CN111986157A - Digital pathological image quality evaluation system - Google Patents
Digital pathological image quality evaluation system Download PDFInfo
- Publication number
- CN111986157A CN111986157A CN202010703756.8A CN202010703756A CN111986157A CN 111986157 A CN111986157 A CN 111986157A CN 202010703756 A CN202010703756 A CN 202010703756A CN 111986157 A CN111986157 A CN 111986157A
- Authority
- CN
- China
- Prior art keywords
- image
- area
- tissue
- preprocessing
- digital pathological
- 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.)
- Granted
Links
- 230000001575 pathological effect Effects 0.000 title claims abstract description 78
- 238000013441 quality evaluation Methods 0.000 title claims abstract description 34
- 238000001514 detection method Methods 0.000 claims abstract description 85
- 238000007781 pre-processing Methods 0.000 claims abstract description 58
- 239000003292 glue Substances 0.000 claims description 32
- 238000002372 labelling Methods 0.000 claims description 32
- 230000037303 wrinkles Effects 0.000 claims description 28
- 238000009826 distribution Methods 0.000 claims description 24
- 238000010586 diagram Methods 0.000 claims description 19
- 238000004043 dyeing Methods 0.000 claims description 18
- 230000007170 pathology Effects 0.000 claims description 13
- 238000012952 Resampling Methods 0.000 claims description 12
- 238000001914 filtration Methods 0.000 claims description 12
- 238000003475 lamination Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000005520 cutting process Methods 0.000 claims description 7
- 239000006185 dispersion Substances 0.000 claims description 6
- 238000010186 staining Methods 0.000 claims description 4
- 238000001303 quality assessment method Methods 0.000 claims description 3
- 238000011160 research Methods 0.000 abstract description 6
- 239000012141 concentrate Substances 0.000 abstract description 2
- 238000011156 evaluation Methods 0.000 description 4
- 238000000034 method Methods 0.000 description 3
- 238000010827 pathological analysis Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000004040 coloring Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- 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
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
- G06T2207/30108—Industrial image inspection
-
- 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
- G06T2207/30168—Image quality inspection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Image Analysis (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention provides a digital pathological image quality evaluation system, which is characterized by comprising the following components: a digital pathological image preprocessing module; a stain detection module; a scanning quality evaluation module; and the scoring module is used for calculating a final effective area, and simultaneously collecting all the subentry detection scores given by the digital pathological image preprocessing module, the stain detection module and the scanning quality evaluation module to give overall quality evaluation of the pathological image. The invention provides a digital pathological image quality evaluation system, which is used for carrying out unified quality evaluation on conventionally acquired medical pathological images, effectively removing useless information in the digital pathological images, enabling doctors to concentrate on effective pathological tissue structures and providing reliable data guarantee for academic research based on the medical pathological images.
Description
Technical Field
The invention relates to an image processing method, in particular to a digital pathological image quality evaluation system.
Background
Noise is inevitably introduced during conventional slide preparation and pathological image digitization, and is commonly: slide glass cutting marks/cracks, slice folds/lamination, bubbles/glue solution/manual labeling of doctors, and focusing blurring. The noises not only influence the judgment of a clinician, but also are not beneficial to research of scientific research personnel, for example, related academic problems are intelligently interpreted by digital pathological images developed based on a deep learning algorithm in the future, and the quality of the digital pathological images is greatly limited. Therefore, the correct evaluation of the quality of the digital pathological image can provide reliable guarantee for pathological diagnosis and medical academic research.
At present, the existing medical pathological image processing is limited in the aspect of image quality evaluation, for example, chinese patent CN109191457A discloses a pathological image quality effectiveness identification method, and the method utilizes labeled data to train a two-classifier to obtain whether a pathological image meets the use requirements of a clinician. Chinese patent CN104408717A discloses a pathological image quality comprehensive evaluation method based on coloring separation, which separates the colors of three channels of an image, respectively calculates the pixel characteristics of each channel to be used as evaluation indexes, utilizes neural network training prediction to compare the predicted value with the true value, and realizes the evaluation of the pathological image color quality. Therefore, the digital pathological image quality evaluation system has profound significance for pathological diagnosis and medical research.
