents the collagen quantity value in the invasion edge area, and coef 2 is 0.081;
  • collagen_feature 3 represents the collagen length value of the invasion edge area, and coef 3 is 0.303;
  • collagen_feature 4 represents the collagen arrangement in the normal tissue area, and coef 4 is -0.515;
  • collagen_feature 5 represents the difference between the first regions of collagen cross-linking density, coef 5 is -0.126;
  • collagen_feature 6 represents the correlation in the gray co-occurrence matrix of the tumor center area with a pixel distance of 3 in the 0° direction, and coef 6 is 0.018;
  • collagen_feature 7 represents the average Gabor filter image convolution with a scale of 3 and a direction of 60° in the central area of the tumor, and coef 7 is 0.477;
  • collagen_feature 8 represents the average convolution value of the Gabor filter image with a scale of 1 and a direction of 60° in the invasion edge area, and coef 8 is -0.119;
  • collagen_feature 9 represents the average convolution value of the Gabor filter image with a scale of 3 and a direction of 60° in the invasion edge area, and coef 9 is 0.232;
  • collagen_feature 10 represents the third inter-regional difference of the Gabor filter image convolution variance with scale 3 and direction 60°, coef 10 is 0.016;
  • collagen_feature 11 represents the difference between the first regions of contrast in the 45° direction with a pixel distance of 3 in the gray co-occurrence matrix, and coef 11 is 0.141;
  • collagen_feature 12 represents the first inter-regional difference of the Gabor filter image convolution variance with scale 2 and direction 90°, coef 12 is 0.048;
  • collagen_feature 13 represents the difference between the first regions of the Gabor filter image convolution variance with a scale of 4 and a direction of 90°, and coef 13 is 0.12;
  • collagen_feature 14 represents the inter-regional variation value of the correlation in the 90° direction with a pixel distance of 1 in the gray co-occurrence matrix, and coef 14 is 0.24;
  • collagen_feature 15 represents the inter-regional variation value of the correlation in the 135° direction with a pixel distance of 2 in the gray co-occurrence matrix, coef 15 is 0.145;
  • Collagen_feature 16 represents the inter-regional variation value of the correlation with a pixel distance of 4 in the 0° direction in the gray co-occurrence matrix, and coef 16 is 0.303.
  • doctors can combine the patient's TNM staging information and collagen characteristic score to predict prognosis. Specifically include:
  • the survival nomogram established in step 141 is shown in Figure 4.
  • the survival nomogram includes parallel establishment of sub-score line segments, T-stage line segments, N-stage line segments, and collagen characteristic score line segments; wherein each score value between 0 and 10 points is evenly arranged on the sub-score line segment;
  • the T1/2 points, T3 points, T4a points and T4b points in the T phase line segment respectively correspond to the 0 minute point, 4.8 minute point, 6.6 minute point and 8.3 minute point of the itemized fractional line segment;
  • the N stage segment line segment The N0 point, N1/N1c point, N2a point and N2b point respectively correspond to the 0 point, 2.5 point, 3.4 point and 5.3 point of the sub-score line segment; they are evenly arranged on the collagen characteristic score line segment.
  • the survival nomogram also includes parallel establishment of a total score line segment, a 1-year disease-free survival rate line segment, a 3-year disease-free survival rate line segment, and a 5-year disease-free survival rate line segment; wherein, 0 points are evenly distributed on the total score line segment.
  • 0.9 points, 0.8 points, 0.7 points and 0.6 points in the 1-year disease-free survival rate line segment correspond to 12.8 points, 15.6 points and 16.8 points respectively in the total score line segment points and 18 points
  • 0.9 points, 0.8 points, 0.7 points and 0.6 points in the 3-year disease-free survival rate line segment correspond to 10.4 points, 12.4 points, 14.4 points and 15.2 points respectively in the total score line segment points
  • 0.9 points, 0.8 points, 0.7 points and 0.6 points in the 5-year disease-free survival rate line segment correspond to 9.6 points, 12 points, 13.2 points and 14.4 points respectively in the total score line segment.
  • the calculated collagen profile score of patient A is 2.305
  • the T stage is T4a
  • the N stage is N1.
  • the patient's TNM stage it was found through the nomogram evaluation shown in Figure 4 that the patient is more likely to have recurrence, metastasis or tumor-related death after surgery (especially after 3 years), and it is recommended to strengthen follow-up. density, increase the content and frequency of postoperative review, and choose a more active adjuvant treatment strategy.
  • the patient's B collagen characteristic score is 0.820
  • his T stage is T3
  • his N stage is N1.
  • the patient's TNM stage it was found through the nomogram evaluation shown in Figure 4 that the patient had recurrence and metastasis after surgery. Or tumor-related death is less likely, it is recommended to conduct routine follow-up and postoperative review, and choose conventional adjuvant treatment strategies.
  • the present application provides a device for colorectal cancer collagen evaluation and prognosis prediction.
  • the colorectal cancer collagen evaluation and prognosis prediction device includes one or more processors 601; one or more input devices 602, one or more output devices 603 and a memory 604.
  • the above-mentioned processor 601, input device 602, output device 603 and memory 604 are connected through a bus.
  • the memory 604 may be the computer-readable storage medium of any one of the first aspects, and the processor 601 is configured to invoke the program instructions.
  • the so-called processor 601 can be a central processing unit (Central Processing Unit, CPU), and the processor can also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP) , Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.
  • the input device 602 may be used to acquire a multi-photon imaging image of the target imaging area on the pathological tissue section.
  • the output device 603 may be used to output the prognosis prediction obtained by any method of the first aspect to the target terminal for display.
  • the memory 604 may include read-only memory and random access memory and provides instructions and data to the processor 601 .
  • a portion of memory 604 may also include non-volatile random access memory.
  • memory 604 may also store device type information.
  • the processor 601, input device 602, and output device 603 described in the embodiment of the present invention can execute the implementation described in any method in the first aspect, and can also execute the implementation of the terminal device described in the embodiment of the invention. The implementation method will not be described again here.
  • the above-mentioned computer-readable storage medium may be an internal storage unit of the terminal device of any of the foregoing embodiments, such as a hard disk or memory of the terminal device.
  • the above-mentioned computer-readable storage medium can also be an external storage device of the above-mentioned terminal device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), or a secure digital (Secure Digital, SD) card equipped on the above-mentioned terminal device. Flash Card, etc.
  • the computer-readable storage medium may also include both an internal storage unit of the terminal device and an external storage device.
  • the above computer-readable storage medium is used to store the above computer program and other programs and data required by the above terminal device.
  • the above-mentioned computer-readable storage media can also be used to temporarily store data that has been output or is to be output.
  • the disclosed terminal equipment and methods can be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the above units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components may be combined or may be integrated into another device, or some features can be ignored, or not implemented.
  • the coupling or direct coupling or communication connection between each other shown or discussed may be an indirect coupling or communication connection through some interfaces, devices or units, or may be an electrical, mechanical or other form of connection.
  • the units described above as separate components may or may not be physically separated.
  • the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.
  • each functional unit in various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.
  • the above integrated units can be implemented in the form of hardware or software functional units.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.
  • the technical solution of the present invention is essentially or contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to cause a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method in various embodiments of the present invention.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program code. .
  • first,” “second,” “the first,” or “the second” may modify various components regardless of order and/or importance. , but these expressions do not limit the corresponding components.
  • the above representations are only for the purpose of distinguishing one element from other elements.
  • a first user equipment and a second user equipment represent different user equipments, although both are user equipments.
  • a first element could be termed a second element, and similarly, a second element could be termed a first element, without departing from the scope of the disclosure.
  • An element e.g., a first element is said to be "(operably or communicatively) coupled to” or “(operably or communicatively) coupled to" another element (e.g., a second element).
  • another element e.g., a second element
  • one element e.g., a second element
  • another element e.g., a second element

