CN114052724B - Orthopedics traction abnormity detection system based on artificial intelligence - Google Patents
Orthopedics traction abnormity detection system based on artificial intelligence Download PDFInfo
- Publication number
- CN114052724B CN114052724B CN202210035001.4A CN202210035001A CN114052724B CN 114052724 B CN114052724 B CN 114052724B CN 202210035001 A CN202210035001 A CN 202210035001A CN 114052724 B CN114052724 B CN 114052724B
- Authority
- CN
- China
- Prior art keywords
- detected
- bandage
- affected limb
- diseased
- moment
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 39
- 230000000399 orthopedic effect Effects 0.000 title claims abstract description 39
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 14
- 230000033001 locomotion Effects 0.000 claims abstract description 97
- 201000010099 disease Diseases 0.000 claims abstract description 88
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 88
- 230000008859 change Effects 0.000 claims abstract description 71
- 230000002159 abnormal effect Effects 0.000 claims abstract description 43
- 238000004364 calculation method Methods 0.000 claims abstract description 23
- 230000005856 abnormality Effects 0.000 claims abstract description 16
- 238000000605 extraction Methods 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 4
- 239000013598 vector Substances 0.000 claims description 34
- 238000006073 displacement reaction Methods 0.000 claims description 30
- 208000016285 Movement disease Diseases 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 25
- 206010025482 malaise Diseases 0.000 abstract description 2
- 201000003152 motion sickness Diseases 0.000 abstract 1
- 210000003414 extremity Anatomy 0.000 description 174
- 230000008569 process Effects 0.000 description 11
- 238000011084 recovery Methods 0.000 description 9
- 230000000474 nursing effect Effects 0.000 description 4
- 238000001931 thermography Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 230000002087 whitening effect Effects 0.000 description 3
- 210000000988 bone and bone Anatomy 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000001575 pathological effect Effects 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 206010010264 Condition aggravated Diseases 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
- 230000017531 blood circulation Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 230000003750 conditioning effect Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000013036 cure process Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000035876 healing Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 210000003141 lower extremity Anatomy 0.000 description 1
- 230000002503 metabolic effect Effects 0.000 description 1
- 230000004060 metabolic process Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006386 neutralization reaction Methods 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000008961 swelling Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1118—Determining activity level
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/01—Measuring temperature of body parts ; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4866—Evaluating metabolism
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F5/00—Orthopaedic methods or devices for non-surgical treatment of bones or joints; Nursing devices; Anti-rape devices
- A61F5/01—Orthopaedic devices, e.g. splints, casts or braces
- A61F5/04—Devices for stretching or reducing fractured limbs; Devices for distractions; Splints
- A61F5/042—Devices for stretching or reducing fractured limbs; Devices for distractions; Splints for extension or stretching
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F5/00—Orthopaedic methods or devices for non-surgical treatment of bones or joints; Nursing devices; Anti-rape devices
- A61F5/37—Restraining devices for the body or for body parts, e.g. slings; Restraining shirts
- A61F5/3715—Restraining devices for the body or for body parts, e.g. slings; Restraining shirts for attaching the limbs to other parts of the body
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- 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
- G06T7/60—Analysis of geometric attributes
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Animal Behavior & Ethology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Heart & Thoracic Surgery (AREA)
- Biomedical Technology (AREA)
- Medical Informatics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Surgery (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Physiology (AREA)
- General Engineering & Computer Science (AREA)
- Nursing (AREA)
- Orthopedic Medicine & Surgery (AREA)
- Vascular Medicine (AREA)
- Geometry (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Obesity (AREA)
- Dentistry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention relates to the field of data processing, in particular to an orthopedic traction abnormity detection system based on artificial intelligence. The system comprises a feature extraction module, a bandage difference calculation module and an abnormal degree calculation module, wherein: a feature extraction module: the motion sickness characteristic and the loosening sickness characteristic are obtained; a bandage difference calculation module: the method is used for obtaining disease change characteristics of the to-be-detected affected limb under the condition that the affected limb moves and the bandage is loosened, and recording the disease change characteristics as first disease change characteristics; obtaining disease change characteristics of the to-be-detected affected limb under the condition that the affected limb moves and the bandage is not loosened, and recording the disease change characteristics as second disease change characteristics; calculating a difference between the first condition change characteristic and the second condition change characteristic; an abnormality degree calculation module: and the method is used for acquiring various types of predicted movement characteristics corresponding to the to-be-detected affected limb, and calculating the abnormal degree of traction at the current moment according to the difference and the various types of predicted movement characteristics. The invention improves the efficiency of detecting the abnormal traction of the orthopedics department.
Description
Technical Field
The invention relates to the field of data processing, in particular to an orthopedic traction abnormity detection system based on artificial intelligence.
Background
The orthopedic traction aims at restoring the fractured and dislocated limb structure, maintaining the stability after dislocation and correcting the spasmodic and distorted limb. When the patient is subjected to orthopedic traction, the patient needs to apply two forces in opposite directions to the affected limb by using a traction tool, and the affected limb needs to be fixed by using a bandage. Affected limb often needs medical personnel to regularly inspect and detect when orthopedics is pull, and whether the bandage on the in time inspection affected limb appears not hard up or the abnormal conditions such as whether each position of affected limb appears removing prevent to lead to affected limb disease aggravation because of the bandage is not hard up or the removal of affected limb. These traction abnormalities are sudden and require frequent patient examinations by medical personnel, which is inefficient and requires too much time and effort by medical personnel.
Disclosure of Invention
In order to solve the problem of low efficiency in acquiring abnormal situations of the traction of an affected limb in the existing method, the invention aims to provide an orthopedic traction abnormality detection system based on artificial intelligence, and the adopted technical scheme is as follows:
the invention provides an artificial intelligence-based orthopedic traction abnormity detection system, which comprises a feature extraction module, a bandage difference calculation module and an abnormity degree calculation module, wherein the feature extraction module comprises a first detection module, a second detection module and a third detection module, the first detection module comprises a first detection module, the second detection module comprises a second detection module, the third detection module comprises a third detection module, a fourth detection module and a fourth detection module, the third detection module comprises a fourth detection module, the fourth detection module comprises a fourth detection module and a fourth detection module, the fourth detection module comprises a feature extraction module, a bandage difference calculation module and an abnormity degree calculation module, the fourth detection module comprises a fourth detection module and an abnormity degree calculation module, the fourth detection module comprises a fourth detection module:
a feature extraction module: the three-dimensional information acquisition module is used for acquiring the movement characteristics of the to-be-detected affected limb and the loosening characteristics of the corresponding bandage at each moment according to the three-dimensional information of the to-be-detected affected limb and the three-dimensional information of the corresponding bandage at each moment; according to the disease characteristics of the limb to be detected at adjacent moments, the disease change characteristics of the limb to be detected at each moment are obtained; acquiring mobile diseased characteristics of the diseased limb to be detected when the bandage is not loosened under different moving conditions, and clustering to obtain mobile diseased characteristics of each category; acquiring loosening diseased characteristics of the affected limb to be detected when the bandage is loosened under different moving conditions, and clustering to obtain loosening diseased characteristics of each category; the movement diseased feature comprises a condition changing feature and a movement feature, and the loosening diseased feature comprises a condition changing feature and a loosening feature;
a bandage difference calculation module: the system is used for obtaining disease change characteristics of the to-be-detected affected limb under the conditions that the affected limb moves and the bandage is loosened according to the movement disease characteristics of each category and the loosening disease characteristics of each category, and recording the disease change characteristics as first disease change characteristics; according to the movement diseased characteristics of each category and the loosening diseased characteristics of each category, obtaining disease change characteristics of the diseased limb to be detected under the condition that the diseased limb moves and the bandage does not loosen, and recording the disease change characteristics as second disease change characteristics; calculating a difference between the first condition change characteristic and the second condition change characteristic;
an abnormality degree calculation module: and the method is used for acquiring various types of predicted movement characteristics corresponding to the to-be-detected affected limb, and calculating the abnormal degree of traction at the current moment according to the difference and the various types of predicted movement characteristics.
Preferably, the obtaining of the movement characteristic of the limb to be detected and the loosening characteristic of the corresponding bandage at each moment according to the three-dimensional information of the limb to be detected and the three-dimensional information of the corresponding bandage at each moment includes:
recording a set of three-dimensional coordinates of pixel points in the image of the affected limb to be detected at the previous moment corresponding to each moment as a first set of the affected limb, and recording a set of three-dimensional coordinates of pixel points in the image of the affected limb to be detected at each moment as a second set of the affected limb; matching the pixel points in the first affected limb set and the second affected limb set to obtain the displacement vector of each pixel point in the image of the affected limb to be detected at each moment; obtaining the movement characteristics of the to-be-detected affected limb at each moment according to the displacement vector;
recording a set of three-dimensional coordinates of pixel points in a bandage image at a previous moment corresponding to each moment as a first bandage set; recording a set of three-dimensional coordinates of each pixel point in the bandage image at each moment as a second bandage set; matching pixel points in the first bandage set and the second bandage set to obtain a displacement vector of each pixel point in each bandage image at each moment;
and processing the displacement vector of each pixel point in the image of the affected limb to be detected at each moment and the displacement vector of each pixel point in the bandage image by adopting a maximum mean difference algorithm to obtain the loosening characteristic of the bandage corresponding to the affected limb to be detected at each moment.
Preferably, the obtaining of the disease change characteristics of the affected limb to be detected at each moment according to the disease characteristics of the affected limb to be detected at adjacent moments includes: and (4) differentiating the disease characteristics of the limb to be detected at each moment from the disease characteristics of the limb to be detected at the previous moment to obtain the disease change characteristics of the limb to be detected at each moment.
Preferably, the obtaining of the predicted movement characteristics of each category corresponding to the to-be-detected affected limb includes:
judging Euclidean distances between the movement features of each historical moment of the limb to be detected and the movement features of the current moment, and constructing a first set of movement features according to the movement features of a plurality of historical moments of which the Euclidean distances are smaller than a preset threshold value;
and acquiring the movement characteristics of the next moment corresponding to each movement characteristic in the first set of movement characteristics, and clustering the movement characteristics of the next moment to obtain the predicted movement characteristics of each category corresponding to the to-be-detected affected limb.
Preferably, the following formula is used to calculate the abnormal degree of traction at the current moment:
wherein,
is the abnormal degree of the traction at the current moment,
in order to be a hyper-parameter,
the disease change degree of the affected limb to be detected at the current moment,
to move the number of categories of diseased features,
to loosen the number of categories of diseased features,
in order to predict the number of classes of moving features,
is the parameter of the disease state at the current moment,
to the extent of concern over a loose feature of the nth loose feature category,
the difference between the disease change characteristics of the affected limb to be detected only under the movement characteristic and the disease change characteristics under the conditions of the movement characteristic and the bandage loosening characteristic,
is the m-th moving diseased characteristic category corresponding to the diseased limb to be detected,
is the h-th predicted movement characteristic category corresponding to the affected limb to be detected,
is composed of
The movement characteristics and
the maximum mean difference of the moving features contained in (1).
Preferably, the mean shift algorithm is used for clustering the mobile diseased characteristics of the diseased limb to be detected when the bandage is not loosened under different mobile conditions to obtain the mobile diseased characteristics of each category.
Preferably, the mean shift algorithm is used for clustering loosening diseased features of the diseased limb to be detected when the bandage is loosened under different moving conditions to obtain loosening diseased features of various categories.
The invention has the following beneficial effects: according to the three-dimensional information of the to-be-detected affected limb and the three-dimensional information of the bandage at each moment, the moving condition of the affected limb and the loosening condition of the bandage are analyzed to obtain the abnormal degree of traction at the current moment. The invention can ensure that medical personnel can timely treat the affected limb by monitoring the abnormal traction state of the affected limb in real time, thereby enabling the patient to recover smoothly. The invention does not need to rely on medical staff to detect the abnormal degree of traction, and solves the problem of lower efficiency existing in the existing method of relying on medical staff to detect the abnormal degree of traction.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a schematic structural diagram of an artificial intelligence-based orthopedic traction abnormality detection system provided by the invention;
FIG. 2 shows the test results of the present invention.
Detailed Description
In order to further explain the technical means and effects of the present invention adopted to achieve the predetermined invention purpose, the following detailed description of an orthopedic traction abnormality detection system based on artificial intelligence according to the present invention will be made with reference to the accompanying drawings and preferred embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following describes a specific scheme of the orthopedic traction abnormality detection system based on artificial intelligence in detail with reference to the accompanying drawings.
Embodiment of orthopedics traction abnormity detection system based on artificial intelligence
When an orthopedic patient is in orthopedic traction, the affected limb needs to be kept braked to maintain a certain body position; however, the patient inevitably moves limbs during long-term orthopedic traction, which may cause the fractured end to be stressed and not to be stable, affect bone reduction and seriously deform the bone joints.
In addition, when the skeleton is pulled, the wounded limb pulling part of the patient needs to be wound by a bandage, on one hand, the fracture part of the wounded limb is fixed and protected, and on the other hand, a pulling appliance, such as a pulling bow or a pulling belt, is fixed, so that the pulling force and the counter-pulling force are maintained to be in the right direction; but because the bandage is not firm, is disturbed by the movement of the limbs of the patient, and the swelling of the affected limb is faded away, the bandage can be loosened or slipped, on one hand, the braking effect of the affected limb cannot be ensured, on the other hand, the direction of the traction force and the direction of the counter traction force can be changed, so that the orthopedic traction effect is poor, and the recovery progress is influenced. The above situations are abnormal states of the patient in orthopedic traction, and in order to ensure smooth rehabilitation of the patient, the embodiment provides an orthopedic traction abnormality detection system for timely detecting the abnormal states of the patient in orthopedic traction and timely informing medical staff to nurse the patient when abnormality occurs.
As shown in fig. 1, the system for detecting abnormal traction in orthopedics department of the present embodiment includes a feature extraction module, a bandage difference calculation module and an abnormal degree calculation module, which are described below.
I, characteristic extraction module
The feature extraction module of the embodiment is used for obtaining the movement feature of the to-be-detected affected limb at each moment and the loosening feature of the corresponding bandage according to the three-dimensional information of the to-be-detected affected limb at each moment and the three-dimensional information of the corresponding bandage; and obtaining the disease change characteristics of the diseased limb to be detected at each moment according to the disease characteristics of the diseased limb to be detected at adjacent moments.
The system for detecting the abnormal orthopedic traction is suitable for a scene that a patient performs orthopedic traction on an orthopedic traction sickbed, and the embodiment provides a mode for acquiring three-dimensional information of a to-be-detected limb and three-dimensional information of a corresponding bandage at each moment, and the method comprises the following steps: installing an RGBD camera and a thermal imaging camera on a traction sickbed to realize real-time monitoring of the affected limb; the RGBD camera can obtain RGB image data and depth information; the RGBD camera is used for acquiring three-dimensional information of each position on an affected limb of a patient to be detected and three-dimensional information of a bandage corresponding to the affected limb to be detected in real time, specifically, the RGBD camera is used for acquiring RGB images of the patient in real time, the RGB images acquired in real time are input into the semantic segmentation network, a semantic region and a bandage semantic region of the affected limb of the patient to be detected are obtained, and three-dimensional coordinates of each pixel point in the semantic region are acquired, so that the three-dimensional information of the affected limb to be detected and the three-dimensional information of the bandage can be acquired in real time. The semantic segmentation network may be deep labv3, which is a well-known technology and will not be described herein.
The process that the characteristic extraction module obtains the moving characteristic of the to-be-detected affected limb and the loosening characteristic of the corresponding bandage according to the three-dimensional information of the to-be-detected affected limb and the three-dimensional information of the corresponding bandage at each moment is as follows:
acquiring set of three-dimensional coordinates of all pixel points on affected limb of patient to be detected at t-1 moment
Acquiring a set of three-dimensional coordinates of all pixel points on the to-be-detected affected limb at the time t
. And matching the pixel points in the two sets by using a KM algorithm so as to ensure that the sum of Euclidean distances of all matched pixel point pairs is minimum. The displacement vector formed by the three-dimensional coordinates of one pixel point pair is regarded as the displacement of the same position on the affected limb at the adjacent moment, and the displacement vectors of all the pixel point pairs form a set
The displacement of the affected limb at different positions at the front and rear time points is shown.
Similarly, a set of three-dimensional coordinates of all pixel points on the bandage wound by the limb to be detected at the moment t-1 is obtained
Acquiring a set of three-dimensional coordinates of all pixel points on the bandage at the time t
. And matching the pixel points in the two sets by using a KM algorithm so as to ensure that the sum of Euclidean distances of all matched pixel point pairs is minimum. The displacement vectors formed by one pixel point pair are regarded as the displacements of the same position at different moments, and the displacement vectors of all the pixel point pairs form a set
The displacement of the bandage at different positions on the affected limb at the two moments before and after the affected limb is shown. If the bandage is not loosened, the bandage will move the same as the affected limb, and if the bandage is loosened, the bandage will not move the same as the affected limb.
The method for acquiring the movement characteristics of the affected limb comprises the following steps: and establishing a three-dimensional Gaussian model, wherein the mean value of the Gaussian model is the mean value of all the displacements in the set A1, and the covariance matrix is the covariance matrix of all the displacements in the set A1. The gaussian model is used to describe the displacement distribution in the set a1, and each displacement vector in the set a1 corresponds to a probability on the three-dimensional gaussian distribution, which is called the importance of each displacement vector, i.e. the more important the displacement vector is closer to the average displacement vector. In the embodiment, the importance is taken as the weight to perform weighted summation on all the displacement vectors in the set a1, and the obtained result is a vector, i.e. the movement characteristic of the affected limb, which is used to represent the movement condition of the affected limb to be detected at time t.
The method for acquiring the loosening degree of the bandage comprises the following steps: taking the displacement vector in the set A1 as a data sample, taking the displacement vector in the set A1 as a data sample, obtaining the distribution difference of the two data sample sets by using a maximum mean difference algorithm, wherein the result is a scalar and represents the difference between the bandage displacement and the affected limb displacement, and the result is taken as the loosening characteristic of the bandage at the time t.
Therefore, the moving characteristics of the affected limb to be detected and the loosening characteristics of the bandage can be obtained at any time.
The infrared image can reflect the metabolism of the affected limb or the heat generated by the blood, and can be used for reflecting the affected condition and the recovery condition of the affected limb of the patient. In order to detect the recovery condition of the patient during the treatment process, the present embodiment provides a way to obtain the disease characteristics of the to-be-detected affected limb at each moment, as follows: acquiring an infrared image of a diseased limb of a patient to be detected by using a thermal imaging camera, enabling the thermal imaging camera and an RGBD (red, green and blue) camera to have the same visual field, acquiring the infrared image of the limb by using the thermal imaging camera, and inputting the infrared image into a new semantic segmentation network to acquire a semantic region of the diseased limb; and then obtaining gray values of all pixel points in the affected semantic area on the infrared image, wherein the gray values are used for representing the heat generated by different positions of the affected limb, when the affected limb has symptoms such as unsmooth blood flow and the like, the metabolic capacity or the generated heat of the positions can be changed, and the affected condition of the patient is evaluated by utilizing the heat of each position of the affected limb. And acquiring a gradient vector of each pixel point on the infrared image by using a Sobel operator, wherein the gradient vector is used for expressing the gradient amplitude and the gradient direction of each pixel point. And performing principal component analysis on gradient vectors of all pixel points on the infrared image, wherein the vectors are 2-dimensional, so that two principal component directions can be obtained, each principal component direction is a unit vector and corresponds to a characteristic value, the characteristic value represents the projection variance of the gradient vector of the pixel point in each principal component direction, the principal component direction with the maximum characteristic value and the two characteristic values are combined into one vector, the vector is used for describing the distribution condition of pixel gray values on the infrared image, the distribution condition is used for representing the diseased condition of a patient, and the vector is recorded as the disease characteristic of the diseased limb to be detected.
The process that the characteristic extraction module obtains the disease change characteristics of the diseased limb to be detected at each moment according to the disease characteristics of the diseased limb to be detected at adjacent moments is as follows: the affected limb to be detected at the time of t-1Is characterized by
And the disease characteristics of the affected limb to be detected at the moment t are recorded as
,
Will be
Is called the disease change characteristic of the affected limb to be detected at the time t. According to the gray distribution characteristics of the periphery of each pixel point on the infrared images at any two moments, the disease change characteristics of each pixel point at any moment are obtained, and according to the disease change characteristics of each pixel point at any moment, the disease change characteristics of the to-be-detected affected limb at any moment are obtained and are used for describing the disease change condition of the affected limb in the orthopedic traction process.
In the process of orthopedic traction of an affected limb, the disease condition of a wound needs to be changed slowly and healed slowly, but a patient inevitably moves the affected limb, the disease condition of the wound may be changed, the recovery of the affected limb is not affected if the patient is light, the disease condition of the affected limb is aggravated if the patient is heavy, and in order to judge whether the affected limb heals gradually or deteriorates gradually in the process of orthopedic traction, the disease condition change condition of the affected limb when the affected limb moves needs to be detected.
The feature extraction module of the embodiment is further configured to obtain moving diseased features of the to-be-detected diseased limb when the bandage is not loosened under different moving conditions, and obtain moving diseased features of each category after clustering; the movement-diseased features include condition-changing features and movement features.
Specifically, the present embodiment obtains the disease change characteristics of the affected limb to be detected at each moment, and supposing that the current moment is t, the disease change characteristics are
Then the degree of change of the condition is
，
Is composed of
The L2 norm of (a), which is used to indicate whether the disease characteristics of the affected limb have changed greatly at the front and back time points; the larger the value is, the movement of the patient at the current moment is likely to have great interference and influence on the original healing process, so that the medical staff is very necessary to carry out nursing or further disease confirmation; the smaller the value is, the condition of the patient does not change obviously, i.e. the affected limb of the patient does not move or the recovery and cure process of orthopedic traction of the affected limb is not changed in the moving process of the affected limb, and the nursing of medical staff is unnecessary.
In order to accurately describe the abnormal condition of the traction of the affected limb to be detected at the current time, the present embodiment uses the average value of the disease change degrees at three times closest to the current time as the initial abnormal degree of the affected limb at the current time, and is used for indicating whether the recovery process of the affected limb suddenly changes due to the movement of the affected limb of the patient in a short time. When the value is larger, the abnormality of the affected limb occurs in the orthopedic traction process, and medical care personnel need to timely handle the abnormality; a smaller value does not mean that no treatment is required, and this embodiment also takes into account that the limb will not move more seriously at a later time, that the movement of the limb may be accompanied by loosening of the bandage, the loose bandage further creates a conditioning agent for the movement of the affected limb, so that the subsequent illness state of the affected limb has stronger uncertainty, or the affected limb has larger instability in the recovery process of orthopedic traction, namely, the abnormal state of the traction of the affected limb is also large, so the influence of the affected limb movement and bandage looseness of the patient on the disease condition at the future time needs to be pre-judged, then, the initial abnormal degree is corrected, the corrected abnormal degree is used as the final abnormal degree, and then whether timely nursing is needed is judged according to the final abnormal degree, and the condition of illness is prevented from being seriously deteriorated, and then medical staff are informed to carry out nursing.
In this embodiment, the influence relationship between the movement characteristic of the affected limb and the loosening characteristic of the bandage needs to be obtained according to the data of the affected limb to be detected in the previous orthopedic traction process.
For a patient, the disease change characteristics of a diseased limb, the movement characteristics of the diseased limb and the loosening characteristics of the bandage are obtained at each moment. The loosening characteristic of the bandage is 0, namely the bandage is not loosened, and the loosening characteristic of the bandage is not 0, namely the bandage is loosened. The present embodiment obtains the loosening characteristic of the bandage as 0 and all the time when the affected limb moves, and the disease change characteristics at these times are only related to the movement characteristic of the affected limb. In this embodiment, the movement characteristic of the affected limb is not 0, i.e. the movement characteristic of the affected limb is 0. Combining the disease change characteristics of the affected limb and the movement characteristics of the affected limb into a vector, and calling the vector as the movement diseased characteristics; and performing ZCA whitening treatment on all the mobile diseased features at the moments, wherein the aim is to enable all the mobile diseased features to have consistent variance distribution in different dimensions, and the problem that the distribution difference of each dimension is too large to cause the inaccurate subsequent calculation result is prevented. The ZCA whitening process is a well-known technique and will not be described here.
Clustering all the obtained mobile diseased features by using a mean shift algorithm to obtain a plurality of categories, wherein each category is a set of some mobile diseased features, the mobile diseased features in the same category are distributed together in a centralized manner and have stronger similarity, and the mobile diseased features in different categories have larger differences; that is, the moving features in the same category always appear together with the diseased features, and the moving features in the same category have a strong correlation with the diseased features. For example, movement characteristics within the same category can lead to the development of pathological features, colloquially, patients can develop certain diseased changes in their affected limb due to certain limb movements. Assuming that M categories are obtained in total, the categories are collectively called as the category of the mobile disease characteristics, wherein the mth category is marked as
。
The feature extraction module of the embodiment is further configured to obtain loosening diseased features of the to-be-detected affected limb when the bandage is loosened under different movement conditions, and obtain loosening diseased features of each category after clustering; the loosening diseased characteristic comprises a disease change characteristic and a loosening characteristic. The method specifically comprises the following steps: all moments when the loosening characteristic of the bandage is not 0 are acquired, and the change characteristic of the disease at the moments is related to the moving characteristic of the affected limb and the loosening characteristic of the bandage. Combining the disease change characteristics of the affected limb and the loosening characteristics of the bandage into a vector, which is called loosening diseased characteristics; and performing ZCA whitening treatment on all the loosening diseased features at the moments, wherein the aim is to enable all the loosening diseased features to have consistent variance distribution in different dimensions, and the condition that the distribution difference of each dimension is too large to cause the inaccurate subsequent calculation result is prevented.
Clustering all the acquired loosening diseased features by using a mean shift algorithm to acquire a plurality of categories, wherein each category is a set of some loosening diseased features, and the loosening diseased features in the same category are distributed together in a centralized manner and have strong similarity; the loosening sickness characteristics in different categories have large differences; that is, the loosening features and the diseased features in the same category always appear together, and the loosening features in the same category have stronger relevance to the diseased features. For example, loosening characteristics in the same category lead to the development of pathological features, colloquially, the presence of certain loosening characteristics of wound bandages on affected limbs leads to certain diseased changes in the affected limbs. Suppose that N categories are obtained, and these categories are collectively called the loosening disease characteristic category, wherein the nth category is expressed as
。
The embodiment acquires the moving diseased characteristics of each category and the loosening diseased characteristics of each category corresponding to the diseased limb to be detected, and is used for subsequently calculating the influence degree of the moving characteristics and the loosening characteristics.
Bandage difference calculating module
The bandage difference calculation module is used for obtaining disease change characteristics of the to-be-detected affected limb under the conditions that the affected limb moves and the bandage is loosened according to the movement disease characteristics of each category and the loosening disease characteristics of each category corresponding to the to-be-detected affected limb, and recording the disease change characteristics as first disease change characteristics; according to the movement diseased characteristics of each category and the loosening diseased characteristics of each category corresponding to the diseased limb to be detected, obtaining disease change characteristics of the diseased limb to be detected under the condition that the diseased limb moves and the bandage is not loosened, and recording the disease change characteristics as second disease change characteristics; a difference is calculated between the first condition change characteristic and the second condition change characteristic.
、
The moving characteristic and the loose characteristic in the two categories have strong relevance to some diseased change characteristics, and the data of the two categories play an important role in acquiring the influence degree of the moving characteristic and the loose characteristic.
III, abnormal degree calculation module
And the abnormal degree calculation module is used for acquiring various types of predicted movement characteristics corresponding to the to-be-detected affected limb and calculating the abnormal degree of traction at the current moment according to the difference and the various types of predicted movement characteristics.
The movement of the patient's limb is determined by the subjective awareness of the patient,The moving characteristics of the affected limb at the next moment can be estimated according to the moving characteristic data of the affected limb at the historical moment. In this embodiment, the method needs to acquire the moving feature at the next time of the current time, and includes: suppose that the movement characteristics of the affected limb to be detected at the current moment t are
(ii) a Obtaining and comparing the moving features obtained at all times in history
All moving features having a euclidean distance smaller than the threshold th, and the set of these moving features is denoted as
The movement characteristics are
Are all quite similar. Obtaining
Assuming that the movement characteristic of the diseased limb to be detected at the moment q is obtained
Acquiring a moving feature with a mode length different from 0 and closest to the moving feature in time length, and recording the moving feature as the moving feature
，r>q，
Is called as
Adjacent moving features of (a).
Wherein adjacent moving features of all moving features constitute oneAnd carrying out mean shift clustering on the moving features in the new moving feature set to obtain a plurality of classes, wherein the number of the classes is H, each class is a set of some moving features, the moving features in the same class are distributed together in a centralized manner, the similarity is high, and the distribution difference of the moving features in different classes is large. The obtained H categories are collectively called as predicted movement characteristic categories, and the value of the movement characteristic of the affected limb to be detected at the next moment of the current moment is determined by the predicted movement characteristic categories.
The abnormal degree of traction at the current moment t is as follows:
wherein,
the degree of abnormality of the traction at the present time,
the disease change degree of the affected limb to be detected at the current moment,
is the disease parameter at the current moment,
in order to be a hyper-parameter,
the acquisition method comprises the following steps: obtaining Euclidean distance of two moving features with the maximum Euclidean distance in history, wherein the reciprocal of the Euclidean distance is
Aim to ensure
The value of (c) is not too large.
The embodiment has acquired M moving diseased feature classes and N loosening diseased feature classes according to historical data, wherein the M moving diseased feature class
And nth loosening diseased characteristic category
Corresponding to a moving loose influence degree
The method is used for representing the influence of the patient on the condition of illness when the affected limb moves under the condition that the bandage is loosened. In addition, H predicted movement characteristic classes are obtained, and each class represents the possible movement characteristics of the affected limb to be monitored at the next moment.
The following key point of the embodiment is to obtain the disease parameters at the current moment
：
Wherein,
the class number of the moving diseased characteristics corresponding to the diseased limb to be detected,
the number of the types of the loosening diseased characteristics corresponding to the diseased limb to be detected,
in order to predict the category of the moving features,
is the disease parameter at the current moment,
to the extent of concern for a loose feature of the nth loose feature class,
the difference between the disease change characteristics of the affected limb only under the movement characteristic and the disease change characteristics under the conditions of the movement characteristic and the bandage loosening characteristic is detected,
for the h-th predicted moving feature class,
is composed of
The movement characteristics and
the larger the value of the maximum mean difference of the moving features contained in (1), the larger the value is
The movement characteristics and
the greater the difference in the movement characteristics contained in (a). The purpose of this embodiment to calculate this value is to do
The present embodiment focuses more on the degree of influence of motion looseness corresponding to the motion-affected feature category having a small difference from the predicted motion feature
，
For the influence degree of the movement looseness with the attention coefficient, a movement characteristic for characterizing the next moment is
The degree of influence under the loosening characteristic in (1),
to the extent of concern for a loose feature of the nth loose feature class,
are normalized coefficients.
Therefore, the abnormal degree of the traction at the current moment is obtained, and if the abnormal degree of the traction at the current moment is larger than a set threshold value, the abnormal condition of the affected limb in the orthopedic traction process is suddenly caused, medical personnel are required to timely handle the abnormal condition, and the smooth recovery of a patient is ensured. The threshold value of the degree of abnormality is set by a person skilled in the art.
According to the three-dimensional information of the limb to be detected and the three-dimensional information of the bandage at each moment, the moving condition of the limb to be detected and the loosening condition of the bandage are analyzed to obtain the abnormal degree of traction at the current moment. This embodiment is through drawing abnormal state real-time supervision to the affected limb to guarantee that medical personnel in time handle the affected limb, thereby make the patient recovered smoothly. This embodiment need not to rely on medical personnel to detect the abnormal degree of traction again, has solved the current problem that relies on medical personnel to detect the efficiency that exists to the abnormal degree of traction.
In order to verify the accuracy of the detection result of the embodiment, 300 patients are selected in the embodiment, and the detection system of the embodiment is used to detect the 300 patients, as shown in fig. 2, the experimental result indicates that the detection system according to the embodiment has 26 detection errors and 91.33 detection accuracy; therefore, the detection system of the embodiment has higher accuracy.
It should be noted that: the above description is only for the purpose of illustrating the preferred embodiments of the present invention and should not be taken as limiting the scope of the present invention, which is intended to cover any modifications, equivalents, improvements, etc. within the spirit and scope of the present invention.
Claims (5)
1. The utility model provides an orthopedics traction anomaly detection system based on artificial intelligence, its characterized in that, the system includes feature extraction module, bandage difference calculation module and abnormal degree calculation module, wherein:
a feature extraction module: the three-dimensional information acquisition module is used for acquiring the movement characteristics of the to-be-detected affected limb and the loosening characteristics of the corresponding bandage at each moment according to the three-dimensional information of the to-be-detected affected limb and the three-dimensional information of the corresponding bandage at each moment; obtaining disease change characteristics of the diseased limb to be detected at each moment according to the disease characteristics of the diseased limb to be detected at adjacent moments; acquiring mobile diseased characteristics of the diseased limb to be detected when the bandage is not loosened under different movement conditions, and clustering to obtain mobile diseased characteristics of each category; acquiring loosening diseased characteristics of the affected limb to be detected when the bandage is loosened under different moving conditions, and clustering to obtain loosening diseased characteristics of each category; the movement diseased feature comprises a condition changing feature and a movement feature, and the loosening diseased feature comprises a condition changing feature and a loosening feature;
a bandage difference calculation module: the system is used for obtaining disease change characteristics of the to-be-detected affected limb under the conditions that the affected limb moves and the bandage is loosened according to the movement disease characteristics of each category and the loosening disease characteristics of each category, and recording the disease change characteristics as first disease change characteristics; according to the movement diseased characteristics of each category and the loosening diseased characteristics of each category, obtaining disease change characteristics of the diseased limb to be detected under the condition that the diseased limb moves and the bandage does not loosen, and recording the disease change characteristics as second disease change characteristics; calculating a difference between the first condition change characteristic and the second condition change characteristic;
an abnormality degree calculation module: the system is used for acquiring various types of predicted movement characteristics corresponding to the to-be-detected affected limb, and calculating the abnormal degree of traction at the current moment according to the difference and the various types of predicted movement characteristics;
the acquiring of the various types of predicted movement characteristics corresponding to the to-be-detected affected limb comprises the following steps:
judging Euclidean distances between the movement features of each historical moment of the limb to be detected and the movement features of the current moment, and constructing a first set of movement features according to the movement features of a plurality of historical moments of which the Euclidean distances are smaller than a preset threshold value;
acquiring the movement characteristics of the next moment corresponding to each movement characteristic in the first set of movement characteristics, and clustering the movement characteristics of the next moment to obtain each category of predicted movement characteristics corresponding to the to-be-detected affected limb;
the abnormal degree calculating module calculates the abnormal degree of the traction at the current moment by adopting the following formula:
wherein,
the degree of abnormality of the traction at the present time,
in order to be a super-parameter,
the disease change degree of the affected limb to be detected at the current moment,
to move the number of categories of diseased features,
to loosen the number of categories of diseased features,
for predicting number of moving feature classesThe amount of the (B) component (A),
is the disease parameter at the current moment,
to the extent of concern for a loose feature of the nth loose feature class,
the difference between the disease change characteristics of the affected limb only under the movement characteristic and the disease change characteristics under the conditions of the movement characteristic and the bandage loosening characteristic is detected,
is the mth mobile diseased characteristic category corresponding to the diseased limb to be detected,
is the h-th predicted movement characteristic category corresponding to the affected limb to be detected,
is composed of
The movement characteristics and
the maximum mean difference of the moving features contained in (1).
2. The system for detecting the abnormal traction in orthopedics department based on the artificial intelligence as claimed in claim 1, wherein the obtaining of the movement characteristic of the diseased limb to be detected and the loosening characteristic of the corresponding bandage at each moment according to the three-dimensional information of the diseased limb to be detected and the three-dimensional information of the corresponding bandage at each moment comprises:
recording a set of three-dimensional coordinates of pixel points in the image of the affected limb to be detected at the previous moment corresponding to each moment as a first set of the affected limb, and recording a set of three-dimensional coordinates of pixel points in the image of the affected limb to be detected at each moment as a second set of the affected limb; matching the pixel points in the first affected limb set and the second affected limb set to obtain the displacement vector of each pixel point in the image of the affected limb to be detected at each moment; obtaining the movement characteristics of the to-be-detected affected limb at each moment according to the displacement vector;
recording a set of three-dimensional coordinates of pixel points in a bandage image at a previous moment corresponding to each moment as a first bandage set; recording a set of three-dimensional coordinates of each pixel point in the bandage image at each moment as a second bandage set; matching pixel points in the first bandage set and the second bandage set to obtain a displacement vector of each pixel point in each bandage image at each moment;
and processing the displacement vector of each pixel point in the image of the affected limb to be detected at each moment and the displacement vector of each pixel point in the bandage image by adopting a maximum mean difference algorithm to obtain the loosening characteristic of the bandage corresponding to the affected limb to be detected at each moment.
3. The system for detecting abnormal traction in orthopedics department based on artificial intelligence as claimed in claim 1, wherein the obtaining of the disease change characteristics of the affected limb to be detected at each moment according to the disease characteristics of the affected limb to be detected at adjacent moments comprises: and (4) differentiating the disease characteristics of the to-be-detected affected limb at each moment from the disease characteristics of the to-be-detected affected limb at the last moment to obtain the disease change characteristics of the to-be-detected affected limb at each moment.
4. The system of claim 1, wherein the mean shift algorithm is used to cluster the moving diseased features of the affected limb to be detected when the bandage is not loosened under different moving conditions, so as to obtain moving diseased features of each category.
5. The system for detecting abnormal traction in orthopedics department based on artificial intelligence as claimed in claim 1, wherein mean shift algorithm is used for clustering the loosening diseased features of the affected limb to be detected when the bandage is loosened under different moving conditions to obtain loosening diseased features of each category.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210035001.4A CN114052724B (en) | 2022-01-13 | 2022-01-13 | Orthopedics traction abnormity detection system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210035001.4A CN114052724B (en) | 2022-01-13 | 2022-01-13 | Orthopedics traction abnormity detection system based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114052724A CN114052724A (en) | 2022-02-18 |
| CN114052724B true CN114052724B (en) | 2022-09-09 |
Family
ID=80231065
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210035001.4A Active CN114052724B (en) | 2022-01-13 | 2022-01-13 | Orthopedics traction abnormity detection system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114052724B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115424186B (en) * | 2022-11-05 | 2023-01-13 | 深圳市亿康医疗技术有限公司 | Intelligent processing method for monitoring data of orthopedic traction |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102805677A (en) * | 2003-11-25 | 2012-12-05 | 康复米斯公司 | Patient selectable knee joint arthroplasty devices |
| CN103750888A (en) * | 2014-01-21 | 2014-04-30 | 四川聚能核技术工程有限公司 | Automatic limb fracture reduction method and automatic limb fracture reduction system |
| CN203749664U (en) * | 2014-03-19 | 2014-08-06 | 魏彩虹 | Limb traction protection sleeve |
| CN106132348A (en) * | 2014-04-03 | 2016-11-16 | 阿斯卡拉波股份有限公司 | Medical science is with reference to equipment, medical navigation system and method flexibly |
| CN106175778A (en) * | 2016-07-04 | 2016-12-07 | 中国科学院计算技术研究所 | A kind of method setting up gait data collection and gait analysis method |
| WO2017208480A1 (en) * | 2016-06-02 | 2017-12-07 | 株式会社スペース・バイオ・ラボラトリーズ | Walking motion assist device |
| CN108433859A (en) * | 2018-02-01 | 2018-08-24 | 复旦大学附属中山医院 | A kind of restraint device and its constrained procedure |
| CN109273078A (en) * | 2018-10-30 | 2019-01-25 | 贵州医科大学附属医院 | A kind of ankle foot healing pull-in control system and method |
| CN109414221A (en) * | 2016-06-29 | 2019-03-01 | 视觉探索工业股份有限公司 | Measurement and custom-built system for apparatus for correcting |
| CN109452941A (en) * | 2018-11-23 | 2019-03-12 | 中国科学院自动化研究所 | Limb circumference measurement method and system based on picture correction and Boundary Extraction |
| CN109996515A (en) * | 2016-07-28 | 2019-07-09 | 安纳特思科 | Method for designing and producing a customized device having a shape adapted to a user's form |
| CN110381898A (en) * | 2017-02-13 | 2019-10-25 | 奥索冰岛有限公司 | Orthopedic appliance and method and system for manufacturing orthopedic appliance |
| CN110974232A (en) * | 2019-12-26 | 2020-04-10 | 上海交通大学 | Wearable load monitoring and rehabilitation training intelligent auxiliary device |
| CN111839859A (en) * | 2020-08-18 | 2020-10-30 | 苏州逸动健康科技有限公司 | Self-adaptive adjusting strap and adjusting method |
| CN112084967A (en) * | 2020-09-12 | 2020-12-15 | 周美跃 | Limb rehabilitation training detection method and system based on artificial intelligence and control equipment |
| CN112734746A (en) * | 2021-01-21 | 2021-04-30 | 常德海 | Orthopedic injury spectrum matching system and method based on image recognition |
