CN111738997A - Method for calculating new coronary pneumonia lesion area ratio based on deep learning - Google Patents
Method for calculating new coronary pneumonia lesion area ratio based on deep learning Download PDFInfo
- Publication number
- CN111738997A CN111738997A CN202010531416.1A CN202010531416A CN111738997A CN 111738997 A CN111738997 A CN 111738997A CN 202010531416 A CN202010531416 A CN 202010531416A CN 111738997 A CN111738997 A CN 111738997A
- Authority
- CN
- China
- Prior art keywords
- mask
- lung
- calculating
- lesion
- region
- 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.)
- Pending
Links
- 230000003902 lesion Effects 0.000 title claims abstract description 46
- 238000000034 method Methods 0.000 title claims abstract description 23
- 206010035664 Pneumonia Diseases 0.000 title claims abstract description 19
- 238000013135 deep learning Methods 0.000 title claims abstract description 12
- 210000004072 lung Anatomy 0.000 claims abstract description 41
- 238000012549 training Methods 0.000 claims abstract description 12
- 238000013136 deep learning model Methods 0.000 claims abstract description 7
- 238000012545 processing Methods 0.000 claims abstract description 5
- 238000010606 normalization Methods 0.000 claims abstract description 4
- 210000004872 soft tissue Anatomy 0.000 claims description 13
- 238000001914 filtration Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000003672 processing method Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 abstract description 2
- 230000036285 pathological change Effects 0.000 description 4
- 231100000915 pathological change Toxicity 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 238000004445 quantitative analysis Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/30061—Lung
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Geometry (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method for calculating the proportion of new coronary pneumonia lesion areas based on deep learning, which belongs to the technical field of lung measurement and comprises the following steps: carrying out normalization processing on the original CT image set so as to adapt to data input of a deep learning model; respectively inputting the CT image data in the training set into two network learning models of 2DUnet and 2.5DUnet, and predicting a binary mask of a lung lesion region and a binary mask of the whole lung region; calculating the similarity between the binary mask predicted by the training set and the real label mask, and selecting an optimal network learning model; and predicting a lung lesion region mask and a lung whole region mask by using the optimal network learning model for the CT images in the training set, and calculating the ratio of the lesion region mask to the whole region mask. The invention automatically segments the lesion area of the lung and the effective mask of the whole lung by utilizing the deep learning technology, thereby quickly and accurately calculating the volume ratio of the lesion area.
Description
Technical Field
The invention relates to the technical field of lung measurement, in particular to a method for calculating the ratio of new coronary pneumonia lesion areas based on deep learning.
Background
In the process of treating the new coronary pneumonia, the CT image plays an important role in diagnosing the new coronary pneumonia, a patient regularly shoots the CT image and observes the development change of the disease condition, a unified measurement standard is not provided in the quantitative analysis of the diseased region of the patient at present, the diseased trend of the patient cannot be quantitatively analyzed, errors are easily caused by the adoption of manual observation of the change of the diseased region, a large amount of manpower is wasted, and the efficiency of the quantitative analysis is reduced.
Disclosure of Invention
The invention aims to solve the problems that quantitative analysis of the pneumonia lesion area lacks a measuring standard, has large error and low efficiency, and provides a method for calculating the proportion of the new coronary pneumonia lesion area based on deep learning.
In order to achieve the purpose, the invention adopts the technical scheme that: a method for calculating the proportion of new coronary pneumonia lesion areas based on deep learning comprises the following steps:
carrying out normalization processing on the original CT image set so as to adapt to data input of a deep learning model;
respectively inputting the CT image data in the training set into two network learning models of 2DUnet and 2.5DUnet, and predicting a binary mask of a lung lesion region and a binary mask of the whole lung region;
calculating the similarity between the binary mask predicted by the training set and the real label mask, and selecting an optimal network learning model;
and predicting a lung lesion region mask and a lung whole region mask by using the optimal network learning model for the CT images in the training set, and calculating the ratio of the lesion region mask to the whole region mask.
Preferably, the original CT image processing method includes:
a. converting the Hu value of the original CT image into an interval of [ -1200,600 ];
b. hu values less than-1200 and greater than 600 are set to-1200 and 600, respectively;
c. and normalizing the Hu value matrix of the interval to [0,255] to adapt to the input of deep learning model data.
Preferably, the real label mask adopted by the network learning model comprises a lesion area label mask and a total area label mask, wherein:
using two thresholds of-750 and-200 to perform lung segmentation, subtracting a mask obtained by-750 from a mask obtained by-200, and taking the final mask as a label mask of a lesion area;
the entire lung area is manually labeled as a full area label mask.