Disclosure of Invention
The purpose of the invention is: a system capable of performing uniform quality evaluation of conventionally acquired medical pathology images is provided.
In order to achieve the above object, an aspect of the present invention provides a digital pathological image quality evaluation system, including:
the digital pathological image preprocessing module is used for resampling the digital pathological image, generating an image size suitable for system detection, detecting whether the number of the tissue sections is in compliance or not, and giving a detection score;
the stain detection module is used for detecting a cutter mark/crack/wrinkle/lamination area and a bubble/glue solution/manual labeling area and giving detection scores of all items;
the scanning quality evaluation module is used for detecting the fuzzy area, detecting the dyeing quality and giving out each item detection score;
and the scoring module is used for calculating a final effective area, and simultaneously collecting all the subentry detection scores given by the digital pathological image preprocessing module, the stain detection module and the scanning quality evaluation module to give overall quality evaluation of the pathological image.
Preferably, the digital pathology image preprocessing module comprises a digital pathology image preprocessing unit and a tissue section quantity compliance detection unit, wherein:
the implementation of the digital pathological diagram preprocessing unit comprises the following steps:
step 101: inputting a digital pathological diagram with original size, analyzing the maximum acquisition multiplying power of the digital pathological diagram, combining the preprocessing target multiplying power to obtain a resampling ratio,
step 102: resampling the original-size digital pathology map based on the resampling ratio to obtain a global preprocessing map;
the implementation of the tissue section quantity compliance detection unit comprises the following steps:
step 201: inputting a global preprocessing chart obtained by a digital pathological chart preprocessing unit, identifying a tissue part area, and acquiring the coordinate distribution of the tissue part area;
step 202: cutting the tissue part according to the coordinates of the tissue part area to obtain a cutting picture;
step 203: counting the number N of the available tissue sections, giving the score of the available sections,
preferably, the stain detection module comprises a knife mark \ crack \ wrinkle \ laminated area detection unit and a bubble \ glue solution \ manual labeling area detection unit, wherein:
the implementation of the cutter mark \ crack \ wrinkle \ laminated region detection unit comprises the following steps:
step 301: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
step 302: performing convolution on the original tissue area by using a convolution kernel with the size of 10 multiplied by 10 and the filling value of 0.01 to obtain a convolved image;
step 303: calculating the difference value between the convolved image and the original tissue area image to obtain a difference value image;
step 304: performing opening operation on the difference image, taking intersection of the calculation result and the original tissue area image, and obtaining a knife mark \ crack \ wrinkle \ lamination problem area image;
step 305: taking a difference set between the original tissue area and the knife mark \ crack \ wrinkle \ lamination problem area image to obtain an effective tissue area image;
step 306: calculating the ratio of the effective tissue area image to the original tissue area image as r, and giving a score of a knife mark \ crack \ wrinkle \ laminated area, wherein the score of the knife mark \ crack \ wrinkle \ laminated area is 100 multiplied by r;
the realization of the bubble \ glue solution \ manual labeling area detection unit comprises the following steps:
step 401: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
step 402: frangi filtering is carried out on the original tissue area image to obtain a bubble \ glue solution \ manual labeling edge outline image;
step 403: after the bubble \ glue solution \ manual labeling edge contour image is filled, intersecting with the original tissue area image to obtain a bubble \ glue solution \ manual labeling problem area image;
step 404: taking a difference set between the original tissue area image and the bubble \ glue solution \ manual labeling problem area image to obtain an effective tissue area image;
step 405: and calculating the ratio of the effective tissue area image to the original tissue area image as r, and giving a bubble \ glue solution \ manual labeling area score which is 100 multiplied by r.
