CN107451385A - A kind of the nervous system disease monitoring and early warning system based on daily necessities - Google Patents
A kind of the nervous system disease monitoring and early warning system based on daily necessities Download PDFInfo
- Publication number
- CN107451385A CN107451385A CN201610371171.4A CN201610371171A CN107451385A CN 107451385 A CN107451385 A CN 107451385A CN 201610371171 A CN201610371171 A CN 201610371171A CN 107451385 A CN107451385 A CN 107451385A
- Authority
- CN
- China
- Prior art keywords
- nervous system
- daily necessities
- data
- system disease
- action
- 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.)
- Granted
Links
Abstract
The invention discloses a kind of the nervous system disease monitoring based on daily necessities and early warning system.By the way that pervasive sensor is combined with daily necessities, monitoring user uses the situation of daily necessities in daily life.By the collection to data, segmentation and analysis, the upper extremity exercise function with the nervous system disease personnel is would be possible to assess, early warning or auxiliary diagnosis further are carried out to the incidence of the nervous system disease.By the data of quantification, illustrate the motor function situation of the elderly, avoid the uncertainty artificially checked, and by the way that system is implanted in daily life, solution must not in real time monitor and the problem of early warning.
Description
Technical field
The invention belongs to man-machine interaction and domestic medicine field, and in particular to a kind of nervous system disease based on daily necessities
Disease monitoring and early warning system.
Background technology
China has progressed into aging population society, and the problem of being triggered by it is also more and more prominent, endangers elderly population
Chronic disease turn into the important diseases for threatening human health, the nervous system disease one of is exactly.The nervous system disease, such as
Parkinson, small vessel disease, dementia etc., the incidence of disease in worldwide are presented ascendant trend, can cause serious motion work(
Energy obstacle, has the characteristics that high mortality and high disability rate, has a strong impact on patient's life-span and quality of life, to society and family's band
Carry out heavy burden.
Existing diagnosis of nervous system diseases method depend on to nervous function and image change it is empirical
And qualitative judgement.Typically after corresponding health has occurred in the elderly, go to see a doctor, and the coherence check carried out.
Its result has very big uncertainty, and can not accomplish detection and early warning in real time, also runs counter to the early hair of medical treatment now
The principle of existing early treatment.
In addition, also have the assessment that some medical science scales are used for the nervous system disease, such as Action Research Arm
Test (ARAT) scale, Wolf Motor Function Test (WMFT) scales and Fugl-Meyer Assessment
(FMA) scale etc..These scales are used to check the motor function of subject, for example grab, hold, kneading big athletic performance etc., can
With the nervous system disease situation at the current family of excrescence.In the inspection project of these scales, some is daily accessible
Article, such as cup, stone, user need according to scale and doctor requirement, carry out a series of operation, doctor passes through sight
The action of patient is examined, it is given a mark, and subsequent analysis inspection is carried out according to fraction.But these detection methods very consume
It is time-consuming, and need doctor's full name to participate in and score.Meanwhile the method and unresolved it can not be asked in real time with early warning
Topic.
Thus it is possible to be monitored as early as possible and accurately and easily to the nervous system disease and early warning is very important.
The content of the invention
In order to reach the purpose for mitigating working doctor amount, raising inspection efficiency and acceptable degree, the present invention devises one
The nervous system disease monitoring and early warning system of the kind based on daily necessities.By the way that pervasive sensor is combined with daily necessities,
Monitor the situation that user uses daily necessities in daily life.By the collection to data, segmentation and analysis, will have to assess
The upper extremity exercise function of the nervous system disease personnel may be suffered from, early warning further is carried out to the incidence of the nervous system disease
Or auxiliary diagnosis.By the data of quantification, illustrate the motor function situation of the elderly, avoid the uncertainty artificially checked,
And by the way that system is implanted in daily life, solution must not in real time monitor and the problem of early warning.