| CN113180647A (en) * | 2021-06-07 | 2021-07-30 | 郑州大学 | AI limb function rehabilitation pre-intervention behavior acquisition system and method |
| CN113677264A (en) * | 2019-03-11 | 2021-11-19 | 佩德拉科技私人有限公司 | System and method for providing real-time perfusion guidance target for peripheral intervention |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| BR112014028389A2 (en) * | 2012-05-16 | 2017-06-27 | Koninklijke Philips Nv | training apparel to track movement in rehabilitation training for upper limb disorders |
| US20180220966A1 (en) * | 2015-08-06 | 2018-08-09 | Upright Technologies Ltd. | Body movement feedback system and method |
| WO2017189926A1 (en) * | 2016-04-27 | 2017-11-02 | Radial Medical, Inc. | Adaptive compression therapy systems and methods |
| WO2020033997A1 (en) * | 2018-08-14 | 2020-02-20 | Travis Eadie | Customized insole manufacturing system |
-
2022
- 2022-01-13 CN CN202210035001.4A patent/CN114052724B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102805677A (en) * | 2003-11-25 | 2012-12-05 | 康复米斯公司 | Patient selectable knee joint arthroplasty devices |
| CN103750888A (en) * | 2014-01-21 | 2014-04-30 | 四川聚能核技术工程有限公司 | Automatic limb fracture reduction method and automatic limb fracture reduction system |
| CN203749664U (en) * | 2014-03-19 | 2014-08-06 | 魏彩虹 | Limb traction protection sleeve |
| CN106132348A (en) * | 2014-04-03 | 2016-11-16 | 阿斯卡拉波股份有限公司 | Medical science is with reference to equipment, medical navigation system and method flexibly |
| WO2017208480A1 (en) * | 2016-06-02 | 2017-12-07 | 株式会社スペース・バイオ・ラボラトリーズ | Walking motion assist device |
| CN109414221A (en) * | 2016-06-29 | 2019-03-01 | 视觉探索工业股份有限公司 | Measurement and custom-built system for apparatus for correcting |
| CN106175778A (en) * | 2016-07-04 | 2016-12-07 | 中国科学院计算技术研究所 | A kind of method setting up gait data collection and gait analysis method |
| CN109996515A (en) * | 2016-07-28 | 2019-07-09 | 安纳特思科 | Method for designing and producing a customized device having a shape adapted to a user's form |
| CN110381898A (en) * | 2017-02-13 | 2019-10-25 | 奥索冰岛有限公司 | Orthopedic appliance and method and system for manufacturing orthopedic appliance |
| CN108433859A (en) * | 2018-02-01 | 2018-08-24 | 复旦大学附属中山医院 | A kind of restraint device and its constrained procedure |
| CN109273078A (en) * | 2018-10-30 | 2019-01-25 | 贵州医科大学附属医院 | A kind of ankle foot healing pull-in control system and method |
| CN109452941A (en) * | 2018-11-23 | 2019-03-12 | 中国科学院自动化研究所 | Limb circumference measurement method and system based on picture correction and Boundary Extraction |
| CN113677264A (en) * | 2019-03-11 | 2021-11-19 | 佩德拉科技私人有限公司 | System and method for providing real-time perfusion guidance target for peripheral intervention |
| CN110974232A (en) * | 2019-12-26 | 2020-04-10 | 上海交通大学 | Wearable load monitoring and rehabilitation training intelligent auxiliary device |
| CN111839859A (en) * | 2020-08-18 | 2020-10-30 | 苏州逸动健康科技有限公司 | Self-adaptive adjusting strap and adjusting method |
| CN112084967A (en) * | 2020-09-12 | 2020-12-15 | 周美跃 | Limb rehabilitation training detection method and system based on artificial intelligence and control equipment |
| CN112734746A (en) * | 2021-01-21 | 2021-04-30 | 常德海 | Orthopedic injury spectrum matching system and method based on image recognition |
| CN113180647A (en) * | 2021-06-07 | 2021-07-30 | 郑州大学 | AI limb function rehabilitation pre-intervention behavior acquisition system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114052724A (en) | 2022-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020207377A1 (en) | Method, device, and system for image recognition model training and image recognition | |
| US11776680B2 (en) | Method and system for real-time and offline de-identification of facial regions from regular and occluded color video streams obtained during diagnostic medical procedures | |
| Al Arif et al. | Shape-aware deep convolutional neural network for vertebrae segmentation | |
| AU2017292642A1 (en) | System and method for automatic detection, localization, and semantic segmentation of anatomical objects | |
| Liu et al. | Magnetic resonance image diagnosis of femoral head necrosis based on ResNet18 network | |
| Kutlu et al. | Detection of coronavirus disease (COVID-19) from X-ray images using deep convolutional neural networks | |
| Bourbakis | Detecting abnormal patterns in WCE images | |
| US20200170710A1 (en) | Surgical decision support using a decision theoretic model | |
| Kyrollos et al. | Noncontact neonatal respiration rate estimation using machine vision | |
| CN114052724B (en) | Orthopedics traction abnormity detection system based on artificial intelligence | |
| Choi et al. | Video recognition of simple mastoidectomy using convolutional neural networks: Detection and segmentation of surgical tools and anatomical regions | |
| Cotton et al. | Transforming gait: video-based spatiotemporal gait analysis | |
| CN115019388A (en) | Full-automatic gait analysis method for shooting gait video by using monocular camera | |
| Luo et al. | Monocular endoscope 6-DoF tracking with constrained evolutionary stochastic filtering | |
| Zhu et al. | A kalman filter based approach for markerless pose tracking and assessment | |
| Mechmeche et al. | Data correction to determine a representative pattern of a set of 3D knee kinematic measurements | |
| Li et al. | MsgeCNN: Multiscale geometric embedded convolutional neural network for ONFH segmentation and grading | |
| CN114140749A (en) | Chest postoperative respiratory recovery monitoring devices | |
| Xu et al. | Video-based tracking and quantified assessment of spontaneous limb movements in neonates | |
| Figueiredo et al. | Dissimilarity measure of consecutive frames in wireless capsule endoscopy videos: a way of searching for abnormalities | |
| Ng et al. | Medical image segmentation using feature-based GVF snake | |
| Wirth et al. | ShaRPy: Shape Reconstruction and Hand Pose Estimation from RGB-D with Uncertainty | |
| Yazici et al. | In-Bed Pose Estimation: A Review | |
| CN118552793B (en) | Postoperative incision healing state identification system based on artificial intelligence | |
| CN118521966B (en) | Bladder irrigation state monitoring system and method based on image recognition |
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 |