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Biomedical Technology (AREA)
  • Multimedia (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Pathology (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Molecular Biology (AREA)
  • Image Processing (AREA)
  • Investigating Or Analysing Biological Materials (AREA)

Abstract

本申请公开了一种计算机可读存储介质。该计算机可读存储介质的内容当被处理器执行时，可以对包含肿瘤环境信息的病理组织切片进行多光子成像，并进一步通过图像处理进行肿瘤微环境的病理学划分；分别从肿瘤组织、侵袭边缘、正常组织三个区域提取各项胶原特征参数值，比如形态学、能量、纹理特征参数；根据区域特征参数计算区域间差异和变异程度；建立胶原特征评分模型；通过模型纳入的胶原特征参数计算出胶原特征评分；该胶原特征评分能够很好地补充TNM分期系统对肿瘤微环境"土壤"的评估缺失，更加有效地为结直肠癌预后预测提供参考。

Description

一种结直肠癌胶原评价和预后预测装置和存储介质
本申请要求于2022年03月08日提交中国专利局、申请号为202210218325.1、发明名称为“一种结直肠癌胶原评价和预后预测装置和存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及医学成像技术领域，特别涉及一种结直肠癌胶原评价和预后预测装置和存储介质。
背景技术
结直肠癌是世界范围内最常见的恶性肿瘤之一，且近年来发病率上升趋势明显。在结直肠癌预测判断中TNM分期系统是当前临床最广泛运用的胶原评价和预后预测方法，即通过评估原发灶(Tumor)、淋巴结受累(Nodes)和远处转移(Metastasis)情况对肿瘤进行评估和分期，进而预测预后。
但是TNM分期系统仍具有一定的局限性。一方面，相同分期的患者，在接受相似的治疗方案后，其临床结局也差异很大，在Ⅱ、Ⅲ期结直肠癌中，这一差异更为突出。另一方面，TNM分期装置关注肿瘤自身特征，而缺乏对肿瘤微环境“土壤”的评估。因此，基于肿瘤微环境的预后策略有待进一步补充。
发明内容
本申请的目的在于提供一种结直肠癌胶原评价和预后预测装置和存储介质，其能够改善上述问题。
本申请的实施例是这样实现的：
第一方面，本申请提供一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有计算机程序，所述计算机程序包括程序指令，所述程序指令当被处理器执行时使所述处理器执行以下步骤：
获取病理组织切片上目标成像区域的多光子成像图像，对所述多光子成像图像进行图像处理，将所述多光子成像图像划分为肿瘤中心区域、侵袭边缘区域和正常组织区域；
提取各个区域中的各项胶原特征参数值，根据所述胶原特征参数值分别计算出各个评分参数，所述评分参数包括各个区域的各项胶原特征参数平均值、区域间差异值和变异程度值；其中，所述区域间差异值为任意两个区域之间的所述胶原特征参数平均值之差值，所述变异程度值为其中两个所述区域间差异值的比值；
选取目标项胶原特征对应的所述评分参数进行胶原特征评分计算，得到所述病理组织切片的胶原特征评分；
联合患者的TNM分期信息和所述胶原特征评分进行预后预测。
可以理解，本申请公开了一种计算机可读存储介质。该计算机可读存储介质的内容被执行时可以对包含肿瘤环境信息的病理组织切片进行多光子成像，并进一步通过图像处理进行肿瘤微环境的病理学划分；分别从肿瘤组织、侵袭边缘、正常组织三个区域提取各项胶原特征参数值，比如形态学、能量、纹理特征参数；根据区域特征参数计算区域间差异和变异程度；建立胶原特征评分模型，通过模型纳入的胶原特征参数计算出胶原特征评分；该胶原特征评分能够很好地补充TNM分期系统对肿瘤微环境“土壤”的评估缺失，更加有效地为结直肠癌预后预测提供参考。
在本申请可选的实施例中，所述对所述多光子成像图像进行图像处理，将所述多光子成像图像划分为肿瘤中心区域、侵袭边缘区域和正常组织区域，包括：
提取出肿瘤和正常组织的交界线，确定包含所述肿瘤的区域为初始肿瘤中心区域，确定包含所述正常组织的区域为初始正常组织区域；
将所述交界线向所述初始肿瘤中心区域的方向平移第一距离后，形成第一边界线，将所述交界线向所述初始正常组织区域的方向平移第二距离后，形成第二边界线；
确定所述第一边界线和所述第二边界线之间的区域为所述侵袭边缘区域；
确定所述初始肿瘤中心区域中除所述侵袭边缘区域以外的区域为所述肿瘤中心区域，确定所述初始正常组织区域中除所述侵袭边缘区域以外的区域为所述正常组织区域。
在本申请可选的实施例中，所述提取各个区域中的各项胶原特征参数 值，根据所述胶原特征参数值分别计算出各个评分参数，包括：
在所述肿瘤中心区域内随机选取三个肿瘤中心子区域ROI1、ROI2、ROI3；计算三个所述肿瘤中心子区域的各项所述胶原特征参数值的平均值，作为所述肿瘤中心区域的各项所述胶原特征参数平均值；
在所述侵袭边缘区域内随机选取三个侵袭边缘子区域ROI4、ROI5、ROI6；计算三个所述侵袭边缘子区域的各项所述胶原特征参数值的平均值，作为所述侵袭边缘区域的各项所述胶原特征参数平均值；
在所述正常组织区域内随机选取三个正常组织子区域ROI7、ROI8、ROI9；计算三个所述正常组织子区域的各项所述胶原特征参数值的平均值，作为所述正常组织区域的各项所述胶原特征参数平均值。
其中，所述提取各个区域中的各项胶原特征参数值，根据所述胶原特征参数值分别计算出各个评分参数，还包括：
计算所述肿瘤中心区域和所述侵袭边缘区域的各项所述胶原特征参数平均值之差，作为第一区域间差值；