Specifically, the technical solution adopted by the present invention is as follows：
A kind of the nervous system disease monitoring and early warning system based on daily necessities, including：
The motion sensor being integrated in daily necessities, for gathering exercise data of user when using daily necessities；
The aiding sensors being integrated in daily necessities, for gathering accessory channel number of user when using daily necessities
According to；
Cutting module is acted, cutting is carried out to exercise data for the change according to secondary channel data, user is used
Continuous sexual act during daily necessities is cut into single sub- action (such as stretch out one's hand, turn wrist, translation and hovering)；
Environmental context analysis module, analyzed by the exercise data to having gathered and secondary channel data, with reference to
Current daily necessities use state carries out environmental context analysis；
Sub- action recognition module, for carrying out language to single sub- action according to Environmental context information and secondary channel data
Reason and good sense solution, it is identified as standard son action；
Characteristic extracting module, after carrying out time window division for exercise data corresponding to the standard son action to identifying
Extract time domain and frequency domain character vector；
Model fitting module, for the characteristic vector of the time domain of extraction and frequency domain and the nervous system disease established is auxiliary
Help diagnosis and Early-warning Model to be matched, analyze feature generic, and then judge the nervous system disease situation and classification.
Further, the daily necessities, refer to extensive daily necessities concept, including may move daily necessities and other
It is any can integrated sensor daily necessities, such as interactivity desk, chair, door, tea table, cupboard etc..The removable life
Tableware, kitchen tools, decoration piece, electronic product (such as mobile phone, tablet personal computer, remote control in articles for use, including but not limited to daily life
Device etc.) etc..These articles are built-in or can integrate motion sensor, while user uses them, can capture user action.
In these daily necessitiess, some aiding sensors are integrated, for judging whether user interacts with them, and and its
The use state for other the removable daily necessitiess being in contact.
Further, the motion sensor, including accelerometer, gyroscope and magnetometer, and they are combined
The inertia motion capture systems formed.These motion sensors typically have three axles, but according to be actually needed usable single shaft or
The sensor of twin shaft.Meanwhile according to required precision, can also adjust the firmware program of chip, with set the precision of sensor and
The parameters such as range.
Further, athletic performance of the motion sensor to user records, and sensing data will react user
Action situation, these action situation data will be transferred in computer system, and and be handled and analyzed, according to the system
The diagnosis and Early-warning Model pre-established such as is identified at the processing.Its starting point recorded is triggered by aiding sensors.
Further, the aiding sensors, refer to be integrated in daily necessities, daily necessities process is used in user
In, the sensor that its reading produces significant change can be triggered, including but not limited to pressure sensor, infrared sensor, distance passes
Sensor, temperature sensor and capacitance sensor etc..It can be configured according to actual object and service condition, its installation site also has
Difference, such as bottom of article is installed on for detecting whether being lifted, it is installed on article surface and is used to detect contact situation etc..Its
Numerical value forms secondary channel data, and numerical value change is demonstrated by user and uses the state of daily necessities or action starting/terminating point.
Further, the environmental context analysis module can be according to newest daily necessities use state to environmentally
Hereafter analysis prediction result is modified and adjusted.
Further, the identification technology that the sub- action recognition module uses includes secondary channel data feature recognition, reality
Border life timing information analysis and the analysis of other contextual informations etc..
Further, the characteristic extracting module (is such as trembled, coordination) according to the characteristics of the nervous system disease
The exercise data of motion sensor collection is analyzed, individual features are extracted from the exercise data being originally inputted.
Further, the characteristic extracting module divides correspondingly sized time window according to the species and property of action,
Statistic in time domain level analysis window, form temporal signatures vector；The characteristic extracting module is done quickly to time window
Fourier transformation, the frequency domain distribution information of data in window is obtained, and in the feature of frequency domain level analysis frequency domain, formation frequency domain character
Vector.Extraction of the characteristic extracting module to characteristic vector is to be based on carrying out statistical analysis to large-scale consumer data, before building most
Machine learning, deep learning and the data mining algorithm and model on edge, carry out what successive ignition optimization obtained.
Further, the nervous system disease auxiliary diagnosis and Early-warning Model are that the present invention is acquiring nervous system disease
After the data of patient and normal person, feature (maximum, minimum value, average value system such as in time window in data are extracted
Count frequency distribution feature of feature and frequency domain etc.), then the methods of by the classification and recurrence of machine learning, to the feature selected
Extracted, and train diagnosis and the Early-warning Model of formation.
Further, said system also includes data-optimized module, for being done to the data after action cutting module cutting
The further time preferably (including removes the head and the tail section of data segment after partial preliminary cutting, the length specifically removed with secondary cutting
Degree is determined by action duration and connecting moves length) to ensure in the data of acquisition not switch the number of boundary brought by action
According to.
