CN115409834B - Feature extraction method, system and storage medium for tendinopathy diagnosis - Google Patents
Feature extraction method, system and storage medium for tendinopathy diagnosis Download PDFInfo
- Publication number
- CN115409834B CN115409834B CN202211341602.4A CN202211341602A CN115409834B CN 115409834 B CN115409834 B CN 115409834B CN 202211341602 A CN202211341602 A CN 202211341602A CN 115409834 B CN115409834 B CN 115409834B
- Authority
- CN
- China
- Prior art keywords
- diagnosis
- tendinopathy
- achilles
- machine learning
- feature extraction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06T2207/20081—Training; Learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Radiology & Medical Imaging (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Multimedia (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Quality & Reliability (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The invention belongs to the field of medical image analysis and diagnosis, and particularly relates to a feature extraction method, a feature extraction system and a storage medium for achilles tendinopathy diagnosis. The feature extraction method for diagnosing the achilles tendinopathy comprises the following steps of: step 1, inputting a musculoskeletal ultrasonic image, and segmenting a musculoskeletal region of interest; step 2, extracting and quantizing the iconography characteristics from the R, G and B single-channel image modes and the gray image mode respectively; and 3, determining the characteristics for inputting the machine learning model to diagnose the achilles tendinopathy based on the accuracy of the machine learning model to diagnose the achilles tendinopathy. The invention further provides a system for diagnosing the achilles tendinopathy by using the characteristics obtained by the characteristic extraction method. The invention obtains the achilles tendon disease diagnosis model with better predictive performance, can more accurately carry out intelligent diagnosis on the achilles tendon disease and has good application prospect.
Description
Technical Field
The invention belongs to the field of medical image analysis and diagnosis, and particularly relates to a feature extraction method, a feature extraction system and a storage medium for tendinopathy diagnosis.
Background
The achilles tendon is the largest tendon in humans, and achilles tendinopathy is usually accompanied by pain, swelling and limited function of the achilles tendon and surrounding areas, unlike achilles tendonitis, which does not necessarily have inflammatory cells present, but rather is a chronic, long-term degeneration of tendon tissue, which is essentially a degeneration of collagen fibers. In general, some high-strength athletes, such as skiers, are more prone to tendinopathy because they typically exercise in cold conditions, stiff ankles, vasoconstriction, which increases the risk of tendinopathy. Sports medical experts at home and abroad recommend the Doppler ultrasonic examination of the Achilles tendon as the 'gold standard' for diagnosing the Achilles tendon disease.
The ultrasonic manifestations of achilles tendinopathy include thickening of the achilles tendon, vascular hyperplasia, echogenic attenuation, local calcification, increased fatty pad echogenicity, thickening of the achilles tendon aponeurosis, etc. Among them, echo is a commonly used parameter in ultrasonic examination, and echo of achilles tendinopathy is related to extracellular matrix composition and cell structure thereof. Currently, the AIUM guidelines do not describe how to assess echogenicity in achilles tendon tissue. In defining the achilles tendon echo intensity, sonographers always rely on subjective visual judgment, which is influenced by various instrument settings, gains, depth ranges, and sonographer experience, resulting in different sonographers' ultrasound description and diagnosis of the same lesion.
Imaging omics, also known as radiology, has been successfully applied to diagnosis and treatment research of multiple parts and multiple diseases of human body as a novel medical image analysis technology. Clinical research and application based on ultrasound imaging omics are more and more available, but are still lacking in orthopedic related disease research.
The problem of difference in subjective judgment of sonographers can be effectively solved by diagnosing the indexes of the image group through methods such as machine learning and the like. However, in machine learning, what features are input for calculation is a key factor affecting the accuracy of diagnosis. Therefore, it is highly desirable to help the sonographer provide effective musculoskeletal ultrasound image markers to enable achilles tendinopathy identification and diagnosis based on musculoskeletal ultrasound images, to reduce the sonographer's clinical subjectivity, and to provide speed and accuracy of diagnosis. However, the related research techniques are still lacking.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a feature extraction method, a system and a storage medium for diagnosing achilles tendinopathy, aiming at realizing accurate identification and diagnosis of the achilles tendinopathy based on an ultrasonic image of a musculoskeletal bone by selecting features.
