CN112842261B - Intelligent evaluation system for three-dimensional spontaneous movement of infant based on complex network - Google Patents
Intelligent evaluation system for three-dimensional spontaneous movement of infant based on complex network Download PDFInfo
- Publication number
- CN112842261B CN112842261B CN202011616536.8A CN202011616536A CN112842261B CN 112842261 B CN112842261 B CN 112842261B CN 202011616536 A CN202011616536 A CN 202011616536A CN 112842261 B CN112842261 B CN 112842261B
- Authority
- CN
- China
- Prior art keywords
- module
- motion
- dimensional
- vector
- output
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/40—Detecting, measuring or recording for evaluating the nervous system
- A61B5/4076—Diagnosing or monitoring particular conditions of the nervous system
- A61B5/4088—Diagnosing of monitoring cognitive diseases, e.g. Alzheimer, prion diseases or dementia
-
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/40—Detecting, measuring or recording for evaluating the nervous system
- A61B5/4058—Detecting, measuring or recording for evaluating the nervous system for evaluating the central nervous system
- A61B5/4064—Evaluating the brain
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Neurology (AREA)
- Surgery (AREA)
- Biomedical Technology (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Physiology (AREA)
- Physics & Mathematics (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Neurosurgery (AREA)
- Psychology (AREA)
- Hospice & Palliative Care (AREA)
- Child & Adolescent Psychology (AREA)
- Developmental Disabilities (AREA)
- Psychiatry (AREA)
- Dentistry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Processing Or Creating Images (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
A baby three-dimensional spontaneous movement intelligent evaluation system based on a complex network comprises a three-dimensional somatosensory information input module, a movement information feature extraction module, a computer with a complex network analysis module and a movement quality output module; the output of the three-dimensional motion sensing information input module is connected with the input of the motion information characteristic extraction module, the output of the motion characteristic extraction module is connected with the input of a computer with a complex network analysis module, and the output of the computer with the complex network analysis module is connected with the input of the motion quality output module; the system can comprehensively reflect the movement characteristics of the baby, performs targeted complexity characteristic evaluation on the spontaneous movement of the baby, and meets the large-scale and intelligent popularization and application of the baby cerebral palsy screening.
Description
Technical Field
The invention belongs to the technical field of human motion assessment, and particularly relates to an intelligent infant three-dimensional spontaneous motion assessment system based on a complex network.
Background
A professional doctor can analyze movement characteristics through clinical observation of spontaneous movement of the infant to realize early prediction of neurodevelopmental disorders such as cerebral palsy and the like, and can promote nerve remodeling through intervention treatment in a brain development strengthening process, so that negative effects of the cerebral palsy on the infant are reduced, but the training period of the early diagnosis mode is long and the efficiency is low. In order to improve the early screening efficiency of the infantile cerebral palsy, a computer and a visual sensor can be adopted to assist a doctor to finish the observation and analysis process, and powerful technical support is provided for early screening and rehabilitation training of the cerebral palsy.
At present, with the rapid development of somatosensory technology, the digital evaluation of human body movement also becomes a research hotspot. The existing human motion digital assessment mainly comprises two main categories of identification evaluation of specific actions and clinical assessment of typical actions. The standardized evaluation of specific actions can only be applied to conscious exercise learning, such as evaluation of execution degree of standard actions of sports, dancing, rehabilitation training and the like, and the motion evaluation method is not suitable for evaluation of the motion of the infant in consideration of the involuntary motion of the infant; the characteristic evaluation of typical movement uses more traditional time-frequency domain analysis, is mainly used for dyskinesia evaluation caused by medical diseases, can describe the dynamic characteristics of movement, but does not evaluate the nonlinear characteristics of spontaneous movement of infants.
However, the computer system is used for assisting a professional physician to perform early screening of cerebral palsy, and the core problem is to digitally represent the infant movement and extract and evaluate corresponding characteristics. However, the existing movement evaluation multiple bases on the statistical distribution of movement for early prediction of infantile cerebral palsy cannot consider the change process of the movement per se on the time sequence, can not comprehensively and essentially reflect the characteristics of the whole body movement, and then has low prediction accuracy on the neural development state.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide the infant three-dimensional spontaneous movement intelligent evaluation system based on the complex network, which can comprehensively reflect the movement characteristics of the infant, carry out targeted complexity characteristic evaluation on the spontaneous movement of the infant and meet the large-scale and intelligent popularization and application of the infant cerebral palsy screening.
In order to achieve the purpose, the invention adopts the technical scheme that:
a baby three-dimensional spontaneous movement intelligent evaluation system based on a complex network comprises a three-dimensional somatosensory information input module, a movement information feature extraction module, a computer with a complex network analysis module and a movement quality output module; the output of the three-dimensional motion sensing information input module is connected with the input of the motion information characteristic extraction module, the output of the motion characteristic extraction module is connected with the input of a computer with a complex network analysis module, and the output of the computer with the complex network analysis module is connected with the input of the motion quality output module.
The three-dimensional somatosensory information input module realizes acquisition of three-dimensional somatosensory information, wherein the three-dimensional somatosensory information is human skeleton three-dimensional coordinates output by Kinect equipment, namely three-dimensional key point information of a human body.
The motion information characteristic extraction module comprises a human body structure vector representation module and a human body motion characteristic representation module, the human body structure vector representation module represents three-dimensional key point information output by the three-dimensional body feeling information input module as human body structure vectors, and the human body motion characteristic representation module realizes key point angle calculation between the human body structure vectors.
The computer with the complex network analysis module comprises an association dimension analysis module and a small-world network reconstruction module; the correlation dimension analysis module analyzes the key point angle features of the input human body structure and calculates the correlation dimension of the input feature sequence; and the small-world network reconstruction module analyzes the small-world network characteristics of the key point angle characteristics of the input human body structure.
The motion quality output module comprises a comprehensive motion quality generation module and an average comprehensive evaluation index output module; the comprehensive motion quality generation module combines the output of the correlation dimension analysis module and the output of the small-world network reconstruction module to generate a final comprehensive evaluation index; the average comprehensive evaluation index output module repeatedly calculates the comprehensive evaluation index of the computer with the complex network analysis module and outputs the final average comprehensive evaluation index.
The invention has the following beneficial effects:
(1) the system of the invention fully considers the variability of the motion in time, and can evaluate the spontaneous motion quality of the infant in terms of complexity in a targeted manner.
(2) The system has high efficiency, avoids the influence of subjective factors in the evaluation of the spontaneous movement of the infant, and realizes the objective quantitative evaluation of the spontaneous movement quality of the infant.
Drawings
FIG. 1 is a block diagram of the present invention.
FIG. 2 shows the position of the joint point of the human body according to the present invention.
FIG. 3 is a diagram showing the comparison result of solving the correlation dimension by simulating spontaneous movement of a normal infant (complex movement) and a cerebral palsy infant (simple movement), and (a) is a diagram showing the result of the right shoulders of the two infants; panel (b) is a schematic representation of the right wrist results for both infants.
FIG. 4 is a schematic diagram of a correlation coefficient matrix for simulating solution of a group of simple movements and a group of complex movements of normal and cerebral paralyzed infants, wherein (a) is a normal movement correlation matrix; graph (b) an abnormal motion correlation matrix.
FIG. 5 is a schematic diagram of a network structure obtained by reconstructing a group of simple movements and a group of complex movements by simulating normal and cerebral palsy infants, wherein (a) is a normal movement network structure; graph (b) abnormal motion network structure.
FIG. 6 is a diagram of the combined indicators of normal and abnormal infants in the group of normal and cerebral palsy infants 12.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and examples.
Referring to fig. 1, an intelligent evaluation system for three-dimensional spontaneous movement of an infant based on a complex network comprises a three-dimensional somatosensory information input module, a movement information feature extraction module, a computer with a complex network analysis module and a movement quality output module; the output of the three-dimensional motion sensing information input module is connected with the input of the motion information characteristic extraction module, the output of the motion characteristic extraction module is connected with the input of a computer with a complex network analysis module, and the output of the computer with the complex network analysis module is connected with the input of the motion quality output module.
The three-dimensional somatosensory information input module realizes acquisition of three-dimensional somatosensory information, wherein the three-dimensional somatosensory information is human skeleton three-dimensional coordinates output by Kinect equipment, namely three-dimensional key point information of a human body.
The motion information feature extraction module comprises a human body structure vector representation module and a human body motion feature representation module, wherein the human body structure vector representation module represents the three-dimensional key point information input by the three-dimensional body feeling information input module into a human body structure vector; the human motion characteristic representation module realizes the calculation of the key point angle between the human structure vectors.
Different human body structure vectors are constructed in the human body structure vector representation module according to the three-dimensional coordinates of the human body joint points output by the three-dimensional somatosensory information input module;
taking 17 human body structure vectors as an example, the following structure vectors can be constructed: neck → head vectorNeck → right shoulder vectorRight shoulder → right elbow vectorRight elbow → right wrist vectorNeck → left shoulder vectorLeft shoulder → left elbow vectorLeft elbow → left wrist vectorNeck → right hip vectorRight hip → right knee vectorRight Knee → Right ankle vectorNeck → left hip vectorLeft hip → left knee vectorLeft knee → left ankle vectorNeck → right wrist vectorNeck → left wrist vectorRight hip → right ankle vectorLeft hip → left ankle vectorTaking the right upper body structure vector as an example, the calculation method of the right upper body structure vector is as follows:
in the formula: (x)Part,yPart,zPart) Three-dimensional coordinates of corresponding nodes of the body are respectively shown, wherein Part represents a corresponding key point of the human body, H is the head, N is the neck, SR is the right shoulder, ER is the right elbow, and WR is the right wrist, as shown in FIG. 2.
The human motion characteristic representation module takes the similarity of human structures into consideration, extracts human joint angles from the human structure vectors output by the human structure vector representation module, and digitally represents the human motion to obtain angle motion characteristic sequences with different dimensions;
taking the 14-dimensional as an example,
θi=[θNR,θNL,θSL,θEL,θWL,θSR,θER,θWR,θHL,θKL,θAL,θHR,θKR,θAR],
wherein theta isiRepresenting the ith key point angle, NR being the right neck, NL being the left neck, SL being the left shoulder, EL being the left elbow, WL being the left wrist, SR being the right shoulder, ER being the right elbow, WR being the right wrist, HL being the left hip, KL being the left knee, AL being the left ankle, HR being the right hip, KR being the right knee, AR being the right ankle; theta is the angle characteristic of the corresponding joint, and the angle theta of the right shoulder jointSRAnd right elbow joint angle thetaERThe calculation method is as follows:
other joint angles are obtained in the same way.
The computer with the complex network analysis module comprises an association dimension analysis module and a small-world network reconstruction module; the correlation dimension analysis module analyzes the key point angle features of the input human body structure and calculates the correlation dimension of the input feature sequence; and the small-world network reconstruction module analyzes the small-world network characteristics of the key point angle characteristics of the input human body structure.
The correlation dimension analysis module adopts a nonlinear system correlation theory to calculate the nonlinear complexity of each dimension motion characteristic sequence output by each human motion characteristic representation module, and selects the time delay tau at the moment as the time delay of the sequence when the first local minimum value of the mutual information quantity appears along with the increase of the time delay tau; meanwhile, as the embedding dimension m increases, when the pseudo-neighbor rate in the phase space decreases to 0, m at this time is selected as the embedding dimension of the sequence; after determining the embedding dimension m and the time delay τ, the correlation dimension D is solved2:
Wherein D is2For a sequence of features { theta }iThe GP correlation dimension of, C (r) represents the ratio of the number of point pairs with a distance less than r in the feature phase space to all possible number of point pairsTo 14-dimensional correlation dimension information D2. FIG. 3 shows a comparison of the correlation dimension solving effects of the angular feature sequences of the right shoulder joint and the right wrist joint of 30 different groups of infants in different motion states; as can be seen from the figure, different motion patterns have different embedding dimensions, complex motions have a higher embedding dimension, while simple motions have a lower embedding dimension.