A kind of the nervous system disease monitoring and method for early warning, its step based on daily necessities include：
1) gather user and use the exercise data and secondary channel data during the daily necessities for being integrated with pervasive sensor.
2) according to the change of secondary channel data, cutting is carried out to the exercise data of motion sensor collection, user is existed
In daily life single sub- action is cut into using continuous action sequence during daily necessities.
3) the single sub- action segmented is identified, the nervous system disease auxiliary for being identified as having established is examined
Standard in the disconnected and modeling list of Early-warning Model act (difference is had according to application scenarios difference, include but is not limited to stretch out one's hand,
Turn wrist, movement, hover, put, grabbing, holding, pinching), to that can not identify or in model, (the nervous system disease established is not auxiliary
Help diagnosis and Early-warning Model) in single sub- action action recorded, if it is multiple occur if recorded by computer, and
Model is adjusted, if seldom occurring, will be dropped.
4) feature of exercise data corresponding to the action of extraction, the change of continuous timing values is converted into a series of time domains
With the feature of frequency domain.
5) time domain of extraction and the feature of frequency domain are entered with the nervous system disease auxiliary diagnosis and Early-warning Model established
Row compares, and analyzes feature generic, and then judge the nervous system disease situation and classification.
Further, the collection user refers to user using the data during daily necessities for being integrated with pervasive sensor
In daily life, in the case of not by sensor disturbance, it is accustomed to normal use daily necessities according to the daily life of oneself,
In these daily necessitiess during by use, wherein integrated sensor is monitored to the service condition of user, gather
User movement data and secondary channel data.
Further, described to carry out motion sensor data cutting according to secondary channel data, its principle is user's day
During often using daily necessities, obvious boundary is had between different son actions, as numerical value increases over suddenly
Threshold value of setting etc., these boundaries can be demarcated using aiding sensors, and carry out further cutting.
Further, the antithetical phrase action is identified, and refers to according to Environmental context information and secondary channel data pair
The current action of user is further analyzed and identified that its identification technology includes secondary channel data feature recognition, reality
Life timing information is analyzed and other contextual informations are analyzed etc..
Further, the feature of exercise data corresponding to the extraction action, refers to the spy according to the nervous system disease
Point (such as trembling, coordination) analyzes exercise data, and the process of individual features is extracted from original input data.Its feature
Correlated characteristic including time domain and frequency domain.
It is further, described to be contrasted feature with the nervous system disease auxiliary diagnosis and Early-warning Model established,
Refer to characteristic vector being input in the diagnosis of the invention established in advance and Early-warning Model, carry out the mistake of machine learning prediction
Journey.
Compared with prior art, the present invention has the advantage that as follows with good effect：
1) inspection of a series of complex is carried out into hospital without patient；
2) accompanied and attended to inspection without doctor's whole process, action of going forward side by side coaches and given a mark；
3) by pervasive sensor integration into daily living article, the system that forms non-intrusion type, by using these days
During normal daily necessities, disease event is monitored and early warning；
4) complete detection model is formed, there is provided more objective inspection and analysis result.
Brief description of the drawings
Fig. 1：Present system structural representation.
Fig. 2：Flow chart of data processing of the present invention.
Embodiment
To enable objects, features and advantages of the present invention to become apparent, hereafter by specific embodiment, and combine attached
Figure, is described in detail.
China has progressed into aging population society, it is following the problem of it is more and more prominent, one of them is important
Aspect is that the pathogenicity rate of the nervous system disease and the death rate remain high, and many pains, pressure are brought to patient, family and society
And burden.First, under many circumstances, patient does not obtain the main reason for timely and suitable treatment is high mortality.Its
Secondary, current inspection is all centered on hospital, and patient only goes to hospital and could checked accordingly.Again, base at present
In the test mode of scale, time-consuming longer, easily there are the moods such as agitation in numerous and diverse checking process in patient, influences to check
As a result.Finally, whole checking process needs the full name of doctor to accompany and attend to and score, and reduces the efficiency of early screening, and due to
The presence artificially diagnosed, certain subjectivity be present.So at present for establishing an effective mechanism, to nervous system disease
Disease, which carries out early warning and tentative diagnosis, active demand, also has obvious meaning for the heavy burden for alleviating family and society.
