CN106485086B - Human body composition prediction technique based on AIC and improvement entropy assessment - Google Patents
Human body composition prediction technique based on AIC and improvement entropy assessment Download PDFInfo
- Publication number
- CN106485086B CN106485086B CN201610910928.2A CN201610910928A CN106485086B CN 106485086 B CN106485086 B CN 106485086B CN 201610910928 A CN201610910928 A CN 201610910928A CN 106485086 B CN106485086 B CN 106485086B
- Authority
- CN
- China
- Prior art keywords
- aic
- model
- weight
- human body
- body composition
- 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
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
Human body composition prediction technique based on AIC and improvement entropy assessment, comprising: S1: selection impedance model collects data and constructs the primitive character collection F of physiologic information sample；S2: being added the primitive character collection F of physiologic information sample, constructs fisrt feature parameter and second feature parameter；S3: using akaike information criterion, selects AIC stable model；S4: calculating the value of AIC, selects the smallest feature combination of AIC value, obtains characteristic parameter matrix, analyze influence of each characteristic parameter to model of fit, correct characteristic parameter matrix；S5: introducing comentropy, obtains unified matrix, calculates entropy and weight；S6: characteristic parameter matrix coefficient is solved, human body composition model of fit is obtained.Human body composition precision of prediction can be improved in the human body composition prediction model of foundation, provides more efficiently detection means for body composition Study and clinical application.
Description
Technical field
The invention belongs to field of bioinformatics more particularly to a kind of human body compositions based on AIC and improvement entropy assessment
Prediction technique.
Background technique
The variation of human body component reflects the variation of physical condition, the Accurate Prediction of human body component to a certain extent
The prevention important in inhibiting of adjusting and disease to human nutrition situation.The parameter for influencing human body component is numerous, main at present
Including two class of physiology electrical impedance parameter and generic physiological characteristic parameter.Between these physiological parameters there is also nonlinearity,
The characteristics of serious association, existing human body composition model is difficult to meet this needs.
With being constantly progressive for medical measuring technique, detectable physiological characteristic extensive development, and show sample it is few,
The features such as dimension is high, this gives the processing of traditional physiological data and analysis to bring huge challenge, the wherein presence of redundancy feature
Adverse effect has been aggravated indirectly, and human body composition is caused to predict Shortcomings.
In view of the above problems, it is necessary to a kind of new human body composition prediction technique is proposed, to solve the above problems.
Summary of the invention
In view of the deficiencies of the prior art, the invention proposes the human body composition prediction sides based on AIC and improvement entropy assessment
Method selects one group of optimal characteristic parameter from human body physiological parameter, can efficiently reduce the redundancy between characteristic parameter, letter
Change the model of fit of human body component prediction；Secondly, the unknowm coefficient of prediction model is solved using entropy assessment is improved, to obtain
The prediction model of human body composition；Human body composition precision of prediction can be improved in the human body composition prediction model being built such that, is
Human body composition studies and clinical application provides more efficiently detection means.
To achieve the above object, the present invention provides the human body composition prediction technique based on AIC and improvement entropy assessment, packets
It includes:
S1: selection impedance model collects data and constructs the primitive character collection F of physiologic information sample；
S2: being added the primitive character collection F of physiologic information sample, constructs fisrt feature parameter and second feature parameter；
S3: using akaike information criterion, selects AIC stable model；
S4: calculating the value of AIC, selects the smallest feature combination of AIC value, obtains characteristic parameter matrix, analyzes each feature ginseng
Characteristic parameter matrix is corrected in the influence of several pairs of model of fit；
S5: introducing comentropy, calculates unified matrix；
S6: characteristic parameter matrix coefficient is solved, human body composition model of fit is obtained.
Further, five sections of impedance values, gender, age, height, weight, race are fisrt feature；Fisrt feature is put down
The groups such as side, reciprocal and product are combined into second feature；Primitive character collection F is collectively constituted by fisrt feature and second feature.
Further, akaike information criterion AIC are as follows: AIC=2k-ln (L), k are number of parameters, and L is likelihood function.
Further, the AIC stable model of selection are as follows: AICH=log σ2+ (m/n) logn, σ2For the variance of model, m is
The highest parameter of model, n are number of parameters.
Further, it calculates the value of AIC and chooses the smallest feature combination of AIC value, characteristic parameter matrix: [R can be obtained1,R3,
R4,R2R3,R3R5,S,A,H,W,R]T, and influence of each characteristic parameter to model of fit is analyzed, correct characteristic parameter matrix, construction
Final characteristic parameter matrix [R1,R3,R4,R2R3,R3R5,A,H,W]T, wherein R1~R5For impedance value, S is gender, A is year
Age, H are height, W is weight, R is race, RiRjFor impedance value product.