The small world network reconstruction module analyzes the small world network characteristics of the key point angle characteristics of the input human body structure; taking the synchronicity of the motion of each limb of different human bodies into consideration, and solving the Spearman correlation coefficient among the key point angle characteristic dimensional motion characteristic sequences of the human body structure:
wherein for the original data { theta }iAnd { theta }jRecording theta 'according to the sequence from big to small'iAnd θ'jIs the raw data thetaiAnd thetajLocation in the sorted list is named θ'iAnd thetaj' is the raw data thetaiAnd thetajRank of (2), rank difference ht=θ'i-θ'j(ii) a N is data { thetaiLength of time series; rhos(θi,θj) Spearman's correlation coefficient representing the i and j dimensions in the angular features; forming a correlation coefficient matrix for describing the whole body movement on the basis of the physical connectivity of each node in the human anatomy distribution, and simultaneously considering the joint transmission property of each limb, connecting edges of the connected joints in the anatomy and the non-adjacent joints on the single limb to construct an initial network; and then, taking the absolute values of the elements in the Spearman correlation matrix as the probability of adding new connecting edges between nodes, and reconstructing the initial network aiming at the specific motion sequence to obtain the characteristics of the small-world network under different human motion modes. Fig. 4 (a) and (b) show the correlation coefficient matrix solved for a complex set of motions and a simple set of motions, respectively, and it can be seen that the correlation between the key points of the complex motion in (a) is weaker than that of the simple motion in (b); in FIG. 5(a) And (b) respectively showing a network structure obtained by reconstructing a group of complex motions and a group of simple motions, wherein the connection of the small-world network of the complex motions in (a) is relatively simple, and the relevance between the nodes of the small-world network is strong due to the strong relevance between key points in the simple motions in (b).
The motion quality output module comprises a comprehensive motion quality generation module and an average comprehensive evaluation index output module; the comprehensive motion quality generation module combines the output of the correlation dimension analysis module and the output of the small-world network reconstruction module to generate a final comprehensive evaluation index; the average comprehensive evaluation index output module repeatedly calculates the comprehensive evaluation index of the computer with the complex network analysis module and outputs the final average comprehensive evaluation index.
The comprehensive motion quality generation module takes the correlation dimension output by the correlation dimension analysis module and the small-world network characteristics output by the small-world network reconstruction module as the input of the network model, and solves the whole body motion complexity S under the network statei:
Wherein D2iCorrelation dimension calculated for i times, CiAnd C0Respectively, the cluster coefficient of the i-th reconstructed network and the cluster coefficient of the initial network, LiAnd L0The average path length of the i-th reconstructed and the average path length of the initial network, respectively.
The average comprehensive evaluation index output module repeats the comprehensive motion quality generation module for k times, and averages the comprehensive indexes to obtain a complexity comprehensive index describing the motion of the sample:
fig. 6 shows the comprehensive indexes of 12 groups of normal and abnormal infants, and it can be seen from the figure that the positive sample, i.e. the abnormal sample, can be effectively screened out through the complexity comprehensive index S of 12 groups of samples.
Claims (6)
1. A baby three-dimensional spontaneous motion intelligent evaluation system based on a complex network is characterized in that: the motion quality analysis system comprises a three-dimensional somatosensory information input module, a motion information feature extraction module, a computer with a complex network analysis module and a motion quality output module; the output of the three-dimensional motion sensing information input module is connected with the input of the motion information characteristic extraction module, the output of the motion characteristic extraction module is connected with the input of a computer with a complex network analysis module, and the output of the computer with the complex network analysis module is connected with the input of the motion quality output module;
the three-dimensional somatosensory information input module realizes acquisition of three-dimensional somatosensory information, wherein the three-dimensional somatosensory information is three-dimensional coordinates of human skeleton output by Kinect equipment, namely three-dimensional key point information of a human body;
the motion information characteristic extraction module comprises a human body structure vector representation module and a human body motion characteristic representation module, the human body structure vector representation module represents three-dimensional key point information output by the three-dimensional body feeling information input module as human body structure vectors, and the human body motion characteristic representation module realizes the calculation of key point angles among the human body structure vectors;
the computer with the complex network analysis module comprises an association dimension analysis module and a small-world network reconstruction module; the correlation dimension analysis module analyzes the key point angle features of the input human body structure and calculates the correlation dimension of the input feature sequence; the small world network reconstruction module analyzes the small world network characteristics of the key point angle characteristics of the input human body structure;
the small world network reconstruction module analyzes the small world network characteristics of the key point angle characteristics of the input human body structure; taking the synchronicity of the motion of each limb of different human bodies into consideration, and solving the Spearman correlation coefficient among the key point angle characteristic dimensional motion characteristic sequences of the human body structure:
wherein for the original data { theta }iAnd { theta }jRecording theta 'according to the sequence from big to small'iAnd θ'jIs the raw data thetaiAnd thetajLocation in the sorted list is named θ'iAnd θ'jIs the raw data thetaiAnd thetajRank of (2), rank difference ht=θ′i-θ′j(ii) a N is data { thetaiLength of time series; rhos(θi,θj) Spearman's correlation coefficient representing the i and j dimensions in the angular features; forming a correlation coefficient matrix for describing the whole body movement on the basis of the physical connectivity of each node in the human anatomy distribution, and simultaneously considering the joint transmission property of each limb, connecting edges of the connected joints in the anatomy and the non-adjacent joints on the single limb to construct an initial network; and then, taking the absolute values of the elements in the Spearman correlation matrix as the probability of adding new connecting edges between nodes, and reconstructing the initial network aiming at the specific motion sequence to obtain the characteristics of the small-world network under different human motion modes.
2. The intelligent evaluation system for three-dimensional spontaneous infant movement based on complex network as claimed in claim 1, wherein: different human body structure vectors are constructed in the human body structure vector representation module according to the three-dimensional coordinates of the human body joint points output by the three-dimensional somatosensory information input module; adopting 17 human body structure vectors to construct the following structure vectors: neck → head vectorNeck → right shoulder vectorRight shoulder → right elbow vectorElbow →Right wrist vectorNeck → left shoulder vectorLeft shoulder → left elbow vectorLeft elbow → left wrist vectorNeck → right hip vectorRight hip → right knee vectorRight Knee → Right ankle vectorNeck → left hip vectorLeft hip → left knee vectorLeft knee → left ankle vectorNeck → right wrist vectorNeck → left wrist vectorRight hip → right ankle vectorLeft hip → left ankle vector
The calculation method of the right upper body structure vector is as follows:
in the formula: (x)Part,yPart,zPart) And three-dimensional coordinates of corresponding nodes of the body respectively, wherein Part represents a corresponding key point of the human body, H is the head, N is the neck, SR is the right shoulder, ER is the right elbow, and WR is the right wrist.
3. The intelligent evaluation system for baby three-dimensional spontaneous motion based on complex network as claimed in claim 2, wherein: the human motion characteristic representation module takes the similarity of human structures into consideration, extracts human joint angles from the human structure vectors output by the human structure vector representation module, and digitally represents the human motion to obtain angle motion characteristic sequences with different dimensions; by adopting the method of 14-dimensional,
θi=[θNR,θNL,θSL,θEL,θWL,θSR,θER,θWR,θHL,θKL,θAL,θHR,θKR,θAR],
wherein theta isiRepresenting the ith key point angle, NR being the right neck, NL being the left neck, SL being the left shoulder, EL being the left elbow, WL being the left wrist, SR being the right shoulder, ER being the right elbow, WR being the right wrist, HL being the left hip, KL being the left knee, AL being the left ankle, HR being the right hip, KR being the right knee, AR being the right ankle; theta is the angle characteristic of the corresponding joint, and the angle theta of the right shoulder jointSRAnd right elbow joint angle thetaERThe calculation method is as follows:
other joint angles are obtained in the same way.
4. The intelligent evaluation system for three-dimensional spontaneous infant movement based on complex network as claimed in claim 1, wherein: the correlation dimension analysis module adopts a nonlinear system correlation theory to calculate the nonlinear complexity of each dimension motion characteristic sequence output by each human motion characteristic representation module, and selects the time delay tau at the moment as the time delay of the sequence when the first local minimum value of the mutual information quantity appears along with the increase of the time delay tau; meanwhile, as the embedding dimension m increases, when the pseudo-neighbor rate in the phase space decreases to 0, m at this time is selected as the embedding dimension of the sequence; after determining the embedding dimension m and the time delay τ, the correlation dimension D is solved2:
Wherein D is2For a sequence of features { theta }iThe GP correlation dimension of C (r) represents the proportion of the number of the point pairs with the distance less than r in the characteristic phase space to all possible point pairs, and 14-dimensional correlation dimension information D is obtained2。
5. The intelligent evaluation system for three-dimensional spontaneous infant movement based on complex network as claimed in claim 1, wherein: the motion quality output module comprises a comprehensive motion quality generation module and an average comprehensive evaluation index output module; the comprehensive motion quality generation module combines the output of the correlation dimension analysis module and the output of the small-world network reconstruction module to generate a final comprehensive evaluation index; the average comprehensive evaluation index output module repeatedly calculates the comprehensive evaluation index of the computer with the complex network analysis module and outputs the final average comprehensive evaluation index.
6. The intelligent evaluation system for baby three-dimensional spontaneous motion based on complex network as claimed in claim 5, wherein: the comprehensive motion quality generation module takes the correlation dimension output by the correlation dimension analysis module and the small-world network characteristics output by the small-world network reconstruction module as the input of the network model, and solves the whole body motion complexity S under the network statei:
Wherein D2iCorrelation dimension calculated for i times, CiAnd C0Respectively, the cluster coefficient of the i-th reconstructed network and the cluster coefficient of the initial network, LiAnd L0The average path length of the ith reconstructed network and the average path length of the initial network are respectively;
the average comprehensive evaluation index output module repeats the comprehensive motion quality generation module for k times, and averages the comprehensive indexes to obtain a complexity comprehensive index describing the motion of the sample:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011616536.8A CN112842261B (en) | 2020-12-30 | 2020-12-30 | Intelligent evaluation system for three-dimensional spontaneous movement of infant based on complex network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011616536.8A CN112842261B (en) | 2020-12-30 | 2020-12-30 | Intelligent evaluation system for three-dimensional spontaneous movement of infant based on complex network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112842261A CN112842261A (en) | 2021-05-28 |
CN112842261B true CN112842261B (en) | 2021-12-28 |
Family
ID=75998762
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011616536.8A Active CN112842261B (en) | 2020-12-30 | 2020-12-30 | Intelligent evaluation system for three-dimensional spontaneous movement of infant based on complex network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112842261B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114081447B (en) * | 2021-11-22 | 2024-04-02 | 西安交通大学 | Infant brain development state evaluation system based on common video input |
CN114999648B (en) * | 2022-05-27 | 2023-03-24 | 浙江大学医学院附属儿童医院 | Early screening system, equipment and storage medium for cerebral palsy based on baby dynamic posture estimation |
CN114999646B (en) * | 2022-06-01 | 2023-02-28 | 北京大学 | Newborn exercise development assessment system, method, device and storage medium |
Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014144404A1 (en) * | 2013-03-15 | 2014-09-18 | 12Dx, Inc. | Electronic delivery of information in personalized medicine |
CN104360730A (en) * | 2014-08-19 | 2015-02-18 | 西安交通大学 | Man-machine interaction method supported by multi-modal non-implanted brain-computer interface technology |
CN104715150A (en) * | 2015-03-19 | 2015-06-17 | 上海海事大学 | Migraineur cerebral cortex assistant classification analyzing method based on complex network |
CN104765959A (en) * | 2015-03-30 | 2015-07-08 | 燕山大学 | Computer vision based evaluation method for general movement of baby |
CN107174255A (en) * | 2017-06-15 | 2017-09-19 | 西安交通大学 | Three-dimensional gait information gathering and analysis method based on Kinect somatosensory technology |
CN107349570A (en) * | 2017-06-02 | 2017-11-17 | 南京邮电大学 | Rehabilitation training of upper limbs and appraisal procedure based on Kinect |
CN110046675A (en) * | 2019-04-25 | 2019-07-23 | 河北工业大学 | A kind of the exercise ability of lower limbs appraisal procedure based on improved convolutional neural networks |
CN110532697A (en) * | 2019-08-30 | 2019-12-03 | 南昌航空大学 | One kind dividing shape complex network global efficiency estimation method |
CN110547760A (en) * | 2019-08-09 | 2019-12-10 | 西安交通大学 | infant amblyopia electroencephalogram objective detector |
CN110738093A (en) * | 2019-08-16 | 2020-01-31 | 杭州电子科技大学 | Classification method based on improved small world echo state network electromyography |
CN110801226A (en) * | 2019-11-01 | 2020-02-18 | 西安交通大学 | Human knee joint moment testing system method based on surface electromyographic signals and application |
CN111144217A (en) * | 2019-11-28 | 2020-05-12 | 重庆邮电大学 | Motion evaluation method based on human body three-dimensional joint point detection |
CN111709365A (en) * | 2020-06-17 | 2020-09-25 | 成都工业学院 | Automatic human motion posture detection method based on convolutional neural network |
CN111714339A (en) * | 2020-07-15 | 2020-09-29 | 西安交通大学 | Brain-myoelectricity fusion small-world neural network prediction method for human lower limb movement |
CN111931578A (en) * | 2020-07-07 | 2020-11-13 | 杭州电子科技大学 | Electroencephalogram identification method based on minimum spanning tree and regional double-layer network |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1525541A2 (en) * | 2002-07-26 | 2005-04-27 | Ron Everett | Data management architecture associating generic data items using references |
US9095303B2 (en) * | 2009-03-23 | 2015-08-04 | Flint Hills Scientific, Llc | System and apparatus for early detection, prevention, containment or abatement of spread abnormal brain activity |
-
2020
- 2020-12-30 CN CN202011616536.8A patent/CN112842261B/en active Active
Patent Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014144404A1 (en) * | 2013-03-15 | 2014-09-18 | 12Dx, Inc. | Electronic delivery of information in personalized medicine |
CN104360730A (en) * | 2014-08-19 | 2015-02-18 | 西安交通大学 | Man-machine interaction method supported by multi-modal non-implanted brain-computer interface technology |
CN104715150A (en) * | 2015-03-19 | 2015-06-17 | 上海海事大学 | Migraineur cerebral cortex assistant classification analyzing method based on complex network |
CN104765959A (en) * | 2015-03-30 | 2015-07-08 | 燕山大学 | Computer vision based evaluation method for general movement of baby |
CN107349570A (en) * | 2017-06-02 | 2017-11-17 | 南京邮电大学 | Rehabilitation training of upper limbs and appraisal procedure based on Kinect |
CN107174255A (en) * | 2017-06-15 | 2017-09-19 | 西安交通大学 | Three-dimensional gait information gathering and analysis method based on Kinect somatosensory technology |
CN110046675A (en) * | 2019-04-25 | 2019-07-23 | 河北工业大学 | A kind of the exercise ability of lower limbs appraisal procedure based on improved convolutional neural networks |
CN110547760A (en) * | 2019-08-09 | 2019-12-10 | 西安交通大学 | infant amblyopia electroencephalogram objective detector |
CN110738093A (en) * | 2019-08-16 | 2020-01-31 | 杭州电子科技大学 | Classification method based on improved small world echo state network electromyography |
CN110532697A (en) * | 2019-08-30 | 2019-12-03 | 南昌航空大学 | One kind dividing shape complex network global efficiency estimation method |
CN110801226A (en) * | 2019-11-01 | 2020-02-18 | 西安交通大学 | Human knee joint moment testing system method based on surface electromyographic signals and application |
CN111144217A (en) * | 2019-11-28 | 2020-05-12 | 重庆邮电大学 | Motion evaluation method based on human body three-dimensional joint point detection |
CN111709365A (en) * | 2020-06-17 | 2020-09-25 | 成都工业学院 | Automatic human motion posture detection method based on convolutional neural network |
CN111931578A (en) * | 2020-07-07 | 2020-11-13 | 杭州电子科技大学 | Electroencephalogram identification method based on minimum spanning tree and regional double-layer network |
CN111714339A (en) * | 2020-07-15 | 2020-09-29 | 西安交通大学 | Brain-myoelectricity fusion small-world neural network prediction method for human lower limb movement |
Non-Patent Citations (12)
Title |
---|
Collective dynamics of "small-world" networks;Watts, DJ (Watts, DJ) Strogatz, SH (Strogatz, SH);《NATURE》;19980604;第393卷(第6684期);第440-442页 * |
Fault Source Identification of Complex Hydraulic System with Small World Property;Hua ChengZhang QingXu GuanghuaYan DuifengXie Jun;《Journal of Xi"an Jiaotong University》;20100731;第44卷(第7期);第79-84页 * |
Fractals in small-world networks with time-delay;Yang, XS (Yang, XS);《CHAOS SOLITONS & FRACTALS》;20020228;第13卷(第2期);第215-219页 * |