The nervous system disease monitoring and early warning system of the present invention based on daily necessities, as shown in figure 1, mainly including：Collection
Into the motion sensor and aiding sensors in daily necessities, cutting module, environmental context analysis module, sub- action are acted
Identification module, characteristic extracting module and model fitting module.Based on the system, the present embodiment supports user in daily life,
By using the articles for daily use for being integrated with non-intruding sensor, the motor function showed during articles for daily use is used to user
Etc. being evaluated, diagnosis early stage and early warning further are carried out to the nervous system disease.
Specific implementation is based on daily life scene, has a variety of intelligent daily living articles to collectively form a system, from
The all angles and aspect of daily life carry out understanding analysis to user behavior and action.For example when user opens the door, lock core connects
The positions such as axle, door handle are integrated with the IMU for perceiving motion, and the behavior act to be opened the door to user carries out capture analysis.Eaten in user
During meal, its diet articles for use, such as dish chopsticks spoon fork on be implanted motion sensor, capture its action having a meal, these action compared with
Horn of plenty, on medium article and aiding sensors are disposed on desktop, cutting is carried out to user behavior, user touches these matchmakers
Jie's article and it will be all recorded using their mode.In addition, user of toothbrushing, read, play chess when, tooth tooth glass,
Motion sensor and aiding sensors, the row of user in the case of capture is more can be targetedly implanted on spine, chess piece
For.
When using this system, the captured action analyzed includes but is not limited to user：Contact door handle, insert
Enter key (translation is mobile), turn key (turning wrist), revolving door handle (turning wrist), scoop up/put down article in bowl (turning wrist), by one
Article moves on in another bowl (translation) in bowl, holding spoon, motionless/find target (static), pick up/puts down spoon (grasping), scoops food
Thing, brush teeth, lift book (static), pick up chess piece (pinching), put down chess piece (putting) etc..In these actions, there is triggering accordingly to tremble
The action of type, such as hold spoon and can trigger static tremor by book, mobile object can trigger motility and tremble, and these are different
Type of trembling from different the nervous system diseases have obvious association again.
After Fig. 2 gives capture user daily behavior action, the flow analyzed.
To the exercise data (this example is captured by IMU) of capture, using secondary channel data, (this example is connect by daily necessities first
Capacitance sensor capture at the pressure sensor of contact and holding) action to user carries out son action cutting.After cutting
Action message only has a syntactic information, i.e., system can only judge to distinguish every height action, can not know sub- action be all represent it is assorted
Act implication.The environmental context analysis module of this system is according to the action sequence information and current aiding sensors analyzed
The semanteme of current son action is identified analysis result, by sub- action it is qualitative for contact door handle, revolving door handle, contact spoon,
Pick up spoon, scoop ball, put ball, put spoon, brush teeth, by book, pinch chess piece, put chess piece etc. action.
Then for the data segmented, what is carried out is that it is preferably cut through the row further time with secondary
Point, the data boundary not brought with the data ensured in acquisition by action switching.The specific reaction because of human body needs certain
Time, so each action front portion and afterbody be likely to some data for action conversion when exercise data,
There is unstability in this componental movement data, the analytic band of action can be adversely affected.
In secondary dicing process, first, molar behavior data are finely divided, are divided into isometric analysis window, and
The dispersion degree of data in each window is calculated, if dispersion degree is excessive compared to average value, and window is in exercise data stem
Or afterbody, then it is conversion window to mark this window, and the first conversion section is carried out further according to the continuity degree of multiple conversion windows
Overall labeling and removal.Secondly, according to the species and property of action, correspondingly sized time window is divided, in time domain aspect point
The statistic in window, such as maximum, minimum value, average value, kurtosis, degree of skewness, Wave crest and wave trough information etc. are analysed, it is special to form time domain
Sign vector, these vectors represent distribution situation of the data in time domain, have positive meaning for obtaining data overall evaluation index
Justice.Again, Fast Fourier Transform (FFT) is done to time window, obtains the frequency domain distribution information of data in window, and in frequency domain level analysis
Some features of frequency domain, such as crest number and distribution, form frequency domain character vector, and these vectors represent data on frequency domain
Frequency distribution situation, there is positive effect for analyze data spectrum distribution index.Finally, according to the time domain of extraction and the spy of frequency domain
Sign vector, is selected useful feature, and this selection determined based on the analysis to big data, and with having established
Model is matched, and matching process includes but is not limited to machine learning, deep learning, data digging method etc., and final obtain is used
Family whether ill aspect or have what disease aspect classification results.