计算所述侵袭边缘区域和所述正常组织区域的各项所述胶原特征参数平均值之差，作为第二区域间差值；
计算所述肿瘤中心区域和所述正常组织区域的各项所述胶原特征参数平均值之差，作为第三区域间差值；
将所述第一区域间差值与所述第三区域间差值的比值作为区域间变异程度值。
可以理解，计算区域间差值和区域间变异程度值有利于多维地对肿瘤微环境胶原特征进行分析。其中，区域间变异程度值小于或等于1时，说明肿瘤微环境中肿瘤中心区域至侵袭边缘区域至正常组织区域为同一趋势变化，区域间变异程度较低；区域间变异程度值大于1时，说明肿瘤微环境中肿瘤中心区域至侵袭边缘区域至正常组织区域非同一趋势变化，区域间变异程度较高。
在本申请可选的实施例中，所述选取目标项胶原特征对应的所述评分参数进行胶原特征评分计算，包括：
通过LASSO回归算法构建胶原特征评分模型如下式：
Figure PCTCN2022131939-appb-000001
其中，Cf score代表所述胶原特征评分，collagen_feature i代表第i项所述目标项胶原特征对应的所述评分参数，coef i代表参数collagen_feature i对应的计算系数，k代表所述目标项胶原特征的数量。
第二方面，本申请提供一种结直肠癌胶原评价和预后预测装置，包括处理器、输入设备、输出设备和存储器，所述处理器、输入设备、输出设备和存储器相互连接，其中，所述存储器可以是如本发明第一方面中所述的计算机可读存储介质，所述处理器被配置用于调用所述程序指令。
有益效果：
本申请公开了一种计算机可读存储介质。该计算机可读存储介质的内容被执行时可以对包含肿瘤环境信息的病理组织切片进行多光子成像，并进一步通过图像处理进行肿瘤微环境的病理学划分；分别从肿瘤组织、侵袭边缘、正常组织三个区域提取各项胶原特征参数值，比如形态学、能量、纹理特征参数；根据区域特征计算区域间差异和变异程度；建立胶原特征评分模型，通过模型纳入的胶原特征参数计算出胶原特征评分；该胶原特征评分能够很好地补充TNM分期系统对肿瘤微环境“土壤”的评估缺失，更加有效地为结直肠癌预后预测提供参考。
计算区域间差值和区域间变异程度值有利于多维地对肿瘤微环境胶原特征进行分析。其中，区域间变异程度值小于或等于1时，说明肿瘤微环境中肿瘤中心区域至侵袭边缘区域至正常组织区域为同一趋势变化，区域间变异程度较低；区域间变异程度值大于1时，说明肿瘤微环境中肿瘤中心区域至侵袭边缘区域至正常组织区域非同一趋势变化，区域间变异程度较高。
为使本申请的上述目的、特征和优点能更明显易懂，下文特举可选实施例，并配合所附附图，作详细说明如下。
说明书附图
为了更清楚地说明本申请实施例的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，应当理解，以下附图仅示出了本申请的某些实施例，因此不应被看作是对范围的限定，对于本领域普通技术人员来讲， 在不付出创造性劳动的前提下，还可以根据这些附图获得其他相关的附图。
图1是本申请提供的一种计算机可读存储介质所存储的程序指令的流程示意图；
图2是本申请对多光子成像图像的划分示意图；
图3是本申请对胶原特征参数平均值的计算原理示意图；
图4是本申请联合TNM分期项目和胶原特征评分项目建立的生存列线图；
图5是本申请公开的结直肠癌胶原评价和预后预测方法与TNM分期系统胶原评价和预后预测方法的ROC曲线示意图；
图6是本申请提供的一种结直肠癌胶原评价和预后预测装置的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
TNM分期系统是国际上最为通用的肿瘤分期系统。TNM分期系统已经成为临床医生和医学科学工作者对于恶性肿瘤进行分期的标准方法，即通过评估原发灶(Tumor)、淋巴结受累(Nodes)和远处转移(Metastasis)情况对肿瘤进行评估和分期，进而预测预后。
但是TNM分期系统仍具有一定的局限性。一方面，相同分期的患者，在接受相似的治疗方案后，其临床结局也差异很大，在Ⅱ、Ⅲ期结直肠癌中，这一差异更为突出。另一方面，肿瘤并不是单一的肿瘤细胞的集合，还包括各种基质细胞、免疫细胞以及细胞外基质、微脉管等组成的肿瘤微环境，恶性肿瘤细胞相当于“种子”，肿瘤微环境相当于“土壤”，TNM分期系统关注肿瘤自身特征，而缺乏对肿瘤微环境这一“土壤”的评估。因此，基于肿瘤微环境的预后策略有待进一步补充。
第一方面，如图1所示，本申请提供一种计算机可读存储介质，所述 计算机可读存储介质存储有计算机程序，所述计算机程序包括程序指令，所述程序指令当被处理器执行时使所述处理器执行以下步骤：
110、获取病理组织切片上目标成像区域的多光子成像图像，对多光子成像图像进行图像处理，将多光子成像图像划分为肿瘤中心区域、侵袭边缘区域和正常组织区域。
为了能够补充TNM分期系统对肿瘤微环境“土壤”的评估缺失，上述病理组织切片上目标成像区域需要是处于肿瘤交界处的区域。上述病理组织切片的获取方法包括：将结直肠癌手术标本进行常规病理蜡块制备，选取肿瘤交界处的蜡块，连续切片2张，切片厚度为4μm，其中第一张进行HE染色，第二张作为上述病理组织切片准备进行多光子成像。上述目标成像区域的确定方法包括：通过数字切片扫描系统PRECICE 600对第一张切片进行20倍扫描，选择浸润深度较深的肿瘤交界区域作为上述交界区域；在第二张切片上与该交界区域的对应区域确定为目标成像区域，该目标成像区域的大小一般选为5mm*5mm。
上述多光子成像图像可以由Zeiss LSM 880 with Airyscan多光子共聚焦激光扫描系统对目标成像区域进行20倍扫描后获取。
多光子成像是一种基于非线性光学和飞秒激光镭射之上的多光子显微成像技术，其利用活体组织细胞本身内在的自体荧光及胶原产生的二次谐波进行成像，可无需染色直接获得组织及细胞的胶原组织学特征。基质胶原是肿瘤微环境结构的重要组成部分，参与肿瘤发生、发展的多个过程并影响预后，肿瘤微环境中的高胶原含量与不良预后相关。