Human 3D pose estimation in a lying position by RGB-D images for medical diagnosis and rehabilitation;Wu, Qingqiang;Xu, Guanghua;Zhang, Sicong;Li, Yu;Wei, Fan;《IEEE Xplore》;20200827;全文 * |
MiniNet: An extremely lightweight convolutional neural network for real-time unsupervised monocular depth estimation;Liu, J (Liu, Jun) Li, Q (Li, Qing) Cao, R (Cao, Rui);《ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING》;20200831;第166卷;第255-267页 * |
The structure and function of complex networks;Newman, MEJ (Newman, MEJ);《SIAM REVIEW》;20030630;第45卷(第2期);第167-256页 * |
基于图论方法分析原发性闭角型青光眼患者大脑结构网络的拓扑属性;李丹 周福庆 刘海萍 肖晓怡 李声鸿 蔡凤琴 曾献军;《中华放射学杂志》;20181231;第52卷(第7期);第489-494页 * |
基于非线性与复杂网络方法的脑和血流动力学研究;方小玲.;《中国博士学位论文全文数据库》;20080615;全文 * |
复杂性–稳定性研究:数学模型的进展;徐光华,李小玉,施春华.;《生物多样性》;20191231;第27卷(第12期);第1364-1378页 * |
幼儿全脑皮层厚度与小世界神经网络的相关性研究;曲海波 鲍丽 李学胜 张毓瑾 杨冬明 宁刚;《中华放射学学术大会2016、中华医学会第23次全国放射学学术大会暨中华医学会第24次全国影像技术学术大会》;20161012;第1-1页 * |
青少年首发抑郁症患者大脑静息状态功能网络小世界特性研究;陈策,金晨望,马现仓,高成阁,张明.;《西安交通大学学报(医学版)》;20141231;第35卷(第5期);第639-645页 * |
非线性与复杂网络理论在脑电数据分析中的应用研究;曹锐.;《中国博士学位论文全文数据库》;20150815;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN112842261A (en) | 2021-05-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112842261B (en) | Intelligent evaluation system for three-dimensional spontaneous movement of infant based on complex network | |
Yang et al. | A complex network-based broad learning system for detecting driver fatigue from EEG signals | |
O'Malley et al. | Fuzzy clustering of children with cerebral palsy based on temporal-distance gait parameters | |
Xiong et al. | Intelligent prediction of human lower extremity joint moment: an artificial neural network approach | |
Chen et al. | Adversarial learning based node-edge graph attention networks for autism spectrum disorder identification | |
Wu et al. | Rgb-d videos-based early prediction of infant cerebral palsy via general movements complexity | |
CN110934587A (en) | Alzheimer disease auxiliary diagnosis method based on atlas neural network | |
CN110321827A (en) | A kind of pain level appraisal procedure based on face pain expression video | |
Zhang et al. | Cross-modal prostate cancer segmentation via self-attention distillation | |
CN116072265A (en) | Sleep stage analysis system and method based on convolution of time self-attention and dynamic diagram | |
Yu et al. | Multi-graph attention networks with bilinear convolution for diagnosis of schizophrenia | |
CN112309578B (en) | Method and system for improving recovery efficiency of osteoporotic vertebral fracture patient | |
Zuo et al. | Alzheimer’s Disease Prediction via Brain Structural-Functional Deep Fusing Network | |
CN113255734A (en) | Depression classification method based on self-supervision learning and transfer learning | |
CN114081447B (en) | Infant brain development state evaluation system based on common video input | |
CN110473636A (en) | Intelligent doctor's advice recommended method and system based on deep learning | |
Wang et al. | Diagnosis of early mild cognitive impairment based on associated high-order functional connection network generated by multi-modal MRI | |
CN108875580A (en) | A kind of multiclass Mental imagery EEG signal identification method based on because imitating network | |
CN108875563A (en) | A kind of human motion recognition method based on muscle signal | |
CN114983341A (en) | Multi-modal feature fusion based multi-classification prediction system for Alzheimer's disease | |
Guo et al. | Intelligent Recognition of Sports Players’ Wrong Actions in view of Computer Pattern Recognition | |
Liu et al. | MCAN: multimodal causal adversarial networks for dynamic effective connectivity learning from fMRI and EEG data | |
Zhou et al. | Assessment of Parkinson's Motor Severity by Multi-feature and Multi-scale Motion Convolutional Neural Network with Positional Encoding | |
Du et al. | View-invariant Center-of-pressure Metrics Estimation with Monocular RGB Camera | |
Xiang et al. | Quantitative movement analysis using scaled information implied in monocular videos |
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 |