The present embodiment is tested in the crowd of 300 people, and this experiment is a community medicine research project, is led to
Cross and whole-sample acquisition is carried out to more than 30 years old crowd in whole community.In participant, there are 83 males, 217 women,
Age bracket average age 58.6 years old, reached the age distribution requirement needed for experimental analysis at 34 to 86 years old.Wherein there are 16 subjects
For the patient of the nervous system disease.
In example laboratory, the daily living article for being integrated with sensor is placed in participant family by we, allows them
It is accustomed to carrying out the daily life system of oneself according to daily life.After carrying out data acquisition in daily life, using in the system
The structure the nervous system disease monitoring of Data Management Analysis method is (i.e. dynamic to action progress cutting acquisition of collection with Early-warning Model
Make, secondary cutting obtains the son action without action conversion, action window is divided, extracts time domain and frequency domain in window
Feature is simultaneously analyzed, builds model), training data of the system using 80% data instance as structure model, use other 20%
Data instance is monitored to the nervous system disease and assessed with Early-warning Model.Tentative diagnosis of the system to the nervous system disease/pre-
Alert accuracy rate is 92.99%.Meanwhile the habits and customs for subject also have more good differentiation accuracy rate, as strong hand is sentenced
Other accuracy rate is 95.49%, and the differentiation accuracy rate of smoking is 87.97%, and the differentiation accuracy rate drunk is 78.95%.These ginsengs
Early warning and diagnosis of the number for disease have certain reference significance.
Claims (10)
1. a kind of the nervous system disease monitoring and early warning system based on daily necessities, including：
The motion sensor being integrated in daily necessities, for gathering exercise data of user when using daily necessities；
The aiding sensors being integrated in daily necessities, for gathering secondary channel data of user when using daily necessities；
Cutting module is acted, cutting is carried out to exercise data for the change according to secondary channel data, user is used into life
Continuous sexual act during articles for use is cut into single sub- action；
Environmental context analysis module, analyzed by the exercise data to having gathered and secondary channel data, with reference to current
Daily necessities use state carries out environmental context analysis；
Sub- action recognition module, for carrying out semantic reason to single sub- action according to Environmental context information and secondary channel data
Solution, it is identified as standard son action；
Characteristic extracting module, extracted after carrying out time window division for exercise data corresponding to the standard son action to identifying
Time domain and frequency domain character vector；
Model fitting module, for the characteristic vector of the time domain of extraction and frequency domain to be examined with the nervous system disease auxiliary established
Disconnected and Early-warning Model is matched, and analyzes feature generic, and then judge the nervous system disease situation and classification.
2. the nervous system disease monitoring based on daily necessities and early warning system as claimed in claim 1, it is characterised in that institute
Motion sensor, including accelerometer, gyroscope and magnetometer are stated, and the inertia motion capture that they are combined is
System.
3. the nervous system disease monitoring based on daily necessities and early warning system as claimed in claim 1, it is characterised in that institute
Stating aiding sensors includes pressure sensor, infrared sensor, range sensor, temperature sensor and capacitance sensor.
4. the nervous system disease monitoring based on daily necessities and early warning system as claimed in claim 1, it is characterised in that institute
The starting point for stating motion sensor record is triggered by aiding sensors.
5. the nervous system disease monitoring based on daily necessities and early warning system as claimed in claim 1, it is characterised in that institute
State the identification technology that sub- action recognition module uses include secondary channel data feature recognition, the analysis of real life timing information and
Contextual information is analyzed.
6. the nervous system disease monitoring based on daily necessities and early warning system as claimed in claim 1, it is characterised in that institute
The action data that characteristic extracting module analyzes motion sensor collection according to the characteristics of the nervous system disease is stated, from what is be originally inputted
Individual features are extracted in exercise data.
7. the nervous system disease monitoring based on daily necessities and early warning system as claimed in claim 1, it is characterised in that institute
Species and property of the characteristic extracting module according to action are stated, divides correspondingly sized time window, in time domain level analysis window
Statistic, formed temporal signatures vector；The characteristic extracting module does Fast Fourier Transform (FFT) to time window, obtains in window
The frequency domain distribution information of data, and in the feature of frequency domain level analysis frequency domain, formation frequency domain character vector.
8. the nervous system disease monitoring based on daily necessities and early warning system as claimed in claim 7, it is characterised in that
Statistic in time domain layer face-port includes：Maximum, minimum value, average value, kurtosis, degree of skewness and Wave crest and wave trough information.