Preferably, the scan quality evaluation module includes a blurred region detection unit and a staining quality detection unit, wherein:
the implementation of the fuzzy area detection unit comprises the following steps:
step 501: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
step 502: carrying out laplacian filtering on the original tissue area image, calculating laplacian gradient, and obtaining a fuzzy thermal dispersion distribution diagram;
step 503: filtering fuzzy area distribution larger than a threshold value by using a threshold value on the fuzzy thermal dispersion point distribution diagram, and carrying out gaussian filtering on the fuzzy area distribution to obtain a fuzzy area mask;
step 504: taking a difference set between the original tissue area image and the fuzzy area mask to obtain an effective tissue area image;
step 505: calculating the ratio of the effective tissue area image to the original tissue area image as r, and giving a fuzzy area detection score which is 100 multiplied by r;
the realization of the dyeing quality detection unit comprises the following steps:
step 601: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
step 602: calculating RGB pixel distribution of the original tissue region image, and obtaining a distribution histogram and an average value of each channel in R, G, B channels;
step 603: calculating a data difference value of the R channel and the B channel to obtain dyeing red and blue contrast data;
step 604: based on contrast difference thresholdJudging the red and blue channel contrast data, giving dyeing quality detection score, counting RGB pixel distribution histogram of all pathological images in the pathological database, and taking the standard deviation of global R channel and B channel as the dyeing quality standard Wherein R isstdAnd BstdIs the standard deviation, R, of the R and B channels in the databasestdAnd bstdTo detect the standard deviation of the R and B channels in the image.
Preferably, the scoring module comprises a final effective region calculating unit and a quality scoring unit, wherein:
the implementation of the final effective area calculation unit comprises the following steps:
step 701: summarizing effective tissue area images output by a cutter mark/crack/wrinkle/laminated area detection unit, a bubble/glue solution/manual labeling area detection unit and a fuzzy area detection unit;
step 702: and taking intersection of the effective tissue area images to obtain a final effective tissue area image.
The realization of the quality scoring unit comprises the following steps:
step 801: summarizing available section scoring, cutter mark \ crack \ wrinkle \ laminated region scoring, bubble \ glue solution \ manual labeling region scoring, fuzzy region detection scoring and dyeing quality detection scoring;
step 802: weighted average is taken from all the scores obtained in the step 801 to obtain the final quality score;
step 803: and according to the score threshold, giving a final good/bad judgment result and giving a final effective tissue area image.
The invention provides a digital pathological image quality evaluation system, which is used for carrying out unified quality evaluation on conventionally acquired medical pathological images, effectively removing useless information in the digital pathological images, enabling doctors to concentrate on effective pathological tissue structures and providing reliable data guarantee for academic research based on the medical pathological images.
Drawings
FIG. 1 is a schematic structural diagram of a functional module according to the present invention;
FIG. 2 is a schematic diagram of a digital pathological image preprocessing module according to the present invention;
FIG. 3 is a flow chart of steps of digital pathology map preprocessing performed by the digital pathology map preprocessing module of the present invention;
FIG. 4 is a flowchart of the tissue section quantity compliance detection procedure of the digital pathology map preprocessing module according to the present invention;
FIG. 5 is a schematic structural diagram of a stain detection module according to the present invention;
FIG. 6 is a flow chart of the knife mark \ crack \ wrinkle \ laminated area detection steps of the stain detection module of the present invention;
FIG. 7 is a flow chart of the bubble \ glue solution \ manual labeling area detection steps of the stain detection module of the present invention;
FIG. 8 is a schematic structural diagram of a scan quality evaluation module according to the present invention;
FIG. 9 is a flowchart of the fuzzy area detection step of the scan quality assessment module of the present invention;
FIG. 10 is a flow chart of the staining quality detection step of the scanning quality assessment module of the present invention;
FIG. 11 is a schematic structural diagram of a scoring module according to the present invention;
FIG. 12 is a flowchart of the final valid region calculation step of the scoring module of the present invention;
FIG. 13 is a flowchart of the quality scoring steps of the scoring module of the present invention.
Detailed Description
The invention will be further illustrated with reference to the following specific examples. It should be understood that these examples are for illustrative purposes only and are not intended to limit the scope of the present invention. Further, it should be understood that various changes or modifications of the present invention may be made by those skilled in the art after reading the teaching of the present invention, and such equivalents may fall within the scope of the present invention as defined in the appended claims.
As shown in fig. 1, the digital pathological image quality evaluation system disclosed by the present invention includes a digital pathological image preprocessing module, a stain detection module, a scanning quality evaluation module, and a scoring module, wherein:
and the digital pathological image preprocessing module is used for resampling the digital pathological image, generating an image size suitable for system detection, detecting whether the number of the tissue sections is in compliance or not, and giving a detection score. As shown in fig. 2, the digital pathological image preprocessing module includes a digital pathological image preprocessing unit and a tissue section number compliance detecting unit.