A feature extraction method for tendinopathy diagnosis comprises the following steps:
and 3, determining the characteristics for inputting the machine learning model to diagnose the achilles tendinopathy based on the accuracy of the machine learning model to diagnose the achilles tendinopathy.
Preferably, in step 1, the method for segmenting the muscle bone region of interest is manual segmentation or model segmentation.
Preferably, in step 3, the machine learning model adopts a logistic regression, a random forest or a support vector machine algorithm.
Preferably, in step 3, the finally determined characteristics for inputting into the machine learning model for achilles tendinopathy diagnosis include: elongation, major axis length, maximum two-dimensional diameter, high gray level emphasis, small area high gray level emphasis, maximum three-dimensional diameter, 90 percentile, maximum probability, roughness, and run percentage.
The present invention also provides a system for tendinopathy diagnosis comprising:
the data acquisition and storage module is used for acquiring and storing the musculoskeletal ultrasonic image;
the segmentation module is used for segmenting the interesting region of the musculoskeletal bone;
the multi-channel image chemico-logical characteristic extraction module is used for obtaining characteristics according to the characteristic extraction method;
and the achilles tendinopathy diagnosis module is used for inputting the characteristics into the machine learning model to obtain the diagnosis result of the achilles tendinopathy.
Preferably, in the segmentation module, the method for segmenting the region of interest of the musculoskeletal bone is manual segmentation or model segmentation.
Preferably, the machine learning model adopts a logistic regression, random forest or support vector machine algorithm.
Preferably, the features include: elongation, major axis length, maximum two-dimensional diameter, high gray level emphasis, small area high gray level emphasis, maximum three-dimensional diameter, 90 percentile, maximum probability, roughness, and run percentage.
The present invention also provides a computer-readable storage medium having stored thereon a computer program for implementing the above-described feature extraction method for tendinopathy diagnosis.
The invention takes the clinical requirement of ultrasonic diagnosis of the achilles tendinopathy as a starting point, and constructs an automatic diagnosis mode of the achilles tendinopathy with higher accuracy and clinical gain by acquiring and learning multichannel musculoskeletal ultrasonic imaging omics characteristic information and adopting a combination strategy of multiple characteristic screening and machine learning modeling. In addition, the invention can assist the sonographer to make the diagnosis of the achilles tendinopathy more accurately through the automatic diagnosis system of the achilles tendinopathy, makes up the problems of insufficient clinical subjectivity and diagnosis experience and the like, and is suitable for primary hospitals of communities and the sonographers with insufficient experience.
Obviously, many modifications, substitutions, and variations are possible in light of the above teachings of the invention, without departing from the basic technical spirit of the invention, as defined by the following claims.
The present invention will be described in further detail with reference to the following examples. This should not be understood as limiting the scope of the above-described subject matter of the present invention to the following examples. All the technologies realized based on the above contents of the present invention belong to the scope of the present invention.
Drawings
FIG. 1 is a schematic flow chart of example 1.
FIG. 2 is a comparison of diagnostic performance of the multiple feature screening and machine learning algorithm modeling in example 1 at four channels of R, G, B and grayscale; fig. 2 (a) shows the effect of the R channel in the training set, fig. 2 (B) shows the effect of the R channel in the test set, fig. 2 (c) shows the effect of the G channel in the training set, fig. 2 (d) shows the effect of the G channel in the test set, fig. 2 (e) shows the effect of the B channel in the training set, fig. 2 (f) shows the effect of the B channel in the test set, fig. 2 (G) shows the effect of the grayscale channel in the training set, and fig. 2 (h) shows the effect of the grayscale channel in the test set.