Preferably, the binary mask of the lung lesion region predicted by the network learning model is a binary mask for filtering normal soft tissues, and the filtering method includes:
dividing the generated binary mask into different blocks through a connected domain, judging a lesion region mask and a normal soft tissue mask contained in each block mask by using a two-class model based on resnet18, and finally filtering the normal soft tissue mask.
Preferably, the similarity between the binary mask and the real label mask is calculated by using a Dice coefficient for selecting a network learning model with the highest similarity, and the Dice coefficient calculation formula is as follows:
wherein, A represents a binary mask predicted by using a network learning model, and B is a real label mask.
Preferably, the ratio PoIR calculation formula between the lesion area mask and the whole area mask is as follows:
where volume of fed regions represents the predicted binary mask for the lung lesion region and volume of act Lung represents the predicted binary mask for the entire lung region.
Compared with the prior art, the invention has the following beneficial effects:
the optimal network learning model is selected, the lung pathological change area and the effective mask of the whole lung are automatically segmented by utilizing the deep learning technology, a two-classification model is designed when the mask of the pathological change area is generated, whether a single connected domain is the effective mask of the pathological change area or not is judged, the normal soft tissue mask is filtered, and the efficiency and the accuracy of the mask segmentation of the pathological change area are greatly improved.
Drawings
Fig. 1 is a flow chart of the pneumonia lesion area volume ratio method of the present invention.
FIG. 2 is a flow chart of the present invention for filtering out normal soft tissue masks.
Detailed Description
In order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further described with the specific embodiments.
As shown in fig. 1, a method for calculating the new coronary pneumonia lesion area ratio based on deep learning includes the following steps:
and carrying out normalization processing on the original CT image set so as to adapt to data input of the deep learning model. Respectively inputting the CT image data in the training set into two network learning models of 2DUnet and 2.5DUnet, and predicting a binary mask of a lung lesion region and a binary mask of the whole lung region, wherein the two network learning models are predicted by the following methods: firstly, 2, inputting a single image by DUnet, and inputting a binary mask with the same size by output; secondly, in order to reduce the training scale, the three-dimensional characteristics of the CT images are utilized, so that 2.5DUnet is utilized, the 2.5DUnet is input into three continuous images, and three corresponding binary images are output. And comparing the similarity between the predicted binary mask and the real label mask, and selecting an optimal network learning model. And predicting a lung lesion region mask and a lung whole region mask by using the optimal network learning model for the CT images in the training set, and calculating the ratio of the lesion region mask to the whole region mask.
The method can quickly and accurately calculate the volume ratio of the lesion area, so that after multiple CT images of the same case in different periods are input into the calculation method, the lesion area ratio of each period can be generated, a comparison curve is generated through changed data, the development trend of the patient condition is further measured, a reliable basis is provided for clinical diagnosis, and meanwhile, the quantitative accurate analysis can be performed on the weight and the development trend of the patient by combining data such as clinical blood oxygen indexes.
The original CT image processing method comprises the following steps: a. converting the Hu value of the original CT image into an interval of [ -1200,600 ]; b. hu values less than-1200 and greater than 600 are set to-1200 and 600, respectively; c. and normalizing the Hu value matrix of the interval to [0,255] to adapt to the input of deep learning model data.
The real label mask adopted by the network learning model comprises a lesion area label mask and a whole area label mask, wherein the lung segmentation is carried out by using two thresholds of-750 and-200, the mask obtained by subtracting-750 from the mask obtained by-200 is used as the lesion area label mask; the entire lung area is manually labeled as a full area label mask. The similarity between the binary mask and the real label mask is calculated by adopting a Dice coefficient for selecting a network learning model with the highest similarity, and the Dice coefficient calculation formula is as follows:
the method comprises the following steps that A represents a binary mask predicted by using a network learning model, B is a real label mask, the real label mask can be used as a reference object of two network learning models, which prediction mask in the two network learning models is the highest in similarity with the real mask can be judged according to a Dice coefficient, therefore, the optimal network learning model is selected as the prediction model of the method, errors can be avoided, accuracy evaluation can be carried out after calculation of the volume fraction, for example, the correlation degree of the prediction mask and the real mask is evaluated by adopting a Pearson correlation coefficient, and the accuracy of the pneumonia volume fraction of the prediction mask and the pneumonia volume fraction of the real mask is evaluated by utilizing an average absolute percentage error, wherein the Pearson correlation coefficient formula is as follows:
wherein xi and yi respectively represent the mask predicted by the selected network learning model and the label mask of true;
the average absolute percentage error formula is:
wherein PoIRpredictedAnd PoIRground-truthRespectively representing the mask fraction and the true label fraction predicted using the model of the invention.