如图2所示，ZEN软件(Zen 2.3 lite，Carl Zeiss Microscopy GmbH，2011)完成上述肿瘤中心(Center of Tumor，CT)区域、正常组织(Normal tissue，N)区域和侵袭边缘(Invasive Margin，IM)区域的划分。
120、提取各个区域中的各项胶原特征参数值，根据胶原特征参数值分别计算出各个评分参数，评分参数包括各个区域的各项胶原特征参数平均值、区域间差异值和变异程度值。
其中，区域间差异值为任意两个区域之间的胶原特征参数平均值之差值，变异程度值为其中两个区域间差异值的比值。
对各个区域中的各项胶原特征参数值提取可以由Matlab R2015b (MathWorks)软件完成。从上述多光子成像图像中一般可以提取出8项形态学特征参数、6项强度特征参数和128项纹理特征参数，共142项胶原特征参数。
其中，胶原形态学特征参数包括：胶原面积(Collagen area)、胶原纤维数量(Collagen fiber number)、胶原纤维长度(Collagen fiber length)、胶原纤维宽度(Collagen fiber width)、胶原纤维直线度(Collagen fiber straightness)、胶原纤维交联密度(Collagen crosslink density)、胶原纤维交联空间(Collagen crosslink space)和胶原排列方向(Collagen orientation)。
其中，胶原强度特征参数由柱状图来分析，胶原强度特征参数包括：柱状图均值(Histogram mean)、柱状图方差(Histogram Variance)、柱状图偏度(Histogram Skewness)、柱状图峰度(Histogram Kurtosis)、柱状图能量(Histogram Energy)和柱状图熵(Histogram Entropy)。
其中，胶原结构特征参数包括80个灰度共生矩阵(Gray-level co-occurrence matrix，GLCM)的纹理参数和48个Gabor小波变换参数(Gabor wavelet transform)。通过计算1、2、3、4和5号位置的5个像素以及0°、45°、90°和135°的4个不同方向上的对比度、相关性、能量和均匀性来计算上述80个GLCM。再通过4个不同尺度和5个不同方向的均值和方差，得到48个Gabor小波变换参数。
针对各个区域的各项胶原特征参数平均值的计算，如图3所示，可以由ZEN软件(Zen 2.3lite，Carl Zeiss Microscopy GmbH，2011)完成兴趣区域(Region of Interest，ROI)的选择。在肿瘤中心区域、侵袭边缘区域和正常组织区域内分别随机取3个ROI，得到共9个ROI，ROI大小为844像素×844像素(约700μm×700μm)。随后分别计算肿瘤中心、侵袭边缘和正常组织三个区域内ROI所提取的142项胶原特征参数值的平均值作为该区域的各项胶原特征参数平均值。根据区域特征计算出胶原参数的区域间差异和变异程度。
130、选取目标项胶原特征对应的评分参数进行胶原特征评分计算，得到病理组织切片的胶原特征评分。
如图5所示为本申请公开的计算机可读存储介质与TNM分期系统胶 原评价和预后预测方法的接受者操作特性曲线(receiver operating characteristic curve，ROC)示意图。ROC曲线，又称为感受性曲线(sensitivity curve)，得此名的原因在于曲线上各点反映着相同的感受性，它们都是对同一信号刺激的反应，只不过是在几种不同的判定标准下所得的结果而已。接受者操作特性曲线就是以虚惊概率为横轴，击中概率为纵轴所组成的坐标图，和被试在特定刺激条件下由于采用不同的判断标准得出的不同结果画出的曲线。AUC(Area Under Curve)被定义为ROC曲线下与坐标轴围成的面积，显然这个面积的数值不会大于1。又由于ROC曲线一般都处于y＝x这条直线的上方，所以AUC的取值范围在0.5和1之间。AUC越接近1.0，检测方法真实性越高；等于0.5时，则真实性最低，无应用价值。
由图5可以看出，TNM分期系统胶原评价和预后预测方法所得出的患者5年无病生存率(disease free survival，DFS)的AUC仅为0.674(图5中TNM线所示)；通过本申请所公开的计算机可读存储介质所得出的患者5年无病生存率的AUC显著提升为0.742(图5中TNM+CS TME线所示)。由此可以看出，本申请所公开的计算机可读存储介质的预测效能更高。
可以理解，本申请公开了一种计算机可读存储介质。该方法对包含肿瘤环境信息的病理组织切片进行多光子成像，并进一步通过图像处理进行肿瘤微环境的病理学划分；分别从肿瘤组织、侵袭边缘、正常组织三个区域提取各项胶原特征参数值，比如形态学、能量、纹理特征参数；根据区域特征计算区域间差异和变异程度；建立胶原特征评分模型，通过模型纳入的胶原特征参数计算出胶原特征评分；该胶原特征评分能够很好地补充TNM分期系统对肿瘤微环境“土壤”的评估缺失，更加有效地为结直肠癌预后预测提供参考。
在本申请可选的实施例中，步骤“对多光子成像图像进行图像处理，将多光子成像图像划分为肿瘤中心区域、侵袭边缘区域和正常组织区域”可以由ZEN软件(Zen 2.3 lite，Carl Zeiss Microscopy GmbH，2011)完成，该步骤具体包括：
111、提取出肿瘤和正常组织的交界线，确定包含肿瘤的区域为初始 肿瘤中心区域，确定包含正常组织的区域为初始正常组织区域。
其中，可以通过图像算法提取肿瘤和正常组织的交界线，如图2中的虚线所示，虚线左侧为上述初始肿瘤中心区域，虚线右侧则为上述初始正常组织区域。
112、将交界线向初始肿瘤中心区域的方向平移第一距离后，形成第一边界线，将交界线向初始正常组织区域的方向平移第二距离后，形成第二边界线。
一般来说，可以将交界线向初始肿瘤中心区域的方向平移500μm形成第一边界线L1(如图2所示)，将交界线向初始正常组织区域的方向平移500μm形成第二边界线L2(如图2所示)。
113、确定第一边界线和第二边界线之间的区域为侵袭边缘区域。
如图2中的侵袭边缘(Invasive Margin，IM)区域所示。
114、确定初始肿瘤中心区域中除侵袭边缘区域以外的区域为肿瘤中心区域，确定初始正常组织区域中除侵袭边缘区域以外的区域为正常组织区域。
如图2中的肿瘤中心(Center of Tumor，CT)区域和正常组织(Normal tissue，N)区域所示。
在本申请可选的实施例中，步骤120具体包括：
121、计算肿瘤中心区域和侵袭边缘区域的各项胶原特征参数平均值之差，作为第一区域间差值。
122、计算侵袭边缘区域和正常组织区域的各项胶原特征参数平均值之差，作为第二区域间差值。
123、计算肿瘤中心区域和正常组织区域的各项胶原特征参数平均值之差，作为第三区域间差值。
124、将第一区域间差值与第三区域间差值的比值作为区域间变异程度值。
可以理解，计算区域间差值和区域间变异程度值有利于多维地对肿瘤微环境胶原特征进行分析。其中，区域间变异程度值小于或等于1时，说明肿瘤微环境中肿瘤中心区域至侵袭边缘区域至正常组织区域为同一趋势变化，区域间变异程度较低；区域间变异程度值大于1时，说明肿瘤微 环境中肿瘤中心区域至侵袭边缘区域至正常组织区域非同一趋势变化，区域间变异程度较高。
在本申请可选的实施例中，步骤130具体包括：
通过LASSO回归算法构建胶原特征评分模型如下式：
Figure PCTCN2022131939-appb-000002
其中，Cf score代表所述胶原特征评分，collagen_feature i代表第i项所述目标项胶原特征对应的所述评分参数，coef i代表参数collagen_feature i对应的计算系数，k代表所述目标项胶原特征的数量。
其中，k＝16；
collagen_feature 1代表所述侵袭边缘区域的胶原面积值，coef 1为0.181；
collagen_feature 2代表所述侵袭边缘区域的胶原数量值，coef 2为0.081；
collagen_feature 3代表所述侵袭边缘区域的胶原长度值，coef 3为0.303；
collagen_feature 4代表所述正常组织区域的胶原排布，coef 4为-0.515；
collagen_feature 5代表胶原交联密度的所述第一区域间差值，coef 5为-0.126；
collagen_feature 6代表所述肿瘤中心区域的灰度共生矩阵中像素距离为3在0°方向上的相关性，coef 6为0.018；
collagen_feature 7代表所述肿瘤中心区域的尺度为3、方向为60°的Gabor滤器图像卷积平均值，coef 7为0.477；
collagen_feature 8代表所述侵袭边缘区域的尺度为1、方向为60°的Gabor滤器图像卷积平均值，coef 8为-0.119；
collagen_feature 9代表所述侵袭边缘区域的尺度为3、方向为60°的Gabor滤器图像卷积平均值，coef 9为0.232；
collagen_feature 10代表尺度为3、方向为60°的Gabor滤器图像卷积方差的所述第三区域间差值，coef 10为0.016；
collagen_feature 11代表灰度共生矩阵中像素距离为3在45°方向上的对比度的所述第一区域间差值，coef 11为0.141；
collagen_feature 12代表尺度为2、方向为90°的Gabor滤器图像卷积方差的所述第一区域间差值，coef 12为0.048；
collagen_feature 13代表尺度为4、方向为90°的Gabor滤器图像卷积方差的所述第一区域间差值，coef 13为0.12；
collagen_feature 14代表灰度共生矩阵中像素距离为1在90°方向上的相关性的所述区域间变异程度值，coef 14为0.24；
collagen_feature 15代表灰度共生矩阵中像素距离为2在135°方向上的相关性的所述区域间变异程度值，coef 15为0.145；
collagen_feature 16代表灰度共生矩阵中像素距离为4在0°方向上的相关性的所述区域间变异程度值，coef 16为0.303。
获得胶原评分后，医生可以联合患者的TNM分期信息和胶原特征评分进行预后预测。具体包括：
141、联合TNM分期项目和胶原特征评分项目建立生存列线图。
步骤141所建立的生存列线图如图4所示。生存列线图包括平行建立分项分数线段、T分期线段、N分期线段、胶原特征评分线段；其中，所述分项分数线段上均匀排布有0分到10分之间的各个分数值；所述T分期线段中的T1/2点、T3点、T4a点和T4b点分别对应所述分项分数线段的0分点、4.8分点、6.6分点、8.3分点；所述N分期线段中的N0点、N1/N1c点、N2a点和N2b点分别对应所述分项分数线段的0分点、2.5分点、3.4分点、5.3分点；所述胶原特征评分线段上均匀排布有-1.8分到0.2分之间的各个分数值，其中0.2分点和2.4分点分别与所述分项分数线段的0分点和10分点对应。
生存列线图还包括平行建立总分线段、1年无病生存率线段、3年无病生存率线段、5年无病生存率线段；其中，所述总分线段上均匀排布有0分到24分之间的各个分数值；所述1年无病生存率线段中的0.9点、0.8点、0.7点和0.6点分别对应所述总分线段的12.8分点、15.6分点、16.8分点和18分点；所述3年无病生存率线段中的0.9点、0.8点、0.7点和0.6点分别对应所述总分线段的10.4分点、12.4分点、14.4分点和15.2分点；所述5年无病生存率线段中的0.9点、0.8点、0.7点和0.6点分别对应所述总分线段的9.6分点、12分点、13.2分点和14.4分点。
142、将联合患者的TNM分期信息和胶原特征评分带入生存列线图进行预后预测，得到患者的无病生存率。