9. the nervous system disease monitoring based on daily necessities and early warning system as claimed in claim 1, it is characterised in that institute
It is after the data of the nervous system disease patient and normal person are acquired to state the nervous system disease auxiliary diagnosis and Early-warning Model, is carried
Feature of the access in, then the classification by machine learning and homing method, are extracted, and train shape to the feature selected
Into diagnosis and Early-warning Model.
10. the nervous system disease monitoring based on daily necessities and early warning system as claimed in claim 1, it is characterised in that
Also include data-optimized module, preferably cut for doing the further time to the data after action cutting module cutting with secondary
Point, to ensure the data boundary not brought in the data of acquisition by action switching.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610371171.4A CN107451385B (en) | 2016-05-30 | 2016-05-30 | Nervous system disease monitoring and early warning system based on articles for daily use |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610371171.4A CN107451385B (en) | 2016-05-30 | 2016-05-30 | Nervous system disease monitoring and early warning system based on articles for daily use |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107451385A true CN107451385A (en) | 2017-12-08 |
| CN107451385B CN107451385B (en) | 2021-04-27 |
Family
ID=60484818
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610371171.4A Active CN107451385B (en) | 2016-05-30 | 2016-05-30 | Nervous system disease monitoring and early warning system based on articles for daily use |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107451385B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108962397A (en) * | 2018-06-06 | 2018-12-07 | 中国科学院软件研究所 | A kind of multichannel multitask the nervous system disease assistant diagnosis system based on pen and voice |
| CN110967311A (en) * | 2019-11-25 | 2020-04-07 | 中国科学院地球化学研究所 | Porphyry deposit alteration zone identification method based on infrared spectrum and magnetic susceptibility measurement |
| CN111297366A (en) * | 2018-12-12 | 2020-06-19 | 中国科学院软件研究所 | Data processing method and diagnosis device for assisting disease diagnosis based on daily necessities |
| CN111968741A (en) * | 2020-07-15 | 2020-11-20 | 华南理工大学 | Diabetes complication high-risk early warning system based on deep learning and integrated learning |
| CN115881287A (en) * | 2023-03-03 | 2023-03-31 | 四川互慧软件有限公司 | Doctor recommendation method based on data collected by ECG (electrocardiograph) monitor |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103781405A (en) * | 2011-07-05 | 2014-05-07 | 沙特阿拉伯石油公司 | Systems, computer medium and computer-implemented methods for monitoring and improving health and productivity of employees |
| CN103781409A (en) * | 2011-07-05 | 2014-05-07 | 沙特阿拉伯石油公司 | Systems, computer medium and computer-implemented methods for monitoring and improving biomechanical health of employees |
| CN103782300A (en) * | 2011-07-05 | 2014-05-07 | 沙特阿拉伯石油公司 | Systems, computer medium and computer-implemented methods for monitoring health of employees using mobile devices |
| US20140194701A1 (en) * | 2001-06-29 | 2014-07-10 | Philometron, Inc. | Gateway platform for biological monitoring and delivery of therapeutic compounds |
| US20140349257A1 (en) * | 2013-05-23 | 2014-11-27 | Robert A. Connor | Smart Watch and Food Utensil for Monitoring Food Consumption |
| CN105232000A (en) * | 2015-10-29 | 2016-01-13 | 四川大学华西医院 | Epilepsy detection device and method |
-
2016
- 2016-05-30 CN CN201610371171.4A patent/CN107451385B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140194701A1 (en) * | 2001-06-29 | 2014-07-10 | Philometron, Inc. | Gateway platform for biological monitoring and delivery of therapeutic compounds |
| CN103781405A (en) * | 2011-07-05 | 2014-05-07 | 沙特阿拉伯石油公司 | Systems, computer medium and computer-implemented methods for monitoring and improving health and productivity of employees |
| CN103781409A (en) * | 2011-07-05 | 2014-05-07 | 沙特阿拉伯石油公司 | Systems, computer medium and computer-implemented methods for monitoring and improving biomechanical health of employees |
| CN103782300A (en) * | 2011-07-05 | 2014-05-07 | 沙特阿拉伯石油公司 | Systems, computer medium and computer-implemented methods for monitoring health of employees using mobile devices |
| US20140349257A1 (en) * | 2013-05-23 | 2014-11-27 | Robert A. Connor | Smart Watch and Food Utensil for Monitoring Food Consumption |