As shown in fig. 2, the binary mask of the lung lesion region predicted by the network learning model is a binary mask for filtering normal soft tissue, and the filtering method includes: dividing the generated binary mask into different blocks through a connected domain, judging a lesion area mask and a normal soft tissue mask contained in each block mask by using a binary model based on resnet18, and finally filtering the normal soft tissue mask, wherein the connected domain refers to an area which has the same pixel value and is formed by pixels adjacent in position in an image, the connected domain analysis refers to finding out mutually independent connected domains in the image and marking the mutually independent connected domains, for example, marking normal soft tissue in a lung lesion area, and extracting different connected domains in the image is a common method in image processing, and the normal soft tissue mask can be filtered according to the binary model based on resnet18, so that the accuracy of the lesion area mask is improved.
The proportion PoIR calculation formula between the lesion area mask and the whole area mask is as follows:
where volume of fed regions represents the predicted binary mask for the lung lesion region and volume of act Lung represents the predicted binary mask for the entire lung region.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description of the embodiments is for clarity only, and those skilled in the art should make the description as a whole, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (6)
1. A method for calculating the proportion of new coronary pneumonia lesion areas based on deep learning is characterized by comprising the following steps:
carrying out normalization processing on the original CT image set so as to adapt to data input of a deep learning model;
respectively inputting the CT image data in the training set into two network learning models of 2DUnet and 2.5DUnet, and predicting a binary mask of a lung lesion region and a binary mask of the whole lung region;
calculating the similarity between the binary mask predicted by the training set and the real label mask, and selecting an optimal network learning model;
and predicting a lung lesion region mask and a lung whole region mask by using the optimal network learning model for the CT images in the training set, and calculating the ratio of the lesion region mask to the whole region mask.
2. The method for calculating the volume proportion of new coronary pneumonia based on deep learning according to claim 1, wherein the original CT image processing method comprises:
a. converting the Hu value of the original CT image into an interval of [ -1200,600 ];
b. hu values less than-1200 and greater than 600 are set to-1200 and 600, respectively;
c. and normalizing the Hu value matrix of the interval to 0,255 so as to adapt to the input of deep learning model data.
3. The method for calculating the volume fraction of new coronary pneumonia according to claim 1, wherein the real label masks adopted by the network learning model comprise a lesion area label mask and a total area label mask, wherein:
using two thresholds of-750 and-200 to perform lung segmentation, subtracting a mask obtained by-750 from a mask obtained by-200, and taking the final mask as a label mask of a lesion area;
the entire lung area is manually labeled as a full area label mask.
4. The method for calculating the volume ratio of the new coronary pneumonia according to claim 1, wherein the binary mask of the lung lesion region predicted by the network learning model is a binary mask for filtering out normal soft tissue, and the filtering method comprises:
dividing the generated binary mask into different blocks through a connected domain, judging a lesion region mask and a normal soft tissue mask contained in each block mask by using a two-class model based on resnet18, and finally filtering the normal soft tissue mask.
5. The method for calculating the volume ratio of new coronary pneumonia based on deep learning according to claim 1, wherein the similarity between the binary mask and the real label mask is calculated by using a Dice coefficient for selecting the network learning model with the highest similarity, and the Dice coefficient calculation formula is as follows:
wherein, A represents a binary mask predicted by using a network learning model, and B is a real label mask.
6. The method for calculating the volume fraction of new coronary pneumonia according to claim 1, wherein the ratio between the lesion area mask and the whole area mask, PoIR, is calculated as follows:
where volume of fed regions represents the predicted binary mask for the lung lesion region and volume of intake lung Lung represents the predicted binary mask for the entire lung region.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010531416.1A CN111738997A (en) | 2020-06-11 | 2020-06-11 | Method for calculating new coronary pneumonia lesion area ratio based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010531416.1A CN111738997A (en) | 2020-06-11 | 2020-06-11 | Method for calculating new coronary pneumonia lesion area ratio based on deep learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111738997A true CN111738997A (en) | 2020-10-02 |
Family
ID=72650138