FIG. 3 is the main diagnostic performance of the final model in example 1, FIG. 3 (a) is the quintupling cross validation results, and FIG. 3 (b) is the ROC results on the training and testing set.
Fig. 4 is the performance of the final model in example 1, where fig. 4 (a) is the calibration curve on the training set, fig. 4 (b) is the calibration curve on the test set, fig. 4 (c) is the decision curve on the training set, and fig. 4 (d) is the decision curve on the test set.
Detailed Description
It should be noted that, in the embodiment, the algorithm of the steps of data acquisition, transmission, storage, processing, etc. which are not specifically described, as well as the hardware structure, circuit connection, etc. which are not specifically described, can be implemented by the contents disclosed in the prior art.
As shown in fig. 1, the present embodiment provides an automatic achilles tendinopathy diagnosis method based on musculoskeletal ultrasound imaging omics, which includes:
s1, collecting muscle bone ultrasonic image data, dividing the muscle bone ultrasonic image data into a training group and a testing group, and preprocessing the muscle bone ultrasonic image data;
s2, performing multi-channel imagery omics feature extraction and analysis;
s3, integrating multichannel image omics characteristics on the training set, implementing various machine learning modeling and comparison, and constructing a final achilles tendinopathy diagnosis model;
and S4, performing clinical performance evaluation of the achilles tendinopathy diagnosis model on the test group.
Further, in S1, the specific method further includes:
s11, collecting muscle bone ultrasonic image data of the achilles tendon disease patient and the healthy person respectively, and randomly dividing the muscle bone ultrasonic image data into a training group and a testing group according to a ratio of 4;
s12, manually segmenting a musculoskeletal region of interest on the ultrasonic image by a musculoskeletal sonographer with more than ten years of working experience;
and S13, decomposing the musculoskeletal ultrasonic image into three single-channel image modes of R, G and B and a gray level image mode.
Further, in S2, the specific method further includes:
s21, extracting and quantifying the characteristics of the image omics respectively based on the obtained four single-channel ultrasonic images and the manually segmented region of interest;
s22, the extracted image omics features comprise 14 shape features, 18 first-order statistical features, 73 second-order texture features and 728 high-order signal transformation features;
s23, performing correlation analysis on the image omics characteristics of each channel, screening out characteristics with obvious correlation with the achilles tendinopathy identification, and eliminating redundant characteristics.
Further, in S3, the specific method further includes:
s31, on the basis of the correlation image omics characteristics of each channel on a training set, adopting a combination strategy of multiple characteristic screening and machine learning modeling to determine the optimal image omics characteristics;
and S32, integrating the optimal image omics characteristics of the four channels, and constructing a final model.
Further, in S31, the specific method further includes:
s311, screening the characteristics of the image omics by adopting a combination strategy of three characteristic screening and three machine learning modeling, wherein the three characteristic screening algorithms comprise a minimum absolute shrinkage and selection operator (LASSO), a random forest and a recursive characteristic elimination algorithm based on a support vector machine, and the three machine learning modeling algorithms comprise a logistic regression algorithm, a random forest and a support vector machine algorithm;
s312, constructing an achilles tendinopathy diagnosis model through different combination strategies, and evaluating the diagnosis performance between models according to the area under the curve (AUC) of a Receiver Operating Characteristic (ROC) curve and a confusion matrix;
and S313, taking the selected image omics characteristics of the optimal diagnostic model under each channel as the optimal image omics characteristics of the final grouping.
Further, in S32, the specific method further includes:
s321, integrating the optimal image omics characteristics selected by the four channels, respectively implementing logistic regression, random forest and support vector machine algorithms, and constructing a machine learning diagnosis model;
and S322, selecting a final achilles tendinopathy diagnosis model according to the comparison between the ROC curve and the confusion matrix.
Further, in S4, the specific method further includes:
and S41, performing consistency analysis and clinical decision analysis of the order-preserving regression on the test groups, and effectively evaluating the diagnosis accuracy and clinical gain of the final model.
