CN116485758A - Method, system, electronic equipment and medium for determining number of nodules - Google Patents
Method, system, electronic equipment and medium for determining number of nodules Download PDFInfo
- Publication number
- CN116485758A CN116485758A CN202310457382.XA CN202310457382A CN116485758A CN 116485758 A CN116485758 A CN 116485758A CN 202310457382 A CN202310457382 A CN 202310457382A CN 116485758 A CN116485758 A CN 116485758A
- Authority
- CN
- China
- Prior art keywords
- nodule
- interest
- region
- determined
- current
- 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
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000013528 artificial neural network Methods 0.000 claims abstract description 36
- 238000001514 detection method Methods 0.000 claims abstract description 31
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 11
- 238000004590 computer program Methods 0.000 claims description 9
- 238000002604 ultrasonography Methods 0.000 claims description 2
- 210000000056 organ Anatomy 0.000 description 3
- 206010028980 Neoplasm Diseases 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 206010061218 Inflammation Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000001363 autoimmune Effects 0.000 description 1
- 210000000481 breast Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 208000035269 cancer or benign tumor Diseases 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000010224 classification analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 210000004907 gland Anatomy 0.000 description 1
- 210000002216 heart Anatomy 0.000 description 1
- 230000004054 inflammatory process Effects 0.000 description 1
- 210000003734 kidney Anatomy 0.000 description 1
- 210000004185 liver Anatomy 0.000 description 1
- 210000004072 lung Anatomy 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 230000009885 systemic effect Effects 0.000 description 1
- 210000001685 thyroid gland Anatomy 0.000 description 1
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- 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
- G06T2207/10132—Ultrasound image
-
- 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
- G06T2207/30096—Tumor; Lesion
Abstract
The invention discloses a method, a system, electronic equipment and a medium for determining the number of nodules, which relate to the field of determining the number of nodules, and the method comprises the following steps: acquiring a plurality of target ultrasonic images; positioning the nodules in each target ultrasonic image by using a target detection algorithm to obtain the coordinates of a nodule detection frame corresponding to the target ultrasonic image; determining a region of interest in the corresponding target ultrasonic image according to the nodule detection frame coordinates; determining a current confidence level of the nodule in the current region of interest by utilizing the twin neural network; and determining that the nodule in the current region of interest is a determined nodule or a new nodule according to the current confidence, until all the regions of interest are input into the twin neural network, and determining the number of nodules. The invention improves the accuracy of the determination of the number of the junctions.
Description
Technical Field
The present invention relates to the field of nodule number determination, and in particular, to a method, a system, an electronic device, and a medium for determining the number of nodules.
Background
Nodules are bumps formed by abnormal aggregation of cells, and as a systemic disease, can affect any organ, such as thyroid, breast, liver, kidney, heart lung, etc., or glands, and are clinically common conditions, the signs and symptoms of which depend on the organ in question. Clinically, it can be caused by various causes, such as inflammation, autoimmune reaction, neoplasm, even cancer, etc., which can be represented as nodules, single or multiple.
The existing common method for determining the number of the nodules is ultrasonic detection, ultrasonic waves are adopted to scan the organs of a patient, then a doctor manually judges an ultrasonic image through naked eyes to identify whether the nodules exist in the image and determine the number of the nodules, but in different frames of the same image, the same nodules are possibly different in form, whether the same nodules are the same nodules or not is judged by virtue of memory and experience, time and labor are wasted, and the doctor can easily mistakenly consider two different nodules or mistakenly consider the different nodules as the same nodules, so that the calculation of the number of the nodules is wrong.
Disclosure of Invention
The invention aims to provide a method, a system, electronic equipment and a medium for determining the number of nodules, which can accurately determine the number of nodules.
In order to achieve the above object, the present invention provides the following solutions:
a method of nodule number determination, comprising:
acquiring a plurality of target ultrasonic images; the target ultrasonic image is an ultrasonic image of the number of the nodules to be determined;
for any one target ultrasonic image:
positioning a nodule in the target ultrasonic image by using a target detection algorithm to obtain a nodule detection frame coordinate of the target ultrasonic image;
determining a region of interest in the target ultrasound image according to the nodule detection frame coordinates;
determining a plurality of current confidence levels of the nodule in the current region of interest using the twin neural network; the current confidence is determined from a current region of interest and a most recent region of interest of a currently determined nodule set; a determined nodule corresponds to a most recent region of interest;
determining a nodule corresponding to the first region of interest as a first determined nodule, and adding the first determined nodule into the determined nodule set; taking the region of interest where the first determined nodule is located as the latest region of interest of the first determined nodule;
determining that the nodule in the current region of interest is a determined nodule or a new nodule according to the current confidence levels until all the regions of interest are input to the twin neural network, thereby determining the number of nodules;
if the nodule in the current region of interest is a determined nodule, the current region of interest is taken as the latest region of interest of the corresponding determined nodule;
if the nodule in the current region of interest is a new nodule, adding the nodule in the current region of interest to the determined nodule set; and taking the current region of interest as the latest region of interest of the corresponding determined nodule.
Optionally, determining a plurality of current confidence levels of the nodule in the current region of interest using a twin neural network, specifically including:
performing black edge filling treatment on the current region of interest to obtain a treated region of interest;
utilizing the twin neural network, determining a plurality of current confidence levels of nodules in the processed region of interest.
Optionally, the determining process of any current confidence of the nodule in the processed region of interest specifically includes:
inputting the processed region of interest into a first EfficientNetV2B0 network of the twin neural network to obtain a first eigenvector;
inputting the latest interested area of any determined nodule into a second EfficientNetV2B0 network of the twin neural network to obtain a second eigenvector;
calculating Euclidean distance between the first feature vector and the second feature vector;
and determining any current confidence of the nodules in the processed region of interest according to the Euclidean distance.
Optionally, calculating the euclidean distance between the first feature vector and the second feature vector specifically includes:
by means ofCalculating Euclidean distance between the first feature vector and the second feature vector; wherein x is i An ith eigenvalue in the first eigenvector; y is i Is the i-th eigenvalue in the second eigenvector.
Optionally, determining the current confidence of the nodule in the region of interest after processing according to the euclidean distance specifically includes:
by means ofDetermining the current confidence of the nodule in the processed region of interest; wherein E is the Euclidean distance.
Optionally, determining the nodule in the current region of interest as the determined nodule or the new nodule according to the n current confidence degrees specifically includes:
judging whether the n current confidence degrees are all larger than a preset confidence degree or not;
if yes, determining the nodule in the current interested area as a new nodule;
if not, determining the nodule in the current region of interest as the determined nodule corresponding to the minimum current confidence.
A nodule number determination system comprising:
the image acquisition module is used for acquiring a plurality of target ultrasonic images; the target ultrasonic image is an ultrasonic image of the number of the nodules to be determined;
the nodule positioning module is used for positioning nodules in the target ultrasonic image by utilizing a target detection algorithm to obtain nodule detection frame coordinates of the target ultrasonic image;
the clipping module is used for determining a region of interest in the target ultrasonic image according to the nodule detection frame coordinates;
the confidence determining module is used for determining a plurality of current confidence degrees of the nodules in the current region of interest by utilizing the twin neural network; the current confidence is determined from a current region of interest and a most recent region of interest of a currently determined nodule set; a determined nodule corresponds to a most recent region of interest;
determining a nodule corresponding to the first region of interest as a first determined nodule, and adding the first determined nodule into the determined nodule set; taking the region of interest where the first determined nodule is located as the latest region of interest of the first determined nodule;
the node number determining module is used for determining whether the node in the current interested area is a determined node or a new node according to a plurality of the current confidence degrees; until all the regions of interest are input to the twin neural network, so as to determine the number of nodules;
if the nodule in the current region of interest is a determined nodule, the current region of interest is taken as the latest region of interest of the corresponding determined nodule;
if the nodule in the current region of interest is a new nodule, adding the nodule in the current region of interest to the determined nodule set; and taking the current region of interest as the latest region of interest of the corresponding determined nodule.
An electronic device, comprising: the system comprises a memory and a processor, wherein the memory is used for storing a computer program, and the processor runs the computer program to enable the electronic equipment to execute the nodule number determination method.
A computer readable storage medium storing a computer program which when executed by a processor implements the nodule number determination method described above.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
according to the method, the system, the electronic equipment and the medium for determining the number of the nodules, the nodules in each target ultrasonic image are respectively positioned by utilizing a target detection algorithm, each target ultrasonic image is cut according to the coordinates of a nodule detection frame, the interested area of each nodule is determined, the current confidence coefficient of the nodule in the current interested area is determined by utilizing a twin neural network, and the number of the nodules is determined according to the current confidence coefficient.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for determining the number of nodules provided by the invention;
FIG. 2 is a flow chart of a twin neural network process provided by the invention;
FIG. 3 is a schematic view of a region of interest provided by the present invention;
fig. 4 is a schematic diagram of a twin neural network according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention aims to provide a method, a system, electronic equipment and a medium for determining the number of nodules, which can accurately determine the number of nodules.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1
According to the output result of the target detection algorithm in the process of analyzing the ultrasonic image, the invention analyzes and classifies the nodules appearing on different pictures based on the twin neural network, classifies the images of the same nodule on different pictures of different sections together, and classifies the section images of different nodules separately.
As shown in fig. 1, the method for determining the number of nodules provided by the invention comprises the following steps:
step 101: acquiring a plurality of target ultrasonic images; the target ultrasonic image is an ultrasonic image of the number of nodules to be determined.
For any one target ultrasonic image:
step 102: and positioning the nodule in the target ultrasonic image by using a target detection algorithm to obtain the nodule detection frame coordinates of the target ultrasonic image. In practical application, the scanning image (ultrasonic image) of the ultrasonic machine is transmitted to the target detection algorithm through the acquisition card. The target detection algorithm analyzes each frame of the ultrasonic image and transmits the coordinates of the positioned nodule detection frame and the image of the corresponding frame to the twin neural network.
Step 103: and determining a region of interest in the target ultrasonic image according to the nodule detection frame coordinates.
Step 104: determining a plurality of current confidence levels of the nodule in the current region of interest using the twin neural network; the current confidence is determined from a current region of interest and a most recent region of interest of a currently determined nodule set; a determined nodule corresponds to a most recent region of interest.
Determining a nodule corresponding to the first region of interest as a first determined nodule, and adding the first determined nodule into the determined nodule set; the region of interest where the first determined nodule is located is taken as the latest region of interest of the first determined nodule.
After the coordinates of the nodule detection frame of the current frame and the image of the corresponding frame are obtained, the first step is to judge whether the input is the first input, if so, the nodule detection frame is newly added and classified as the No. 1 nodule, and information is stored; if not, the next analysis and comparison are carried out.
In practical application, a twin neural network (siamese network) is used to analyze and compare the nodule image of the current frame with the previous nodule image, and classification is performed after a classification result is obtained.
As an alternative embodiment, step 104, as shown in fig. 2, specifically includes:
step 1041: and performing black edge filling processing on the current region of interest to obtain the processed region of interest.
In practical application, the present invention cuts out a Region of interest (ROI) according to the nodule detection frame coordinates of the current frame and the image of the corresponding frame, and adjusts the size to 64×64×3 by a black edge filling (Letterbox) method on the premise of maintaining the aspect ratio of the original ROI, as shown in fig. 3. And then comparing the processed ROI with the ROI of the latest frame of the existing n nodules one by one through a twin neural network.
Step 1042: utilizing the twin neural network, determining a plurality of current confidence levels of nodules in the processed region of interest. The determining process of any current confidence of the nodule in the processed region of interest specifically comprises the following steps:
step 10421: and inputting the processed region of interest into a first EfficientNetV2B0 network of the twin neural network to obtain a first eigenvector.
Step 10422: and inputting the latest region of interest of any determined nodule into a second EfficientNetV2B0 network of the twin neural network to obtain a second eigenvector.
In practical application, as shown in fig. 4, the inputs of the twin neural network are the ROI of the current frame and the ROI of the existing nodule, 2 eigenvectors (a first eigenvector and a second eigenvector) are obtained through 2 shared weighted afflicientnetv 2B0, euclidean distances (euclidean distances) of the 2 eigenvectors are calculated, and then a full connection layer with 1-layer neuron number of 1 is connected.
Step 10423: and calculating Euclidean distance between the first feature vector and the second feature vector.
In practical application, use is made ofCalculating Euclidean distance between the first feature vector and the second feature vector; wherein the method comprises the steps of，x i An ith eigenvalue in the first eigenvector; y is i Is the i-th eigenvalue in the second eigenvector.
Step 10424: and determining any current confidence of the nodules in the processed region of interest according to the Euclidean distance.
In practical application, use is made ofDetermining the current confidence of the nodule in the processed region of interest; wherein E is the Euclidean distance.
The twin neural network will output a confidence according to the Sigmoid function. The higher the confidence, the greater the representation alignment difference; the lower the confidence, the smaller the representation alignment difference and the higher the similarity.
Step 105: and determining that the nodule in the current region of interest is the determined nodule or new nodule according to the current confidence degrees until all the regions of interest are input into the twin neural network, so as to determine the number of nodules.
And if the nodule in the current region of interest is the determined nodule, taking the current region of interest as the latest region of interest of the corresponding determined nodule.
If the nodule in the current region of interest is a new nodule, adding the nodule in the current region of interest to the determined nodule set; and taking the current region of interest as the latest region of interest of the corresponding determined nodule.
As an alternative embodiment, step 105 specifically includes:
judging whether the n current confidence degrees are all larger than a preset confidence degree or not; if yes, determining the nodule in the current interested area as a new nodule; if not, determining the nodule in the current region of interest as the determined nodule corresponding to the minimum current confidence.
In practical application, if the confidence coefficient is less than 0.5 in the comparison result, classifying the ROI of the current frame into a node with the minimum confidence coefficient; if the comparison result has no confidence less than 0.5, 1 nodule is newly created, namely the ROI of the current frame is classified as the n+1 nodule.
Compared with the prior art, the method for determining the number of the nodules has the following advantages:
the original nodule proportion can be maintained by adjusting the size through a black edge filling method, distortion is avoided, and the similarity characteristics of the images can be extracted more effectively.
The twin neural network sharing weight can ensure that the vector features of two images extracted by the network are kept in a unified standard.
The invention has fast comparison and classification analysis speed and high precision, only needs 10-20 milliseconds for each analysis, and the precision can reach 99.9 percent.
The invention analyzes and classifies the nodules appearing on different pictures based on the twin neural network, classifies the images of different sections of the same nodule on different pictures together, and classifies the section images of different nodules separately so as to rapidly analyze the nodules and bring the nodules into a report.
Example two
In order to perform a corresponding method of the above embodiment to achieve the corresponding functions and technical effects, a nodule number determination system is provided below, including:
the image acquisition module is used for acquiring a plurality of target ultrasonic images; the target ultrasonic image is an ultrasonic image of the number of nodules to be determined.
And the nodule positioning module is used for positioning nodules in the target ultrasonic image by utilizing a target detection algorithm to obtain nodule detection frame coordinates of the target ultrasonic image.
And the clipping module is used for determining a region of interest in the target ultrasonic image according to the nodule detection frame coordinates.
The confidence determining module is used for determining a plurality of current confidence degrees of the nodules in the current region of interest by utilizing the twin neural network; the current confidence is determined from a current region of interest and a most recent region of interest of a currently determined nodule set; a determined nodule corresponds to a most recent region of interest.
Determining a nodule corresponding to the first region of interest as a first determined nodule, and adding the first determined nodule into the determined nodule set; the region of interest where the first determined nodule is located is taken as the latest region of interest of the first determined nodule.
The node number determining module is used for determining whether the node in the current interested area is a determined node or a new node according to a plurality of the current confidence degrees; until all regions of interest are input to the twin neural network, thereby determining the number of nodules.
And if the nodule in the current region of interest is the determined nodule, taking the current region of interest as the latest region of interest of the corresponding determined nodule.
If the nodule in the current region of interest is a new nodule, adding the nodule in the current region of interest to the determined nodule set; and taking the current region of interest as the latest region of interest of the corresponding determined nodule.
Example III
The present invention provides an electronic device, comprising: the apparatus includes a memory for storing a computer program, and a processor that runs the computer program to cause the electronic device to perform the nodule number determination method of embodiment one.
Example IV
The present invention provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the nodule number determination method of embodiment one.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the invention.
Claims (9)
1. A method for determining the number of nodules comprising:
acquiring a plurality of target ultrasonic images; the target ultrasonic image is an ultrasonic image of the number of the nodules to be determined;
for any one target ultrasonic image:
positioning a nodule in the target ultrasonic image by using a target detection algorithm to obtain a nodule detection frame coordinate of the target ultrasonic image;
determining a region of interest in the target ultrasound image according to the nodule detection frame coordinates;
determining a plurality of current confidence levels of the nodule in the current region of interest using the twin neural network; the current confidence is determined from a current region of interest and a most recent region of interest of a currently determined nodule set; a determined nodule corresponds to a most recent region of interest;
determining a nodule corresponding to the first region of interest as a first determined nodule, and adding the first determined nodule into the determined nodule set; taking the region of interest where the first determined nodule is located as the latest region of interest of the first determined nodule;
determining that the nodule in the current region of interest is a determined nodule or a new nodule according to the current confidence levels until all the regions of interest are input to the twin neural network, thereby determining the number of nodules;
if the nodule in the current region of interest is a determined nodule, the current region of interest is taken as the latest region of interest of the corresponding determined nodule;
if the nodule in the current region of interest is a new nodule, adding the nodule in the current region of interest to the determined nodule set; and taking the current region of interest as the latest region of interest of the corresponding determined nodule.
2. The nodule number determination method of claim 1, wherein determining a plurality of current confidence levels for nodules in a current region of interest using a twin neural network, in particular comprises:
performing black edge filling treatment on the current region of interest to obtain a treated region of interest;
utilizing the twin neural network, determining a plurality of current confidence levels of nodules in the processed region of interest.
3. The nodule number determination method of claim 2, wherein the determination of any current confidence of the nodule in the processed region of interest comprises:
inputting the processed region of interest into a first EfficientNetV2B0 network of the twin neural network to obtain a first eigenvector;
inputting the latest interested area of any determined nodule into a second EfficientNetV2B0 network of the twin neural network to obtain a second eigenvector;
calculating Euclidean distance between the first feature vector and the second feature vector;
and determining any current confidence of the nodules in the processed region of interest according to the Euclidean distance.
4. The nodule number determination method of claim 3, wherein calculating the euclidean distance of the first feature vector and the second feature vector comprises:
by means ofCalculating Euclidean distance between the first feature vector and the second feature vector; wherein x is i An ith eigenvalue in the first eigenvector; y is i Is the i-th eigenvalue in the second eigenvector.
5. A method of nodule number determination according to claim 3, wherein determining the current confidence of the nodule in the processed region of interest based on the euclidean distance comprises:
by means ofDetermining the current confidence of the nodule in the processed region of interest; wherein E is the Euclidean distance.
6. The method according to claim 1, wherein determining whether the nodule in the current region of interest is a determined nodule or a new nodule based on the n current confidence levels, specifically comprises:
judging whether the n current confidence degrees are all larger than a preset confidence degree or not;
if yes, determining the nodule in the current interested area as a new nodule;
if not, determining the nodule in the current region of interest as the determined nodule corresponding to the minimum current confidence.
7. A nodule number determination system, comprising:
the image acquisition module is used for acquiring a plurality of target ultrasonic images; the target ultrasonic image is an ultrasonic image of the number of the nodules to be determined;
the nodule positioning module is used for positioning nodules in the target ultrasonic image by utilizing a target detection algorithm to obtain nodule detection frame coordinates of the target ultrasonic image;
the clipping module is used for determining a region of interest in the target ultrasonic image according to the nodule detection frame coordinates;
the confidence determining module is used for determining a plurality of current confidence degrees of the nodules in the current region of interest by utilizing the twin neural network; the current confidence is determined from a current region of interest and a most recent region of interest of a currently determined nodule set; a determined nodule corresponds to a most recent region of interest;
determining a nodule corresponding to the first region of interest as a first determined nodule, and adding the first determined nodule into the determined nodule set; taking the region of interest where the first determined nodule is located as the latest region of interest of the first determined nodule;
the node number determining module is used for determining whether the node in the current interested area is a determined node or a new node according to a plurality of the current confidence degrees; until all the regions of interest are input to the twin neural network, so as to determine the number of nodules;
if the nodule in the current region of interest is a determined nodule, the current region of interest is taken as the latest region of interest of the corresponding determined nodule;
if the nodule in the current region of interest is a new nodule, adding the nodule in the current region of interest to the determined nodule set; and taking the current region of interest as the latest region of interest of the corresponding determined nodule.
8. An electronic device, comprising: a memory for storing a computer program, and a processor that runs the computer program to cause the electronic device to perform the nodule number determination method of any one of claims 1-6.
9. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, implements the nodule number determination method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310457382.XA CN116485758A (en) | 2023-04-25 | 2023-04-25 | Method, system, electronic equipment and medium for determining number of nodules |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310457382.XA CN116485758A (en) | 2023-04-25 | 2023-04-25 | Method, system, electronic equipment and medium for determining number of nodules |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116485758A true CN116485758A (en) | 2023-07-25 |
Family
ID=87221040
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310457382.XA Pending CN116485758A (en) | 2023-04-25 | 2023-04-25 | Method, system, electronic equipment and medium for determining number of nodules |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116485758A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7206462B1 (en) * | 2000-03-17 | 2007-04-17 | The General Hospital Corporation | Method and system for the detection, comparison and volumetric quantification of pulmonary nodules on medical computed tomography scans |
| CN107992826A (en) * | 2017-12-01 | 2018-05-04 | 广州优亿信息科技有限公司 | A kind of people stream detecting method based on the twin network of depth |
| CN111986225A (en) * | 2020-08-14 | 2020-11-24 | 山东大学 | Multi-target tracking method and device based on angular point detection and twin network |
| CN115705932A (en) * | 2021-08-04 | 2023-02-17 | 广州视源电子科技股份有限公司 | Lung nodule follow-up visit prediction method, device, equipment and storage medium |
-
2023
- 2023-04-25 CN CN202310457382.XA patent/CN116485758A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7206462B1 (en) * | 2000-03-17 | 2007-04-17 | The General Hospital Corporation | Method and system for the detection, comparison and volumetric quantification of pulmonary nodules on medical computed tomography scans |
| CN107992826A (en) * | 2017-12-01 | 2018-05-04 | 广州优亿信息科技有限公司 | A kind of people stream detecting method based on the twin network of depth |
| CN111986225A (en) * | 2020-08-14 | 2020-11-24 | 山东大学 | Multi-target tracking method and device based on angular point detection and twin network |
| CN115705932A (en) * | 2021-08-04 | 2023-02-17 | 广州视源电子科技股份有限公司 | Lung nodule follow-up visit prediction method, device, equipment and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| NA AN 等: "Multitarget Tracking Using Siamese Neural Networks", 《ACM TRANSACTIONS ON MULTIMEDIA COMPUTING，COMMUNICATIONS，AND APPLICATIONS》, vol. 17, no. 2, 18 May 2021 (2021-05-18), pages 1 - 16, XP059190797, DOI: 10.1145/3441656 * |
| XIANGQIONG WU 等: "CacheTrack-YOLO: Real-Time Detection and Tracking for Thyroid Nodules and Surrounding Tissues in Ultrasound Videos", 《IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMA TICS》, vol. 25, no. 10, 5 October 2021 (2021-10-05), pages 3812 - 3823, XP011881863, DOI: 10.1109/JBHI.2021.3084962 * |
| 沈文祥;秦品乐;曾建潮;: "基于多级特征和混合注意力机制的室内人群检测网络", 《计算机应用》, vol. 39, no. 12, 15 October 2019 (2019-10-15), pages 3496 - 3502 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210365717A1 (en) | Method and apparatus for segmenting a medical image, and storage medium | |
| CN109146892B (en) | Image clipping method and device based on aesthetics | |
| CN108898086B (en) | Video image processing method and device, computer readable medium and electronic equipment | |
| US6606406B1 (en) | System and method for progressive stereo matching of digital images | |
| US11501435B2 (en) | Unsupervised content-preserved domain adaptation method for multiple CT lung texture recognition | |
| CN110827398B (en) | Automatic semantic segmentation method for indoor three-dimensional point cloud based on deep neural network | |
| WO2021139324A1 (en) | Image recognition method and apparatus, computer-readable storage medium and electronic device | |
| CN110853011B (en) | Method for constructing convolutional neural network model for pulmonary nodule detection | |
| CN109685765B (en) | X-ray film pneumonia result prediction device based on convolutional neural network | |
| CN109165658B (en) | Strong negative sample underwater target detection method based on fast-RCNN | |
| CN114978037B (en) | Solar cell performance data monitoring method and system | |
| CN113989293A (en) | Image segmentation method and training method, device and equipment of related model | |
| US8077956B2 (en) | Orientation detection for chest radiographic images | |
| CN116485817A (en) | Image segmentation method, device, electronic equipment and storage medium | |
| CN114638800A (en) | Improved Faster-RCNN-based head shadow mark point positioning method | |
| CN113902932A (en) | Feature extraction method, visual positioning method and device, medium and electronic equipment | |
| CN116452697B (en) | Method and system for detecting secondary pulmonary tuberculosis focus | |
| CN116485758A (en) | Method, system, electronic equipment and medium for determining number of nodules | |
| CN116229236A (en) | Bacillus tuberculosis detection method based on improved YOLO v5 model | |
| CN116168328A (en) | Thyroid nodule ultrasonic inspection system and method | |
| WO2023035535A1 (en) | Training method and apparatus for semantic segmentation network, device, and storage medium | |
| CN115719335A (en) | Cerebrovascular image-label two-stage generation method, device and storage medium | |
| CN112750124B (en) | Model generation method, image segmentation method, model generation device, image segmentation device, electronic equipment and storage medium | |
| CN111291820B (en) | Target detection method combining positioning information and classification information | |
| Chhabra et al. | Comparison of different edge detection techniques to improve quality of medical images |
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 |