举例说明，患者甲计算得到的胶原特征评分为2.305，T分期为T4a，N分期为N1。根据胶原特征评分与该患者的TNM分期，通过图4所示的列线图评估发现，患者术后(尤其是3年后)出现复发、转移或肿瘤相关死亡的可能性较大，建议加强随访密度、增加术后复查内容和次数，选择较为积极的辅助治疗策略。
举例说明，患者乙胶原特征评分为0.820，T分期为T3，N分期为N1根据胶原特征评分与该患者的TNM分期，通过图4所示的列线图评估发现，患者术后出现复发、转移或肿瘤相关死亡的可能性较小，建议常规随访与术后复查，选择常规辅助治疗策略。
第二方面，本申请提供一种结直肠癌胶原评价和预后预测装置。如图6所示，结直肠癌胶原评价和预后预测装置包括一个或多个处理器601；一个或多个输入设备602，一个或多个输出设备603和存储器604。上述处理器601、输入设备602、输出设备603和存储器604通过总线连接。存储器604可以是第一方面任一项的所述计算机可读存储介质，所述处理器601被配置用于调用所述程序指令。
应当理解，在本发明实施例中，所称处理器601可以是中央处理单元(Central Processing Unit，CPU)，该处理器还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
输入设备602可以用于获取病理组织切片上目标成像区域的多光子成像图像。
输出设备603可以用于将通过第一方面任一项方法得到的预后预测输出给目标终端进行显示。
该存储器604可以包括只读存储器和随机存取存储器，并向处理器601提供指令和数据。存储器604的一部分还可以包括非易失性随机存取存储器。例如，存储器604还可以存储设备类型的信息。
具体实现中，本发明实施例中所描述的处理器601、输入设备602、 输出设备603可执行第一方面任一方法所描述的实现方式，也可执行本发明实施例所描述的终端设备的实现方式，在此不再赘述。
上述计算机可读存储介质可以是前述任一实施例的终端设备的内部存储单元，例如终端设备的硬盘或内存。上述计算机可读存储介质也可以是上述终端设备的外部存储设备，例如上述终端设备上配备的插接式硬盘，智能存储卡(Smart Media Card，SMC)，安全数字(Secure Digital，SD)卡，闪存卡(Flash Card)等。进一步地，上述计算机可读存储介质还可以既包括上述终端设备的内部存储单元也包括外部存储设备。上述计算机可读存储介质用于存储上述计算机程序以及上述终端设备所需的其他程序和数据。上述计算机可读存储介质还可以用于暂时地存储已经输出或者将要输出的数据。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、计算机软件或者二者的结合来实现，为了清楚地说明硬件和软件的可互换性，在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。
在本申请所提供的几个实施例中，应该理解到，所揭露的终端设备和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，上述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个装置，或一些特征可以忽略，或不执行。另外，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口、装置或单元的间接耦合或通信连接，也可以是电的、机械的或其它的形式连接。
上述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本发明实施例方案的目的。
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元 中，也可以是各个单元单独物理存在，也可以是两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
上述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分，或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本发明各个实施例中方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
在本公开的各种实施方式中所使用的表述“第一”、“第二”、“所述第一”或“所述第二”可修饰各种部件而与顺序和/或重要性无关，但是这些表述不限制相应部件。以上表述仅配置为将元件与其它元件区分开的目的。例如，第一用户设备和第二用户设备表示不同的用户设备，虽然两者均是用户设备。例如，在不背离本公开的范围的前提下，第一元件可称作第二元件，类似地，第二元件可称作第一元件。
当一个元件(例如，第一元件)称为与另一元件(例如，第二元件)“(可操作地或可通信地)联接”或“(可操作地或可通信地)联接至”另一元件(例如，第二元件)或“连接至”另一元件(例如，第二元件)时，应理解为该一个元件直接连接至该另一元件或者该一个元件经由又一个元件(例如，第三元件)间接连接至该另一个元件。相反，可理解，当元件(例如，第一元件)称为“直接连接”或“直接联接”至另一元件(第二元件)时，则没有元件(例如，第三元件)插入在这两者之间。
以上描述仅为本申请的可选实施例以及对所运用技术原理的说明。本领域技术人员应当理解，本申请中所涉及的发明范围，并不限于上述技术特征的特定组合而成的技术方案，同时也应涵盖在不脱离上述发明构思的情况下，由上述技术特征或其等同特征进行任意组合而形成的其它技术方案。例如上述特征与本申请中公开的(但不限于)具有类似功能的技术特征 进行互相替换而形成的技术方案。
以上所述仅为本申请的可选实施例而已，并不用于限制本申请，对于本领域的技术人员来说，本申请可以有各种更改和变化。凡在本申请的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。