Fig. 2 shows a comparison of diagnostic performance for multiple feature screening and machine learning algorithm modeling in four channels, R, G, B and grayscale, respectively. Fig. 2 (a) shows the effect of the R channel in the training set, fig. 2 (B) shows the effect of the R channel in the test set, fig. 2 (c) shows the effect of the G channel in the training set, fig. 2 (d) shows the effect of the G channel in the test set, fig. 2 (e) shows the effect of the B channel in the training set, fig. 2 (f) shows the effect of the B channel in the test set, fig. 2 (G) shows the effect of the grayscale channel in the training set, and fig. 2 (h) shows the effect of the grayscale channel in the test set. The diagnostic model based on G-channel, modeled by random forest screening and support vector machine, was found to be optimal.
In this embodiment, the ultrasound imaging omics characteristics of the finally screened traditional Chinese medicine are shown in the following table:
the ROC curve and quintupling cross validation results for the diagnosis of achilles tendinopathy using the method of this example are shown in fig. 3; the calibration and decision curves on the training and test sets are shown in fig. 4. As can be seen from the figure, the achilles tendinopathy diagnosis method provided by the embodiment has good diagnosis accuracy and can bring greater clinical gain.
Example 2 Achilles tendinopathy diagnostic System
The embodiment provides an achilles tendon disease automatic diagnosis system based on musculoskeletal ultrasonography, which comprises:
the data acquisition and storage module is used for acquiring and storing the musculoskeletal ultrasonic image;
the segmentation module is used for segmenting the interesting region of the musculoskeletal bone;
the multi-channel image chemico-logical feature extraction module is used for obtaining features according to the feature extraction method of the embodiment 1;
and the achilles tendinopathy diagnosis module is used for inputting the characteristics into the machine learning model to obtain the diagnosis result of the achilles tendinopathy.
As can be seen from the above embodiments and experimental examples, by optimizing the feature extraction method and the model algorithm, the invention obtains the diagnosis model of the achilles tendinopathy with better predictive performance, can carry out intelligent diagnosis on the achilles tendinopathy more accurately, and has good application prospect.
Claims (3)
1. A system for diagnosing tendinopathy, comprising:
the data acquisition and storage module is used for acquiring and storing the musculoskeletal ultrasonic image;
the segmentation module is used for segmenting the muscle bone region of interest;
the multi-channel image chemico-logical feature extraction module is used for obtaining features;
the achilles tendinopathy diagnosis module is used for inputting the characteristics into the machine learning model to obtain the diagnosis result of the achilles tendinopathy;
the multichannel imaging chemics feature extraction module obtains features by the following steps:
step 1, inputting a musculoskeletal ultrasonic image, and segmenting a musculoskeletal region of interest;
step 2, extracting and quantifying the iconography characteristics from the R, G and B single-channel image modes and the gray level image mode respectively;
step 3, determining the characteristics for inputting the machine learning model to diagnose the achilles tendinopathy based on the accuracy of the machine learning model to diagnose the achilles tendinopathy;
in step 3, the finally determined characteristics for inputting the machine learning model to carry out achilles tendinopathy diagnosis comprise: elongation, major axis length, maximum two-dimensional diameter, high gray level emphasis, small area high gray level emphasis, maximum three-dimensional diameter, 90 percentile, maximum probability, roughness and operation percentage;
the method for constructing the machine learning model comprises the following steps:
s321, integrating the optimal image omics characteristics selected by the R, G, B and gray level channels, respectively implementing logistic regression, random forest and support vector machine algorithms, and constructing a machine learning diagnosis model;
and S322, selecting a final achilles tendinopathy diagnosis model according to the comparison between the ROC curve and the confusion matrix.
2. The system for tendinopathy diagnosis of claim 1, wherein: in the segmentation module, the method for segmenting the muscle bone interesting region is manual segmentation or model segmentation.