As shown in fig. 3, the implementation of the digital pathology map preprocessing unit includes the following steps:
the first step is as follows: inputting a digital pathological diagram with original size, analyzing the maximum acquisition multiplying power of the digital pathological diagram, combining the preprocessing target multiplying power to obtain a resampling ratio,
the second step is that: and resampling the original-size digital pathology map based on the resampling ratio to obtain a global preprocessing map.
As shown in fig. 4, the implementation of the tissue section quantity compliance detection unit includes the following steps:
the first step is as follows: inputting a global preprocessing chart obtained by a digital pathological chart preprocessing unit, identifying a tissue part area, and acquiring the coordinate distribution of the tissue part area;
the second step is that: cutting the tissue part according to the coordinates of the tissue part area to obtain a cutting picture;
the third step: counting the number N of the available tissue sections, giving the score of the available sections,
and the stain detection module is used for detecting a cutter mark/crack/wrinkle/lamination area and a bubble/glue solution/manual labeling area and giving detection scores of all items. As shown in fig. 5, the stain detection module includes a knife mark \ crack \ wrinkle \ laminated area detection unit and a bubble \ glue solution \ manual labeling area detection unit.
As shown in fig. 6, the implementation of the tool crack \ wrinkle \ laminated region detection unit includes the following steps:
the first step is as follows: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
the second step is that: performing convolution on the original tissue area by using a convolution kernel with the size of 10 multiplied by 10 and the filling value of 0.01 to obtain a convolved image;
the third step: calculating the difference value between the convolved image and the original tissue area image to obtain a difference value image;
the fourth step: performing opening operation on the difference image, taking intersection of the calculation result and the original tissue area image, and obtaining a knife mark \ crack \ wrinkle \ lamination problem area image;
the fifth step: taking a difference set between the original tissue area and the knife mark \ crack \ wrinkle \ lamination problem area image to obtain an effective tissue area image;
and a sixth step: and calculating the ratio r of the effective tissue area image to the original tissue area image, and giving a score of a knife mark \ crack \ wrinkle \ laminated area, wherein the score of the knife mark \ crack \ wrinkle \ laminated area is 100 multiplied by r.
As shown in fig. 7, the implementation of the bubble \ glue solution \ manual labeling area detection unit includes the following steps:
the first step is as follows: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
the second step is that: frangi filtering is carried out on the original tissue area image to obtain a bubble \ glue solution \ manual labeling edge outline image;
the third step: after the bubble \ glue solution \ manual labeling edge contour image is filled, intersecting with the original tissue area image to obtain a bubble \ glue solution \ manual labeling problem area image;
the fourth step: taking a difference set between the original tissue area image and the bubble \ glue solution \ manual labeling problem area image to obtain an effective tissue area image;
the fifth step: and calculating the ratio r of the effective tissue area image to the original tissue area image, and giving a bubble \ glue solution \ manual labeling area score which is 100 multiplied by r.
And the scanning quality evaluation module is used for detecting the fuzzy area, detecting the dyeing quality and giving out each item detection score. As shown in fig. 8, the scan quality evaluation module includes a blurred region detection unit and a staining quality detection unit.
As shown in fig. 9, the implementation of the blur area detection unit includes the steps of:
the first step is as follows: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
the second step is that: carrying out laplacian filtering on the original tissue area image, calculating laplacian gradient, and obtaining a fuzzy thermal dispersion distribution diagram;
the third step: filtering fuzzy area distribution larger than a threshold value by using a threshold value on the fuzzy thermal dispersion point distribution diagram, and carrying out gaussian filtering on the fuzzy area distribution to obtain a fuzzy area mask;
the fourth step: taking a difference set between the original tissue area image and the fuzzy area mask to obtain an effective tissue area image;
the fifth step: and calculating the ratio r of the effective tissue area image to the original tissue area image, and giving a fuzzy area detection score which is 100 multiplied by r.
As shown in fig. 10, the implementation of the dyeing quality detecting unit includes the following steps:
the first step is as follows: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
the second step is that: calculating RGB pixel distribution of the original tissue region image, and obtaining a distribution histogram and an average value of each channel in R, G, B channels;
the third step: calculating a data difference value of the R channel and the B channel to obtain dyeing red and blue contrast data;
the fourth step: judging the red and blue channel contrast data according to the contrast difference threshold value, giving a dyeing quality detection score, counting RGB pixel distribution histograms of all pathological images in a pathological database, and taking the standard deviation of the global R channel and B channel as a dyeing quality standard Wherein R isstdAnd BstdIs the standard deviation, R, of the R and B channels in the databasestdAnd bstdTo detect the standard deviation of the R channel and the B channel in the image.
And the scoring module is used for calculating a final effective area, and simultaneously collecting all the subentry detection scores given by the digital pathological image preprocessing module, the stain detection module and the scanning quality evaluation module to give overall quality evaluation of the pathological image.
As shown in fig. 11, the scoring module includes a final effective region calculating unit and a quality scoring unit.
As shown in fig. 12, the implementation of the final effective area calculation unit includes the steps of:
the first step is as follows: summarizing effective tissue area images output by a cutter mark/crack/wrinkle/laminated area detection unit, a bubble/glue solution/manual labeling area detection unit and a fuzzy area detection unit;
the second step is that: and taking intersection of the effective tissue area images to obtain a final effective tissue area image.
As shown in fig. 13, the implementation of the quality scoring unit includes the following steps:
the first step is as follows: summarizing available section scoring, cutter mark \ crack \ wrinkle \ laminated region scoring, bubble \ glue solution \ manual labeling region scoring, fuzzy region detection scoring and dyeing quality detection scoring;
the second step is that: weighted average is taken from all the scores obtained in the first step, and final quality scores are obtained;
the third step: and according to the score threshold, giving a final good/bad judgment result and giving a final effective tissue area image.
Claims (5)
1. A digital pathology image quality evaluation system, comprising:
the digital pathological image preprocessing module is used for resampling the digital pathological image, generating an image size suitable for system detection, detecting whether the number of the tissue sections is in compliance or not, and giving a detection score;
the stain detection module is used for detecting a cutter mark/crack/wrinkle/lamination area and a bubble/glue solution/manual labeling area and giving detection scores of all items;
the scanning quality evaluation module is used for detecting the fuzzy area, detecting the dyeing quality and giving out each item detection score;
and the scoring module is used for calculating a final effective area, and simultaneously collecting all the subentry detection scores given by the digital pathological image preprocessing module, the stain detection module and the scanning quality evaluation module to give overall quality evaluation of the pathological image.
2. The digital pathological image quality evaluation system according to claim 1, wherein the digital pathological image preprocessing module comprises a digital pathological image preprocessing unit and a tissue section quantity compliance detection unit, wherein:
the implementation of the digital pathological diagram preprocessing unit comprises the following steps:
step 101: inputting a digital pathological diagram with original size, analyzing the maximum acquisition multiplying power of the digital pathological diagram, combining the preprocessing target multiplying power to obtain a resampling ratio,
step 102: resampling the original-size digital pathology map based on the resampling ratio to obtain a global preprocessing map; the implementation of the tissue section quantity compliance detection unit comprises the following steps:
step 201: inputting a global preprocessing chart obtained by a digital pathological chart preprocessing unit, identifying a tissue part area, and acquiring the coordinate distribution of the tissue part area;
step 202: cutting the tissue part according to the coordinates of the tissue part area to obtain a cutting picture;
3. the digital pathological image quality evaluation system of claim 2, wherein the stain detection module comprises a knife mark \ crack \ wrinkle \ laminated region detection unit and a bubble \ glue solution \ manual labeling region detection unit, wherein:
the implementation of the cutter mark \ crack \ wrinkle \ laminated region detection unit comprises the following steps:
step 301: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
step 302: performing convolution on the original tissue area by using a convolution kernel with the size of 10 multiplied by 10 and the filling value of 0.01 to obtain a convolved image;
step 303: calculating the difference value between the convolved image and the original tissue area image to obtain a difference value image;
step 304: performing opening operation on the difference image, taking intersection of the calculation result and the original tissue area image, and obtaining a knife mark \ crack \ wrinkle \ lamination problem area image;
step 305: taking a difference set between the original tissue area and the knife mark \ crack \ wrinkle \ lamination problem area image to obtain an effective tissue area image;
step 306: calculating the ratio r of the effective tissue area image to the original tissue area image, and giving a score of a knife mark \ crack \ wrinkle \ laminated area, wherein the score of the knife mark \ crack \ wrinkle \ laminated area is 100 multiplied by r;
the realization of the bubble \ glue solution \ manual labeling area detection unit comprises the following steps:
step 401: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
step 402: frangi filtering is carried out on the original tissue area image to obtain a bubble \ glue solution \ manual labeling edge outline image;
step 403: after the bubble \ glue solution \ manual labeling edge contour image is filled, intersecting with the original tissue area image to obtain a bubble \ glue solution \ manual labeling problem area image;
step 404: taking a difference set between the original tissue area image and the bubble \ glue solution \ manual labeling problem area image to obtain an effective tissue area image;
step 405: and calculating the ratio r of the effective tissue area image to the original tissue area image, and giving a bubble \ glue solution \ manual labeling area score which is 100 multiplied by r.
4. The digital pathology image quality evaluation system of claim 3, wherein the scan quality assessment module comprises a blurred region detection unit and a staining quality detection unit, wherein:
the implementation of the fuzzy area detection unit comprises the following steps:
step 501: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
step 502: carrying out laplacian filtering on the original tissue area image, calculating laplacian gradient, and obtaining a fuzzy thermal dispersion distribution diagram;
step 503: filtering fuzzy area distribution larger than a threshold value by using a threshold value on the fuzzy thermal dispersion point distribution diagram, and carrying out gaussian filtering on the fuzzy area distribution to obtain a fuzzy area mask;
step 504: taking a difference set between the original tissue area image and the fuzzy area mask to obtain an effective tissue area image;
step 505: calculating the ratio r of the effective tissue area image to the original tissue area image, and giving a fuzzy area detection score which is 100 multiplied by r;
the realization of the dyeing quality detection unit comprises the following steps:
step 601: inputting a global preprocessing image obtained by a digital pathological image preprocessing unit, extracting a tissue region, and binarizing to obtain an original tissue region image;
step 602: calculating RGB pixel distribution of the original tissue region image, and obtaining a distribution histogram and an average value of each channel in R, G, B channels;
step 603: calculating a data difference value of the R channel and the B channel to obtain dyeing red and blue contrast data;
step 604: judging the red and blue channel contrast data according to the contrast difference threshold value, giving a dyeing quality detection score, counting RGB pixel distribution histograms of all pathological images in a pathological database, and taking the standard deviation of the global R channel and B channel as a dyeing quality standard Wherein R isstdAnd BstdIs the standard deviation, R, of the R and B channels in the databasestdAnd bstdTo detect the standard deviation of the r and b channels in the image.
5. The digital pathology image quality evaluation system of claim 4, wherein the scoring module comprises a final effective region calculation unit and a quality scoring unit, wherein:
the implementation of the final effective area calculation unit comprises the following steps:
step 701: summarizing effective tissue area images output by a cutter mark/crack/wrinkle/laminated area detection unit, a bubble/glue solution/manual labeling area detection unit and a fuzzy area detection unit;
step 702: and taking intersection of the effective tissue area images to obtain a final effective tissue area image.
The realization of the quality scoring unit comprises the following steps:
step 801: summarizing available section scoring, cutter mark \ crack \ wrinkle \ laminated region scoring, bubble \ glue solution \ manual labeling region scoring, fuzzy region detection scoring and dyeing quality detection scoring;
step 802: weighted average is taken from all the scores obtained in the step 801 to obtain the final quality score;
step 803: and according to the score threshold, giving a final good/bad judgment result and giving a final effective tissue area image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010703756.8A CN111986157B (en) | 2020-07-21 | 2020-07-21 | Digital pathological image quality evaluation system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010703756.8A CN111986157B (en) | 2020-07-21 | 2020-07-21 | Digital pathological image quality evaluation system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111986157A true CN111986157A (en) | 2020-11-24 |
CN111986157B CN111986157B (en) | 2024-02-09 |
Family
ID=73439309
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010703756.8A Active CN111986157B (en) | 2020-07-21 | 2020-07-21 | Digital pathological image quality evaluation system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111986157B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113888529A (en) * | 2021-10-26 | 2022-01-04 | 济南超级计算技术研究院 | Pathological section image quality rating method and system based on deep learning |
CN113962975A (en) * | 2021-01-20 | 2022-01-21 | 赛维森(广州)医疗科技服务有限公司 | System for carrying out quality evaluation on pathological slide digital image based on gradient information |
CN114511559A (en) * | 2022-04-18 | 2022-05-17 | 杭州迪英加科技有限公司 | Multidimensional evaluation method, system and medium for quality of stained nasal polyp pathological section |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103175834A (en) * | 2013-01-28 | 2013-06-26 | 宁波江丰生物信息技术有限公司 | Method and system for evaluating quality of digital pathological section |
WO2016189065A1 (en) * | 2015-05-26 | 2016-12-01 | Ventana Medical Systems, Inc. | Method and system for assessing stain quality for in-situ hybridization and immunohistochemistry |
US20170294010A1 (en) * | 2016-04-12 | 2017-10-12 | Adobe Systems Incorporated | Utilizing deep learning for rating aesthetics of digital images |
KR101789513B1 (en) * | 2016-07-11 | 2017-10-26 | 주식회사 인피니트헬스케어 | Method of determining image quality in digital pathology system |
CN107945156A (en) * | 2017-11-14 | 2018-04-20 | 宁波江丰生物信息技术有限公司 | A kind of method of automatic Evaluation numeral pathology scan image image quality |
CN109191457A (en) * | 2018-09-21 | 2019-01-11 | 中国人民解放军总医院 | A kind of pathological image quality validation recognition methods |
WO2019047949A1 (en) * | 2017-09-08 | 2019-03-14 | 众安信息技术服务有限公司 | Image quality evaluation method and image quality evaluation system |
CN110738658A (en) * | 2019-12-21 | 2020-01-31 | 杭州迪英加科技有限公司 | Image quality evaluation method |
CN110807759A (en) * | 2019-09-16 | 2020-02-18 | 幻想动力(上海)文化传播有限公司 | Method and device for evaluating photo quality, electronic equipment and readable storage medium |
CN111105407A (en) * | 2019-12-25 | 2020-05-05 | 广州金域医学检验中心有限公司 | Pathological section staining quality evaluation method, device, equipment and storage medium |
-
2020
- 2020-07-21 CN CN202010703756.8A patent/CN111986157B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103175834A (en) * | 2013-01-28 | 2013-06-26 | 宁波江丰生物信息技术有限公司 | Method and system for evaluating quality of digital pathological section |
WO2016189065A1 (en) * | 2015-05-26 | 2016-12-01 | Ventana Medical Systems, Inc. | Method and system for assessing stain quality for in-situ hybridization and immunohistochemistry |
US20170294010A1 (en) * | 2016-04-12 | 2017-10-12 | Adobe Systems Incorporated | Utilizing deep learning for rating aesthetics of digital images |
KR101789513B1 (en) * | 2016-07-11 | 2017-10-26 | 주식회사 인피니트헬스케어 | Method of determining image quality in digital pathology system |
WO2019047949A1 (en) * | 2017-09-08 | 2019-03-14 | 众安信息技术服务有限公司 | Image quality evaluation method and image quality evaluation system |
CN107945156A (en) * | 2017-11-14 | 2018-04-20 | 宁波江丰生物信息技术有限公司 | A kind of method of automatic Evaluation numeral pathology scan image image quality |
CN109191457A (en) * | 2018-09-21 | 2019-01-11 | 中国人民解放军总医院 | A kind of pathological image quality validation recognition methods |
CN110807759A (en) * | 2019-09-16 | 2020-02-18 | 幻想动力(上海)文化传播有限公司 | Method and device for evaluating photo quality, electronic equipment and readable storage medium |
CN110738658A (en) * | 2019-12-21 | 2020-01-31 | 杭州迪英加科技有限公司 | Image quality evaluation method |
CN111105407A (en) * | 2019-12-25 | 2020-05-05 | 广州金域医学检验中心有限公司 | Pathological section staining quality evaluation method, device, equipment and storage medium |
Non-Patent Citations (4)
Title |
---|
MAHDI S. HOSSEINI等: "Focus Quqlity Assessment of High-Throughput Whole Slide Imaging in Digital Pathology", IEEE TRANSACTIONS ON MEDICAL IMAGING, vol. 39, no. 1, pages 62 - 74 * |
周超等: "基于深度卷积网络和结合策略的乳腺组织病理图像细胞核异型性自动评分", 中国生物医学工程学报, no. 3, pages 23 - 30 * |
王超: "低剂量CT图像质量评价方法及其性能的研究", 中国优秀硕士学位论文全文数据库 信息科技辑, no. 10, pages 138 - 277 * |
袁媛: "基于深度卷积网络的图像质量评价方法研究", 中国博士学位论文全文数据库 信息科技辑, no. 1, pages 138 - 128 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113962975A (en) * | 2021-01-20 | 2022-01-21 | 赛维森(广州)医疗科技服务有限公司 | System for carrying out quality evaluation on pathological slide digital image based on gradient information |
CN113962976A (en) * | 2021-01-20 | 2022-01-21 | 赛维森(广州)医疗科技服务有限公司 | Quality evaluation method for pathological slide digital image |
CN113962975B (en) * | 2021-01-20 | 2022-09-13 | 赛维森(广州)医疗科技服务有限公司 | System for carrying out quality evaluation on pathological slide digital image based on gradient information |
CN113962976B (en) * | 2021-01-20 | 2022-09-16 | 赛维森(广州)医疗科技服务有限公司 | Quality evaluation method for pathological slide digital image |
CN113888529A (en) * | 2021-10-26 | 2022-01-04 | 济南超级计算技术研究院 | Pathological section image quality rating method and system based on deep learning |
CN114511559A (en) * | 2022-04-18 | 2022-05-17 | 杭州迪英加科技有限公司 | Multidimensional evaluation method, system and medium for quality of stained nasal polyp pathological section |
Also Published As
Publication number | Publication date |
---|---|
CN111986157B (en) | 2024-02-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112508826B (en) | Printed matter defect detection method | |
CN111650220B (en) | Vision-based image-text defect detection method | |
CN107316077B (en) | Automatic adipose cell counting method based on image segmentation and edge detection | |
JP7076698B2 (en) | Image analysis method, image analysis device, program, learned deep learning algorithm manufacturing method and learned deep learning algorithm | |
CN111986157B (en) | Digital pathological image quality evaluation system | |
CN110210448B (en) | Intelligent face skin aging degree identification and evaluation method | |
CN111986190A (en) | Printed matter defect detection method and device based on artifact elimination | |
CN109671068B (en) | Abdominal muscle labeling method and device based on deep learning | |
CN108615239B (en) | Tongue image segmentation method based on threshold technology and gray level projection | |
CN109993099A (en) | A kind of lane line drawing recognition methods based on machine vision | |
CN110189383B (en) | Traditional Chinese medicine tongue color and fur color quantitative analysis method based on machine learning | |
CN107749049B (en) | Vein distribution display method and device | |
CN112381840B (en) | Method and system for marking vehicle appearance parts in loss assessment video | |
CN115546605A (en) | Training method and device based on image labeling and segmentation model | |
CN113393454A (en) | Method and device for segmenting pathological target examples in biopsy tissues | |
CN111325754A (en) | Automatic lumbar vertebra positioning method based on CT sequence image | |
CN103852034A (en) | Elevator guide rail perpendicularity detection method | |
CN115170518A (en) | Cell detection method and system based on deep learning and machine vision | |
Lezoray et al. | Segmentation of cytological image using color and mathematical morphology | |
CN114926635B (en) | Target segmentation method in multi-focus image combined with deep learning method | |
Kunwar et al. | Malaria detection using image processing and machine learning | |
CN110543802A (en) | Method and device for identifying left eye and right eye in fundus image | |
CN109658382B (en) | Tongue positioning method based on image clustering and gray projection | |
CN109934215B (en) | Identification card identification method | |
CN114187241A (en) | Pleural line identification method and system based on lung ultrasound |
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 |