Further, the value of ethnic R is essentially equal, constructs final model of fit are as follows:
Male:f=a1R1+a2R3+a3R4+a4R2R3+a5R3R5+a6A+a7H+a8W+ε1
Female:f=b1R1+b2R3+b3R4+b4R2R3+b5R3R5+b6A+b7H+b8W+ε2
In formula, a1~a8, b1~b8For regression coefficient, ε1, ε1For error.
Further, comentropy calculation formula are as follows: H (R1)=- ∫xp(x)logp(x)dx。
Further, human body composition model of fit solution procedure are as follows:
S51: setting assessment event has m object, n parameter, XijFor j-th of index under i-th of object, according to formulaOr formulaCalculate the decision matrix for determining m row n column, Y=
{Xij}m×n；
S52: eliminating the different dimensional units that the different indexs of object have, and forms unified matrix:
S53: entropy is calculatedE in formulajFor entropy corresponding to j-th of evaluation index；If
Y′ij=0, then ejValue is in [0,1]；
S54: weight is calculatedW in formulajIndicate that the weight of j-th of index, n indicate index number.
As further, human body composition model of fit solution procedure further include:
S55: comprehensive weight is calculated:
After the entropy weight for calculating each evaluation index, sorted the standard scores series to be formed according to each indication information entropy size,
To obtain the comprehensive weight about index x；
Rule set total entropy are as follows:
Since the importance of each evaluation index has been lain in grade scale, conventional weight is determined by grade scale value
λj, the weight calculation formula is as follows:
Wherein λjFor the conventional weight of j-th of index, k is characterized the comentropy row for the parameter index that selection algorithm is selected
The standard scores series of sequence.
As further, comprehensive routine weight λjWith objective weight wjObtain new improvement entropy weight weight:
The present invention due to using the technology described above, can obtain following technical effect: select from human body physiological parameter
One group of optimal characteristic parameter is selected out, the redundancy between characteristic parameter can be efficiently reduced, simplifies the fitting of human body component prediction
Model；Secondly, the unknowm coefficient of prediction model is solved using entropy assessment is improved, to obtain the prediction mould of human body composition
Type；Human body composition precision of prediction can be improved in the human body composition prediction model being built such that, for body composition Study and faces
Bed application provides more efficiently detection means.
Detailed description of the invention
Fig. 1 is the flow chart the present invention is based on AIC and the human body composition prediction technique for improving entropy assessment；
Fig. 2 is human body composition model of fit solution procedure flow chart；
Fig. 3 is the age when being impact factor, the distribution situation of human body composition；
Fig. 4 is weight when being impact factor, the distribution situation of human body composition；
Fig. 5 is height when being impact factor, the distribution situation of human body composition；
Fig. 6 is males ingredient prediction result schematic diagram；
Fig. 7 is women body ingredient prediction result schematic diagram.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, right in the following with reference to the drawings and specific embodiments
The present invention is described in detail.
Before the human body composition prediction technique for being described in detail based on AIC and improving entropy assessment, this method institute is first introduced
Need collection and the relevant calculation formula of parameter.
Firstly, construction human body physiological parameter primitive character collection.Five sections of impedance values are measured using five sections of human body impedance measuring instrument
And the generic physiological characteristic parameter of human body, and five sections of impedance values are selected, gender, at the age, height, weight, race is as the first spy
Sign.Then, second feature is generated by algebraic transformation by fisrt feature, i.e., by fisrt feature square, the groups such as reciprocal and product
Cooperation is second feature.Primitive character is collectively constituted by fisrt feature and second feature, i.e. primitive character collection F is by five sections of impedance values
R1~R5, the combination of impedance value1/Ri,RiRj(1≤i≤5,1≤j≤5), gender S, age A, height H, weight W, ethnic Z
Composition.
It is more than the selection quantity of characteristic parameter due to model of fit and combinations thereof and complex, it is as simple as possible in order to obtain
Single, accurate model, the present invention, as judgment criteria, obtain explanation data that can be best using akaike information criterion AIC
It and include the model of minimum free parameter.
The definition value of AIC are as follows:
AIC=2k-ln (L)
In formula, k is the independent parameter number of model, and L is the maximum likelihood function of model.
When being intended to select an optimal model from one group of alternative model, AIC is selected to be for the smallest model
Desirable, when, there is when sizable difference, this difference comes across the L of above formula, and works as L and do not show between two models
When writing difference, k then works, thus the model that the few model of number of parameters has been.
In view of the complexity of body ingredient prediction process, AIC stable model is selected herein:
AICH=log σ2+(m/n)logn
In formula, σ2For the variance of model, m is the highest parameter of model, and n is number of parameters.
Model selection result is as shown in table 1, arranges from small to large by AIC value, and in formula, No. indicates serial number, and n is variable
Number.
1 model selection result of table
Based on upper table as a result, choosing serial number 1 is the smallest feature combination of AIC value, characteristic parameter matrix can be obtained:
[R1,R3,R4,R2R3,R3R5,S,A,H,W,R]T
Due to this test, selected tester is Han nationality, and the value of ethnic R is essentially equal, therefore to model of fit
Influence is zero, it is possible to remove ethnic R from characteristic parameter matrix.Such as Fig. 3, Fig. 4, shown in Fig. 5, gender, age, body
All there are directly related properties with human body prediction model for high, weight, can be directly used for construction model of fit, in addition to this, Cong Tuzhong
When can be seen that respectively using age, weight, height as impact factor, the difference of gender is always it will be apparent that therefore by gender
Independent processing model, to improve the accuracy of model.
Therefore the characteristic parameter matrix after being simplified:
[R1,R3,R4,R2R3,R3R5,A,H,W]T
Human body composition prediction model is constructed using this feature parameter matrix, obtains two according to the difference of gender here
Prediction model:
Male:f=a1R1+a2R3+a3R4+a4R2R3+a5R3R5+a6A+a7H+a8W+ε1
Female:f=b1R1+b2R3+b3R4+b4R2R3+b5R3R5+b6A+b7H+b8W+ε2
In formula, a1~a8For the unknowm coefficient of male's model of fit, ε1For error；b1~b8For the unknown of women model of fit
Coefficient, ε2For error.
If enabling X=[x1,x2,x3,x4,x5,x6,x7,x8]=[R1,R3,R4,R2R3,R3R5, A, H, W],
A=[a1,a2,a3,a4,a5,a6,a7,a8], B=[b1,b2,b3,b4,b5,b6,b7,b8]
Then model of fit f may be expressed as:
Male:f (x)=AX '+ε
Female:f (x)=BX '+ε
Embodiment 1
Present embodiments provide a kind of human body composition prediction technique based on AIC and improvement entropy assessment, comprising:
S1: considering the difference of partes corporis humani position, selects five sections of impedance models, collects data and constructs physiologic information sample
Primitive character collection F；
S2: consider a series of other physiological parameters for influencing human body component, the primitive character collection of physiologic information sample is added
F constructs fisrt feature parameter and second feature parameter；
Current five sections of human body impedance models are the segment impedance models being commonly used the most, by the difference of partes corporis humani point
It is different to take into account, human body is divided into the totally five sections of impedances of right upper extremity, left upper extremity, trunk, right lower extremity, left lower extremity.Human body composition is built
Mould is in addition to considering five sections of impedance value R1~R5Outside, it is also contemplated that a series of other physiological parameters for influencing human body components, including gender
S, the factors such as age A, height H, weight W, race R.These characteristic parameters are divided into fisrt feature parameter, the second sign parameter, wherein
R1~R5, S, A, H, W, R be first sign parameter；1/R reciprocaliAnd product RiRj(1≤i≤5,1≤j≤5) are the second sign ginseng
Number, the candidate feature parameter of assembly ingredient prediction model.
S3: it uses akaike information criterion as judgment criteria, selects AIC stable model；
Akaike information criterion AIC are as follows: AIC=2k-ln (L), k are number of parameters, and L is likelihood function.
Body ingredient prediction process is complicated, selects AIC stable model:
AICH=log σ2+(m/n)logn
In formula, σ2For the variance of model, m is the highest parameter of model, and n is number of parameters.
S4: calculating the value of AIC, selects the smallest feature combination of AIC value, obtains characteristic parameter matrix: [R1,R3,R4,R2R3,
R3R5,S,A,H,W,R]T.Influence of the sign parameter to model of fit is calculated, characteristic parameter matrix: [R is corrected1,R3,R4,R2R3,R3R5,
A,H,W]T,
Male:f=a1R1+a2R3+a3R4+a4R2R3+a5R3R5+a6A+a7H+a8W+ε1
Female:f=b1R1+b2R3+b3R4+b4R2R3+b5R3R5+b6A+b7H+b8W+ε2
S5: introducing comentropy, improves traditional entropy assessment, calculates unified matrix；Comentropy calculation formula are as follows: H (R1)=- ∫xp
(x)logp(x)dx；
S6: characteristic parameter matrix coefficient is solved, human body composition model of fit is obtained.
Embodiment 2
As the supplement of embodiment 1, human body composition model of fit solution procedure are as follows:
S51: assuming that assessment event has m object, n parameter, XijFor j-th of index under i-th of object, then m row n is arranged
Decision matrix Y={ Xij}m×nIt is calculated according to more bigger more excellent type index:
Or smaller more excellent type index calculates:
S52: eliminating the different dimensional units that the different indexs of object have, and forms unified matrix:In order to
Make ln (Y 'ij) significant, it generally can be assumed that as Y 'ijWhen=0, Y 'ijln(Y′ij)=0.But Y 'ijWhen=1, ln (Y 'ij)
Equal to 0, it is clear that do not conform to the actual conditions, and run counter to entropy meaning this, to Y 'ijIt modifies:
S53: entropy is calculatedE in formulajFor entropy corresponding to j-th of evaluation index；If
Y′ij=0, then ejValue is in [0,1]；
S54: weight is calculatedW in formulajIndicate that the weight of j-th of index, n indicate index number；
S55: comprehensive weight is calculated:
Entropy is probabilistic measurement, and it is objective that entropy weight, which embodies the size for referring to target role of evaluation in objective information,
Weight.After the entropy weight for calculating each evaluation index first with the thinking of above-mentioned entropy assessment, according to each indication information entropy size
Sort the standard scores series formed, to obtain the comprehensive weight about index x；
Rule set total entropy are as follows:
Since the importance of each evaluation index has been lain in grade scale, conventional weight is determined by grade scale value
λj, the weight calculation formula is as follows:
Wherein λjFor the conventional weight of j-th of index, k is characterized the comentropy row for the parameter index that selection algorithm is selected
The standard scores series of sequence.
Comprehensive routine weight λjWith objective weight wjObtain new improvement entropy weight weight:
Body ingredient prediction is carried out to 80 males in sample data and 80 women using the human body composition model of construction,
The body fat percentage measured value of prediction result and South Korea InBody770 do relative error comparison diagram, can be seen that from Fig. 6, Fig. 7
The value of the relative error of male and female is both less than 5%, the results showed that, the human body abdomen based on physiologic information entropy and improvement entropy assessment
Portion's fat content predicted value shows good correlation with measured value, and prediction has comparable accuracy.
Compared to the prior art, the present invention is to provide a kind of human body composition prediction side based on AIC and improvement entropy assessment
Method selects one group of optimal characteristic parameter from human body physiological parameter, can efficiently reduce the redundancy between characteristic parameter, letter
Change the model of fit of human body component prediction；Secondly, the unknowm coefficient of prediction model is solved using entropy assessment is improved, to obtain
The prediction model of human body composition；Human body composition precision of prediction can be improved in the human body composition prediction model being built such that, is
Human body composition studies and clinical application provides more efficiently detection means.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
Anyone skilled in the art within the technical scope of the present disclosure, according to the technique and scheme of the present invention and its
Inventive concept is subject to equivalent substitution or change, should be covered by the protection scope of the present invention.
Claims (1)
1. the human body composition prediction technique based on AIC and improvement entropy assessment characterized by comprising
S1: selection impedance model collects data and constructs the primitive character collection F of physiologic information sample；
S2: being added the primitive character collection F of physiologic information sample, constructs fisrt feature parameter and second feature parameter；
S3: using akaike information criterion, selects AIC stable model；
S4: calculating the value of AIC, selects the smallest feature combination of AIC value, obtains characteristic parameter matrix, analyze each characteristic parameter pair
Characteristic parameter matrix is corrected in the influence of model of fit；
S5: introducing comentropy, obtains unified matrix, calculates entropy and weight；
S6: characteristic parameter matrix coefficient is solved, human body composition model of fit is obtained；
Human body composition model of fit solution procedure are as follows:
S51: setting assessment event has m object, n parameter, XijFor j-th of index under i-th of object, according to formulaOr formulaCalculate the decision matrix Y=for determining m row n column
{Xij}m×n；
S52: eliminating the different dimensional units that the different indexs of object have, and forms unified matrix:
S53: entropy is calculatedE in formulajFor entropy corresponding to j-th of evaluation index；If Y 'ij
=0, then ejValue is in [0,1]；
S54: weight is calculatedW in formulajIndicate that the weight of j-th of index, n indicate index number；
S55: it calculates comprehensive weight: after the entropy weight for calculating each evaluation index, being sorted to be formed according to each indication information entropy size
Standard scores series, to obtain the comprehensive weight about index x；
Rule set total entropy are as follows:
Since the importance of each evaluation index has been lain in grade scale, conventional weight λ is determined by grade scale valuej, the power
Re-computation formula is as follows:
Wherein λjFor the conventional weight of j-th of index, k is characterized the comentropy sequence for the parameter index that selection algorithm is selected
Standard scores series；
Comprehensive routine weight λjWith objective weight wjObtain new improvement entropy weight weight:
Five sections of impedance values, gender, age, height, weight, race are fisrt feature；Fisrt feature square, reciprocal and product group
It is combined into second feature；Primitive character collection F is collectively constituted by fisrt feature and second feature；Akaike information criterion AIC are as follows: AIC
=2k-ln (L), k are number of parameters, and L is likelihood function；The AIC stable model of selection are as follows: AICH=log σ2+ (m/n) logn,
σ2For the variance of model, m is the highest parameter of model, and n is number of parameters；It calculates the value of AIC and chooses the smallest feature of AIC value
Combination, obtains characteristic parameter matrix: [R1,R3,R4,R2R3,R3R5,S,A,H,W,R]T, and each characteristic parameter is analyzed to model of fit
It influences, corrects characteristic parameter matrix, construct final characteristic parameter matrix [R1,R3,R4,R2R3,R3R5,A,H,W]T, wherein R1~
R5For impedance value, S is gender, A is the age, H is height, W is weight, R is race, RiRjFor impedance value product；The value of ethnic R
It is essentially equal, construct final model of fit are as follows:
Male:f=a1R1+a2R3+a3R4+a4R2R3+a5R3R5+a6A+a7H+a8W+ε1
Female:f=b1R1+b2R3+b3R4+b4R2R3+b5R3R5+b6A+b7H+b8W+ε2
In formula, a1~a8, b1~b8For regression coefficient, ε1, ε1For error；
Comentropy calculation formula are as follows: H (R1)=- ∫xp(x)logp(x)dx。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610910928.2A CN106485086B (en) | 2016-10-19 | 2016-10-19 | Human body composition prediction technique based on AIC and improvement entropy assessment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610910928.2A CN106485086B (en) | 2016-10-19 | 2016-10-19 | Human body composition prediction technique based on AIC and improvement entropy assessment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106485086A CN106485086A (en) | 2017-03-08 |
| CN106485086B true CN106485086B (en) | 2019-10-25 |
Family
ID=58269787
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610910928.2A Active CN106485086B (en) | 2016-10-19 | 2016-10-19 | Human body composition prediction technique based on AIC and improvement entropy assessment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106485086B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107845407A (en) * | 2017-08-24 | 2018-03-27 | 大连大学 | Based on filtering type and improve the human body physiological characteristics selection algorithm for clustering and being combined |
| CN108281192B (en) * | 2017-12-29 | 2022-03-22 | 一诺仪器（中国）有限公司 | Human body component prediction method and system based on ensemble learning algorithm |
| CN109801711B (en) * | 2018-12-25 | 2021-04-30 | 浙江工业大学 | Juvenile body composition prediction method based on PSO algorithm |
| CN110363229B (en) * | 2019-06-27 | 2021-07-27 | 岭南师范学院 | Human body characteristic parameter selection method based on combination of improved RReliefF and mRMR |
| CN110706757B (en) * | 2019-09-30 | 2022-06-21 | 武钢资源集团金山店矿业有限公司 | Method for predicting concentration of residual flocculant in mineral separation backwater |
| CN111192681A (en) * | 2019-12-25 | 2020-05-22 | 新绎健康科技有限公司 | Method and system for acquiring target blood glucose characteristics |
| CN113611372B (en) * | 2021-09-01 | 2024-07-19 | 天津仁雨生物科技有限公司 | Method and system for predicting drug effect fitting of effective components of traditional Chinese medicine prescription |
| CN114278362B (en) * | 2021-12-21 | 2024-10-29 | 山西潞安环保能源开发股份有限公司 | Outburst prevention method for high-gas mine area based on hierarchical analysis of coal seam gas content |
| CN115033806A (en) * | 2022-06-24 | 2022-09-09 | 东软睿驰汽车技术(大连)有限公司 | User gender determination method and device based on network comments and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103123669A (en) * | 2013-02-28 | 2013-05-29 | 大连大学 | Human body composition analysis method based on genetic algorithm |
| CN103637800A (en) * | 2013-12-20 | 2014-03-19 | 大连大学 | Eight-section impedance model based body composition analysis method |
| CN105138835A (en) * | 2015-08-18 | 2015-12-09 | 大连大学 | Human body composition prediction method based on physiological information entropy |
-
2016
- 2016-10-19 CN CN201610910928.2A patent/CN106485086B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103123669A (en) * | 2013-02-28 | 2013-05-29 | 大连大学 | Human body composition analysis method based on genetic algorithm |
| CN103637800A (en) * | 2013-12-20 | 2014-03-19 | 大连大学 | Eight-section impedance model based body composition analysis method |
| CN105138835A (en) * | 2015-08-18 | 2015-12-09 | 大连大学 | Human body composition prediction method based on physiological information entropy |
Non-Patent Citations (4)
| Title |
|---|
| Comparison of body composition measurement with whole body multifrequency bioelectrical impedance and air displacement plethysmography in healthy middle-aged women;Pamela D Swan et al;《Health Care for Women International》;20111201;第32卷(第12期);1068-1078 * |
| Measurements of body composition in limbs and trunk using a eight contact electrodes impedancemeter;Michel Y.Jaffrin et al;《Medical Engineering & Physics》;20091130;第31卷(第9期);1079-1086 * |
| Prediction model of human body composition based on physiological information entropy;Bo Chen et al;《 Biomedical Engineering and Informatics (BMEI), 2015 8th International Conference on》;20151016;495-499 * |
| Research on human body composition prediction model based on Akaike Information Criterion and improved entropy method;Bo Chen et al;《 Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), International Congress on》;20161017;1882-1886 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106485086A (en) | 2017-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106485086B (en) | Human body composition prediction technique based on AIC and improvement entropy assessment | |
| CN108597601B (en) | Support vector machine-based chronic obstructive pulmonary disease diagnosis auxiliary system and method | |
| WO2021027295A1 (en) | Human body composition prediction method based on improved adaptive genetic algorithm | |
| Brazier et al. | Estimating a preference-based index from the Japanese SF-36 | |
| Silver et al. | Fast identification of biological pathways associated with a quantitative trait using group lasso with overlaps | |
| CN112435756B (en) | Intestinal flora associated disease risk prediction system based on multi-dataset difference interaction | |
| CN103123669B (en) | A kind of body composition analysis's method based on genetic algorithm | |
| Qiu et al. | Is estimated cardiorespiratory fitness an effective predictor for cardiovascular and all-cause mortality? A meta-analysis | |
| de Andrade Silva et al. | An experimental study on the use of nearest neighbor-based imputation algorithms for classification tasks | |
| Li et al. | Subgroup identification via homogeneity pursuit for dense longitudinal/spatial data | |
| Gong et al. | Profile clustering in clinical trials with longitudinal and functional data methods | |
| CN111798925A (en) | Method for identifying cell types and components in tissue samples based on gene expression profiles | |
| CN105138835B (en) | Human body composition Forecasting Methodology based on physiologic information entropy | |
| Fu et al. | Early sepsis prediction in icu trauma patients with using an improved cascade deep forest model | |
| Panagoulias et al. | Extreme value analysis for dietary intake based on weight class | |
| CN118380160A (en) | Oral cancer medical record data analysis method based on big data | |
| Aggarwal et al. | Performance of standard severity scoring systems for outcome prediction in patients admitted to a respiratory intensive care unit in North India | |
| Koestler et al. | A recursively partitioned mixture model for clustering time-course gene expression data | |
| Antolini et al. | Graphical representations and summary indicators to assess the performance of risk predictors | |
| Varmuza et al. | Redundancy analysis for characterizing the correlation between groups of variables-Applied to molecular descriptors | |
| CN110717629A (en) | Method and device for predicting admission score line, electronic equipment and readable storage medium | |
| KR20190141460A (en) | Method and system for providing personalized aerobic capacity ability information based on genes | |
| Kosinski et al. | Sample size calculations for evaluating a diagnostic test when the gold standard is missing at random | |
| Luo et al. | How is standardized mean difference computed, reported and interpreted in randomized controlled trials: protocol for a meta-epidemiological study | |
| Kumar et al. | Efficient classification of sugarcane genomes |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 |