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010531416.1A Pending CN111738997A (en) | 2020-06-11 | 2020-06-11 | Method for calculating new coronary pneumonia lesion area ratio based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111738997A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112669925A (en) * | 2020-12-16 | 2021-04-16 | 华中科技大学同济医学院附属协和医院 | Report template for CT (computed tomography) reexamination of new coronary pneumonia and forming method |
CN112686898A (en) * | 2021-03-15 | 2021-04-20 | 四川大学 | Automatic radiotherapy target area segmentation method based on self-supervision learning |
CN113344887A (en) * | 2021-06-16 | 2021-09-03 | 南通大学 | Interstitial pneumonia assessment method based on deep learning and fuzzy logic |
CN114820571A (en) * | 2022-05-21 | 2022-07-29 | 东北林业大学 | Pneumonia fibrosis quantitative analysis method based on DLPE algorithm |
CN115115620A (en) * | 2022-08-23 | 2022-09-27 | 安徽中医药大学 | Pneumonia lesion simulation method and system based on deep learning |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110634132A (en) * | 2019-08-30 | 2019-12-31 | 浙江大学 | Method for automatically generating tuberculosis quantitative diagnosis report based on deep learning 3D CT image |
CN110782441A (en) * | 2019-10-22 | 2020-02-11 | 浙江大学 | DR image pulmonary tuberculosis intelligent segmentation and detection method based on deep learning |
-
2020
- 2020-06-11 CN CN202010531416.1A patent/CN111738997A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110634132A (en) * | 2019-08-30 | 2019-12-31 | 浙江大学 | Method for automatically generating tuberculosis quantitative diagnosis report based on deep learning 3D CT image |
CN110782441A (en) * | 2019-10-22 | 2020-02-11 | 浙江大学 | DR image pulmonary tuberculosis intelligent segmentation and detection method based on deep learning |
Non-Patent Citations (1)
Title |
---|
WEI WU等: "Deep learning to estimate the physical proportion of infected region of lung for COVID-19 pneumonia with CT image set", 《HTTPS://ARXIV.ORG/ABS/2006.05018》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112669925A (en) * | 2020-12-16 | 2021-04-16 | 华中科技大学同济医学院附属协和医院 | Report template for CT (computed tomography) reexamination of new coronary pneumonia and forming method |
CN112686898A (en) * | 2021-03-15 | 2021-04-20 | 四川大学 | Automatic radiotherapy target area segmentation method based on self-supervision learning |
CN113344887A (en) * | 2021-06-16 | 2021-09-03 | 南通大学 | Interstitial pneumonia assessment method based on deep learning and fuzzy logic |
CN114820571A (en) * | 2022-05-21 | 2022-07-29 | 东北林业大学 | Pneumonia fibrosis quantitative analysis method based on DLPE algorithm |
CN115115620A (en) * | 2022-08-23 | 2022-09-27 | 安徽中医药大学 | Pneumonia lesion simulation method and system based on deep learning |
CN115115620B (en) * | 2022-08-23 | 2022-12-13 | 安徽中医药大学 | Pneumonia lesion simulation method and system based on deep learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111738997A (en) | Method for calculating new coronary pneumonia lesion area ratio based on deep learning | |
CN110969626B (en) | Method for extracting hippocampus of human brain nuclear magnetic resonance image based on 3D neural network | |
CN111931811B (en) | Calculation method based on super-pixel image similarity | |
Pan et al. | Mitosis detection techniques in H&E stained breast cancer pathological images: A comprehensive review | |
CN111340770B (en) | Method for constructing cancer prognosis model by combining global weighted LBP (local binary pattern) and texture analysis | |
CN112263217B (en) | Improved convolutional neural network-based non-melanoma skin cancer pathological image lesion area detection method | |
CN112927187A (en) | Method for automatically identifying and positioning focal cortical dysplasia epileptic focus | |
CN113706492B (en) | Automatic lung parenchyma segmentation method based on chest CT image | |
CN115496720A (en) | Gastrointestinal cancer pathological image segmentation method based on ViT mechanism model and related equipment | |
CN111724357B (en) | Arm bone density measurement method based on digital radiological image and support vector regression | |
CN113487568B (en) | Liver surface smoothness measuring method based on differential curvature | |
CN113657449A (en) | Traditional Chinese medicine tongue picture greasy classification method containing noise labeling data | |
CN112819765A (en) | Liver image processing method | |
CN113269799A (en) | Cervical cell segmentation method based on deep learning | |
CN117036288A (en) | Tumor subtype diagnosis method for full-slice pathological image | |
CN114549452A (en) | New coronary pneumonia CT image analysis method based on semi-supervised deep learning | |
CN104835155A (en) | Fractal-based early-stage breast cancer calcification point computer auxiliary detection method | |
CN107909577A (en) | Fuzzy C-mean algorithm continuous type max-flow min-cut brain tumor image partition method | |
CN113902738A (en) | Heart MRI segmentation method and system | |
CN116403211B (en) | Segmentation and clustering method and system based on single-cell pathology image cell nuclei | |
CN116228759B (en) | Computer-aided diagnosis system and apparatus for renal cell carcinoma type | |
CN112690815A (en) | System and method for assisting in diagnosing lesion grade based on lung image report | |
CN112348826A (en) | Interactive liver segmentation method based on geodesic distance and V-net | |
Dandan et al. | A multi-model organ segmentation method based on abdominal ultrasound image | |
CN115049605A (en) | Artificial intelligence processing and analyzing method for liver disease pathological tissue image |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201002 |