Claims (7)

  1. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有计算机程序，所述计算机程序包括程序指令，所述程序指令当被处理器执行时使所述处理器执行以下步骤：
    获取病理组织切片上目标成像区域的多光子成像图像，对所述多光子成像图像进行图像处理，将所述多光子成像图像划分为肿瘤中心区域、侵袭边缘区域和正常组织区域；
    提取各个区域中的各项胶原特征参数值，根据所述胶原特征参数值分别计算出各个评分参数，所述评分参数包括各个区域的各项胶原特征参数平均值、区域间差异值和变异程度值；其中，所述区域间差异值为任意两个区域之间的所述胶原特征参数平均值之差值，所述变异程度值为其中两个所述区域间差异值的比值；
    选取目标项胶原特征对应的所述评分参数进行胶原特征评分计算，得到所述病理组织切片的胶原特征评分。
  2. 根据权利要求1所述的计算机可读存储介质，其特征在于，
    所述对所述多光子成像图像进行图像处理，将所述多光子成像图像划分为肿瘤中心区域、侵袭边缘区域和正常组织区域，包括：
    提取出肿瘤和正常组织的交界线，确定包含所述肿瘤的区域为初始肿瘤中心区域，确定包含所述正常组织的区域为初始正常组织区域；
    将所述交界线向所述初始肿瘤中心区域的方向平移第一距离后，形成第一边界线，将所述交界线向所述初始正常组织区域的方向平移第二距离后，形成第二边界线；
    确定所述第一边界线和所述第二边界线之间的区域为所述侵袭边缘区域；
    确定所述初始肿瘤中心区域中除所述侵袭边缘区域以外的区域为所述肿瘤中心区域，确定所述初始正常组织区域中除所述侵袭边缘区域以外的区域为所述正常组织区域。
  3. 根据权利要求1所述的计算机可读存储介质，其特征在于，
    所述提取各个区域中的各项胶原特征参数值，根据所述胶原特征参数值分别计算出各个评分参数，包括：
    在所述肿瘤中心区域内随机选取三个肿瘤中心子区域ROI1、ROI2、 ROI3；计算三个所述肿瘤中心子区域的各项所述胶原特征参数值的平均值，作为所述肿瘤中心区域的各项所述胶原特征参数平均值；
    在所述侵袭边缘区域内随机选取三个侵袭边缘子区域ROI4、ROI5、ROI6；计算三个所述侵袭边缘子区域的各项所述胶原特征参数值的平均值，作为所述侵袭边缘区域的各项所述胶原特征参数平均值；
    在所述正常组织区域内随机选取三个正常组织子区域ROI7、ROI8、ROI9；计算三个所述正常组织子区域的各项所述胶原特征参数值的平均值，作为所述正常组织区域的各项所述胶原特征参数平均值。
  4. 根据权利要求3所述的计算机可读存储介质，其特征在于，
    所述提取各个区域中的各项胶原特征参数值，根据所述胶原特征参数值分别计算出各个评分参数，还包括：
    计算所述肿瘤中心区域和所述侵袭边缘区域的各项所述胶原特征参数平均值之差，作为第一区域间差值；
    计算所述侵袭边缘区域和所述正常组织区域的各项所述胶原特征参数平均值之差，作为第二区域间差值；
    计算所述肿瘤中心区域和所述正常组织区域的各项所述胶原特征参数平均值之差，作为第三区域间差值；
    将所述第一区域间差值与所述第三区域间差值的比值作为区域间变异程度值。
  5. 根据权利要求4所述的计算机可读存储介质，其特征在于，
    所述选取目标项胶原特征对应的所述评分参数进行胶原特征评分计算，包括：
    通过LASSO回归算法构建胶原特征评分模型如下式：
    Figure PCTCN2022131939-appb-100001
    其中，Cf score代表所述胶原特征评分，collagen_feature i代表第i项所述目标项胶原特征对应的所述评分参数，coef i代表参数collagen_feature i对应的计算系数，k代表所述目标项胶原特征的数量。
  6. 根据权利要求5所述的计算机可读存储介质，其特征在于，
    其中，k＝16；
    collagen_feature 1代表所述侵袭边缘区域的胶原面积值，coef 1为0.181；
    collagen_feature 2代表所述侵袭边缘区域的胶原数量值，coef 2为0.081；
    collagen_feature 3代表所述侵袭边缘区域的胶原长度值，coef 3为0.303；
    collagen_feature 4代表所述正常组织区域的胶原排布，coef 4为-0.515；
    collagen_feature 5代表胶原交联密度的所述第一区域间差值，coef 5为-0.126；
    collagen_feature 6代表所述肿瘤中心区域的灰度共生矩阵中像素距离为3在0°方向上的相关性，coef 6为0.018；
    collagen_feature 7代表所述肿瘤中心区域的尺度为3、方向为60°的Gabor滤器图像卷积平均值，coef 7为0.477；
    collagen_feature 8代表所述侵袭边缘区域的尺度为1、方向为60°的Gabor滤器图像卷积平均值，coef 8为-0.119；
    collagen_feature 9代表所述侵袭边缘区域的尺度为3、方向为60°的Gabor滤器图像卷积平均值，coef 9为0.232；
    collagen_feature 10代表尺度为3、方向为60°的Gabor滤器图像卷积方差的所述第三区域间差值，coef 10为0.016；
    collagen_feature 11代表灰度共生矩阵中像素距离为3在45°方向上的对比度的所述第一区域间差值，coef 11为0.141；
    collagen_feature 12代表尺度为2、方向为90°的Gabor滤器图像卷积方差的所述第一区域间差值，coef 12为0.048；
    collagen_feature 13代表尺度为4、方向为90°的Gabor滤器图像卷积方差的所述第一区域间差值，coef 13为0.12；
    collagen_feature 14代表灰度共生矩阵中像素距离为1在90°方向上的相关性的所述区域间变异程度值，coef 14为0.24；
    collagen_feature 15代表灰度共生矩阵中像素距离为2在135°方向上的相关性的所述区域间变异程度值，coef 15为0.145；
    collagen_feature 16代表灰度共生矩阵中像素距离为4在0°方向上的相关性的所述区域间变异程度值，coef 16为0.303。
  7. 一种结直肠癌胶原评价和预后预测装置，其特征在于，
    包括处理器、输入设备、输出设备和存储器，所述处理器、输入设备、 输出设备和存储器相互连接，其中，所述存储器包括如权利要求1至6任一项的所述计算机可读存储介质，所述处理器被配置用于调用所述程序指令。
PCT/CN2022/131939 2022-03-08 2022-11-15 一种结直肠癌胶原评价和预后预测装置和存储介质 WO2023168981A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/286,475 US11990241B2 (en) 2022-03-08 2022-11-15 Apparatus for collagen evaluation and prognostic prediction of colorectal cancer and storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210218325.1A CN114299069B (zh) 2022-03-08 2022-03-08 一种结直肠癌胶原评价和预后预测装置和存储介质
CN202210218325.1 2022-03-08