3. A computer-readable storage medium characterized by: on which a computer program for implementing a system for tendinopathy diagnosis as claimed in claim 1 or 2 is stored.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211341602.4A CN115409834B (en) | 2022-10-30 | 2022-10-30 | Feature extraction method, system and storage medium for tendinopathy diagnosis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211341602.4A CN115409834B (en) | 2022-10-30 | 2022-10-30 | Feature extraction method, system and storage medium for tendinopathy diagnosis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115409834A CN115409834A (en) | 2022-11-29 |
| CN115409834B true CN115409834B (en) | 2023-02-28 |
Family
ID=84168983
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211341602.4A Active CN115409834B (en) | 2022-10-30 | 2022-10-30 | Feature extraction method, system and storage medium for tendinopathy diagnosis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115409834B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101226155A (en) * | 2007-12-21 | 2008-07-23 | 中国人民解放军第八一医院 | Intelligentize lung cancer early cell pathological picture recognition processing method |
| CN109840554A (en) * | 2018-12-26 | 2019-06-04 | 贵州联科卫信科技有限公司 | A kind of Alzheimer's disease MRI image classification method based on SVM-RFE-MRMR algorithm |
| CN110264454A (en) * | 2019-06-19 | 2019-09-20 | 四川智动木牛智能科技有限公司 | Cervical cancer tissues pathological image diagnostic method based on more hidden layer condition random fields |
| CN110895817A (en) * | 2019-11-01 | 2020-03-20 | 复旦大学 | MRI image hepatic fibrosis automatic grading method based on image omics analysis |
| CN111814868A (en) * | 2020-07-03 | 2020-10-23 | 苏州动影信息科技有限公司 | Model based on image omics feature selection, construction method and application |
| CN111860586A (en) * | 2020-06-12 | 2020-10-30 | 南通大学 | Three-stage identification method for fine-grained cervical cell image |
| CN112560911A (en) * | 2020-12-04 | 2021-03-26 | 中国科学院苏州生物医学工程技术研究所 | Tongue image classification method and system in traditional Chinese medicine |
| CN112598661A (en) * | 2020-12-29 | 2021-04-02 | 河北工业大学 | Ankle fracture and ligament injury diagnosis method based on machine learning |
| CN112992346A (en) * | 2021-04-09 | 2021-06-18 | 中山大学附属第三医院（中山大学肝脏病医院） | Method for establishing prediction model for prognosis of severe spinal cord injury |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140233826A1 (en) * | 2011-09-27 | 2014-08-21 | Board Of Regents Of The University Of Texas System | Systems and methods for automated screening and prognosis of cancer from whole-slide biopsy images |
| CN110706220B (en) * | 2019-09-27 | 2023-04-18 | 贵州大学 | Capsule endoscope image processing and analyzing method |
| CN112183237A (en) * | 2020-09-10 | 2021-01-05 | 天津大学 | Automatic white blood cell classification method based on color space adaptive threshold segmentation |
| CN113111874A (en) * | 2021-04-01 | 2021-07-13 | 上海市第一人民医院 | Rapid classification method for lateral images of lumbar vertebrae |
-
2022
- 2022-10-30 CN CN202211341602.4A patent/CN115409834B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101226155A (en) * | 2007-12-21 | 2008-07-23 | 中国人民解放军第八一医院 | Intelligentize lung cancer early cell pathological picture recognition processing method |
| CN109840554A (en) * | 2018-12-26 | 2019-06-04 | 贵州联科卫信科技有限公司 | A kind of Alzheimer's disease MRI image classification method based on SVM-RFE-MRMR algorithm |
| CN110264454A (en) * | 2019-06-19 | 2019-09-20 | 四川智动木牛智能科技有限公司 | Cervical cancer tissues pathological image diagnostic method based on more hidden layer condition random fields |
| CN110895817A (en) * | 2019-11-01 | 2020-03-20 | 复旦大学 | MRI image hepatic fibrosis automatic grading method based on image omics analysis |
| CN111860586A (en) * | 2020-06-12 | 2020-10-30 | 南通大学 | Three-stage identification method for fine-grained cervical cell image |