| CN105232000A (en) * | 2015-10-29 | 2016-01-13 | 四川大学华西医院 | Epilepsy detection device and method |
Non-Patent Citations (1)
| Title |
|---|
| YINENG CHEN ETC: "UbiSpoon: Pervasive Monitoring of Nervous System Diseases through Daily Life", 《CHI 2014》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108962397A (en) * | 2018-06-06 | 2018-12-07 | 中国科学院软件研究所 | A kind of multichannel multitask the nervous system disease assistant diagnosis system based on pen and voice |
| CN111297366A (en) * | 2018-12-12 | 2020-06-19 | 中国科学院软件研究所 | Data processing method and diagnosis device for assisting disease diagnosis based on daily necessities |
| CN111297366B (en) * | 2018-12-12 | 2021-07-06 | 中国科学院软件研究所 | Data processing method and diagnosis device for assisting disease diagnosis based on daily necessities |
| CN110967311A (en) * | 2019-11-25 | 2020-04-07 | 中国科学院地球化学研究所 | Porphyry deposit alteration zone identification method based on infrared spectrum and magnetic susceptibility measurement |
| CN111968741A (en) * | 2020-07-15 | 2020-11-20 | 华南理工大学 | Diabetes complication high-risk early warning system based on deep learning and integrated learning |
| CN111968741B (en) * | 2020-07-15 | 2023-07-18 | 华南理工大学 | Deep learning and integrated learning-based diabetes complication high-risk early warning system |
| CN115881287A (en) * | 2023-03-03 | 2023-03-31 | 四川互慧软件有限公司 | Doctor recommendation method based on data collected by ECG (electrocardiograph) monitor |
| CN115881287B (en) * | 2023-03-03 | 2023-05-23 | 四川互慧软件有限公司 | Doctor recommendation method based on electrocardiograph monitor acquisition data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107451385B (en) | 2021-04-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107451385A (en) | A kind of the nervous system disease monitoring and early warning system based on daily necessities | |
| Amft et al. | Detection of eating and drinking arm gestures using inertial body-worn sensors | |
| Rahman et al. | Predicting" about-to-eat" moments for just-in-time eating intervention | |
| Mirtchouk et al. | Automated estimation of food type and amount consumed from body-worn audio and motion sensors | |
| Tattersall | Species recognition in human paleontology | |
| Chen et al. | Computerized wrist pulse signal diagnosis using modified auto-regressive models | |
| Dong et al. | Detecting periods of eating during free-living by tracking wrist motion | |
| Patel et al. | Tracking motor recovery in stroke survivors undergoing rehabilitation using wearable technology | |
| Plötz et al. | Activity recognition and healthier food preparation | |
| CN108523869A (en) | The method and system of the coronary artery disease in people is detected for using fusion method | |
| CN103927851B (en) | A kind of individualized multi thresholds fall detection method and system | |
| Cömert et al. | Cardiotocography signals with artificial neural network and extreme learning machine | |
| CN104269025B (en) | Wearable single node feature and the position choosing method of monitoring is fallen down towards open air | |
| Zhang et al. | When generalized eating detection machine learning models fail in the field | |
| Singh et al. | Classification of short unsegmented heart sound based on deep learning | |
| Mattfeld et al. | Measuring the consumption of individual solid and liquid bites using a table-embedded scale during unrestricted eating | |
| CN106901689B (en) | Cognitive and motor function detection system based on physical interaction and data processing method | |
| CN110313900A (en) | It is a kind of based on deep learning to the classification method of knee joint swinging signal | |
| Acharya et al. | Automated screening tool for dry and wet age-related macular degeneration (ARMD) using pyramid of histogram of oriented gradients (PHOG) and nonlinear features | |
| CN115040086A (en) | Data processing system and method based on digital biomarkers | |
| Wagner et al. | A wavelet-based approach to monitoring Parkinson's disease symptoms | |
| CN106503667B (en) | A kind of fall detection method based on WISP and pattern-recognition | |
| Bi et al. | Pervasive eating habits monitoring and recognition through a wearable acoustic sensor | |
| KR102632272B1 (en) | Classification method of atrial fibrillation and congestive heart failure using a convolutional artificial neural network | |
| Kim et al. | Eating activity recognition for health and wellness: A case study on asian eating style |
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 |