Publications (1)

Publication Number Publication Date
WO2023168981A1 true WO2023168981A1 (zh) 2023-09-14

Family

ID=80978494

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/131939 WO2023168981A1 (zh) 2022-03-08 2022-11-15 一种结直肠癌胶原评价和预后预测装置和存储介质

Country Status (3)

Country Link
US (1) US11990241B2 (zh)
CN (1) CN114299069B (zh)
WO (1) WO2023168981A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114299069B (zh) * 2022-03-08 2022-05-13 南方医科大学南方医院 一种结直肠癌胶原评价和预后预测装置和存储介质
CN114926465B (zh) * 2022-07-20 2022-09-30 南方医科大学南方医院 一种存储介质和结直肠癌淋巴结转移预测模型构建系统
CN114998348B (zh) * 2022-08-03 2022-11-01 南方医科大学南方医院 一种计算机可读存储介质及直肠癌预后预测模型构建系统
CN115330949B (zh) * 2022-08-16 2024-04-05 中山大学附属第一医院 一种肝细胞癌侵袭边缘预测方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107941760A (zh) * 2017-11-16 2018-04-20 南方医科大学南方医院 直肠癌切除标本的切缘处胶原组织评价方法
CN110391015A (zh) * 2019-06-14 2019-10-29 广东省人民医院(广东省医学科学院) 一种基于影像组学量化肿瘤免疫状态的方法
US20200234442A1 (en) * 2017-07-24 2020-07-23 Ventana Medical Systems, Inc. Methods and systems for evaluation of immune cell infiltrate in tumor samples
CN113234829A (zh) * 2021-06-23 2021-08-10 至本医疗科技（上海）有限公司 结肠癌预后评估基因集及其构建方法
CN114299069A (zh) * 2022-03-08 2022-04-08 南方医科大学南方医院 一种结直肠癌胶原评价和预后预测装置和存储介质

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7805183B2 (en) * 2006-06-22 2010-09-28 Wisconsin Alumni Research Foundation Stromal collagen in the diagnosis and characterization of breast cancer
US8280132B2 (en) * 2006-08-01 2012-10-02 Rutgers, The State University Of New Jersey Malignancy diagnosis using content-based image retreival of tissue histopathology
WO2008109414A2 (en) * 2007-03-02 2008-09-12 Wisconsin Alumni Research Foundation Use of endogenous fluorescence to identify invading metastatic breast tumor cells
SG184389A1 (en) * 2010-03-31 2012-10-30 Agency Science Tech & Res A method and system for determining a stage of fibrosis in a liver
CN105472250B (zh) * 2015-12-23 2018-12-07 浙江宇视科技有限公司 自动聚焦方法及装置
CN107038715B (zh) * 2017-03-21 2022-03-08 腾讯科技（深圳）有限公司 一种图像处理方法及装置
US11627908B2 (en) * 2017-03-31 2023-04-18 University Of Massachusetts Instruments and methods for imaging collagen structure in vivo
CN110632072A (zh) * 2019-09-20 2019-12-31 福建师范大学 基于肿瘤微环境胶原分布进行乳腺癌预后评估的新方法
CN112288704B (zh) * 2020-10-26 2021-09-28 中国人民解放军陆军军医大学第一附属医院 一种基于核密度函数的量化胶质瘤侵袭性的可视化方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200234442A1 (en) * 2017-07-24 2020-07-23 Ventana Medical Systems, Inc. Methods and systems for evaluation of immune cell infiltrate in tumor samples
CN107941760A (zh) * 2017-11-16 2018-04-20 南方医科大学南方医院 直肠癌切除标本的切缘处胶原组织评价方法
CN110391015A (zh) * 2019-06-14 2019-10-29 广东省人民医院(广东省医学科学院) 一种基于影像组学量化肿瘤免疫状态的方法
CN113234829A (zh) * 2021-06-23 2021-08-10 至本医疗科技（上海）有限公司 结肠癌预后评估基因集及其构建方法
CN114299069A (zh) * 2022-03-08 2022-04-08 南方医科大学南方医院 一种结直肠癌胶原评价和预后预测装置和存储介质

Also Published As

Publication number Publication date
US20240096491A1 (en) 2024-03-21
CN114299069A (zh) 2022-04-08
US11990241B2 (en) 2024-05-21
CN114299069B (zh) 2022-05-13

Similar Documents

Publication Publication Date Title
WO2023168981A1 (zh) 一种结直肠癌胶原评价和预后预测装置和存储介质
Sun et al. Deep learning vs. radiomics for predicting axillary lymph node metastasis of breast cancer using ultrasound images: don't forget the peritumoral region
Rajan et al. Brain tumor detection and segmentation by intensity adjustment
CN111127466B (zh) 医学图像检测方法、装置、设备及存储介质
US10339648B2 (en) Quantitative predictors of tumor severity
US9710908B2 (en) Method and system for assessing fibrosis in a tissue
US7599542B2 (en) System and method for detection and display of diseases and abnormalities using confidence imaging
Lee et al. Detection of neovascularization based on fractal and texture analysis with interaction effects in diabetic retinopathy
WO2021129325A1 (zh) 超声图像病灶的分割方法、装置和计算机设备
WO2015040990A1 (ja) 疾患分析装置、制御方法、及びプログラム
CN109978004B (zh) 图像识别方法及相关设备
Hu et al. Non-invasive evaluation for benign and malignant subcentimeter pulmonary ground-glass nodules (≤ 1 cm) based on CT texture analysis
Varma et al. An alternative approach to detect breast cancer using digital image processing techniques
Kaur et al. Computer-aided diagnosis of renal lesions in CT images: a comprehensive survey and future prospects
CN114332132A (zh) 图像分割方法、装置和计算机设备
Firmansyah et al. Detection melanoma cancer using ABCD rule based on mobile device
CN114529541A (zh) 预测胰腺癌微环境中细胞表达的方法、电子设备、介质及系统
JP7270917B2 (ja) コンピュータプログラム、及び画像処理装置
Armya et al. Medical images segmentation based on unsupervised algorithms: a review
WO2014006421A1 (en) Identification of mitotic cells within a tumor region
Kamalakannan et al. Classification of breast abnormality using decision tree based on GLCM features in mammograms
Wang et al. FSOU-Net: Feature supplement and optimization U-Net for 2D medical image segmentation
Bhagavan et al. A compressive survey on different image processing techniques to identify the brain tumor.
Swami et al. An algorithmic detection of brain tumour using image filtering and segmentation of various radiographs
Wang et al. HER2-MCNN: a HER2 classification method based on multi convolution neural network

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 18286475

Country of ref document: US

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22930604

Country of ref document: EP

Kind code of ref document: A1