| CN111814868A (en) * | 2020-07-03 | 2020-10-23 | 苏州动影信息科技有限公司 | Model based on image omics feature selection, construction method and application |
| CN112560911A (en) * | 2020-12-04 | 2021-03-26 | 中国科学院苏州生物医学工程技术研究所 | Tongue image classification method and system in traditional Chinese medicine |
| CN112598661A (en) * | 2020-12-29 | 2021-04-02 | 河北工业大学 | Ankle fracture and ligament injury diagnosis method based on machine learning |
| CN112992346A (en) * | 2021-04-09 | 2021-06-18 | 中山大学附属第三医院（中山大学肝脏病医院） | Method for establishing prediction model for prognosis of severe spinal cord injury |
Non-Patent Citations (3)
| Title |
|---|
| Automated design for recognition of blood cells diseases from hematopathology using classical features selection and ELM;Muhammad Attique Khan等;《Microsc Res Tech》;20201231;第202-216页 * |
| 基于细胞核特征的宫颈癌细胞图像的识别与分类;宁梓淯等;《自动化与仪器仪表》;20161025(第10期);第197-199页 * |
| 多特征聚类与粘连分离模型的细胞抹片图像分割与分类;王品等;《生物医学工程学杂志》;20170825(第04期);第614-621页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115409834A (en) | 2022-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110010219B (en) | Intelligent detection system and detection method for retinopathy by optical coherence tomography | |
| CN101626726B (en) | Identification and analysis of lesions in medical imaging | |
| CN112070119A (en) | Ultrasonic tangent plane image quality control method and device and computer equipment | |
| JP2016531709A (en) | Image analysis technology for diagnosing disease | |
| CN109948671B (en) | Image classification method, device, storage medium and endoscopic imaging equipment | |
| CN110946619B (en) | Intravascular ultrasound automatic image histology analysis system and analysis method | |
| CN108766555A (en) | The computer diagnosis method and system of Pancreatic Neuroendocrine Tumors grade malignancy | |
| CN106530294A (en) | Method for carrying out processing on meibomian gland image to obtain gland parameter information | |
| CN112819818B (en) | Image recognition module training method and device | |
| CN111368827A (en) | Medical image processing method, medical image processing device, computer equipment and storage medium | |
| CN111528918B (en) | Tumor volume change trend graph generation device after ablation, equipment and storage medium | |
| CN115409834B (en) | Feature extraction method, system and storage medium for tendinopathy diagnosis | |
| Goudarzi et al. | Segmentation of arm ultrasound images in breast cancer-related lymphedema: A database and deep learning algorithm | |
| CN110827255A (en) | Plaque stability prediction method and system based on coronary artery CT image | |
| CN116433620A (en) | CT image-based bone mineral density prediction and osteoporosis intelligent screening method and system | |
| CN110890139A (en) | Intelligent pediatric surgical wound cleaning and drainage control system and method | |
| CN114864095A (en) | Analysis method for blood circulation change of narrow coronary artery under combination of multiple exercise strengths | |
| US20080107317A1 (en) | Method and apparatus for calculating coarseness level in target organ from ultrasonic image | |
| CN111481233B (en) | Thickness measuring method for transparent layer of fetal cervical item | |
| Marcomini et al. | Quantitative evaluation of automatic methods for lesions detection in breast ultrasound images | |
| Zhang et al. | Medical image fusion based a densely connected convolutional networks | |
| CN111265234A (en) | Method and system for judging properties of lung mediastinal lymph nodes | |
| CN111951219A (en) | Thyroid eye disease screening method, system and equipment based on orbit CT image | |
| CN117036302B (en) | Method and system for determining calcification degree of aortic valve | |
| Fatema et al. | Development of an automated optimal distance feature-based decision system for diagnosing knee osteoarthritis using segmented X-ray 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |