CN109887605A - A kind of mechanism long-term care hospital stays pattern analysis method based on model - Google Patents
A kind of mechanism long-term care hospital stays pattern analysis method based on model Download PDFInfo
- Publication number
- CN109887605A CN109887605A CN201910144639.XA CN201910144639A CN109887605A CN 109887605 A CN109887605 A CN 109887605A CN 201910144639 A CN201910144639 A CN 201910144639A CN 109887605 A CN109887605 A CN 109887605A
- Authority
- CN
- China
- Prior art keywords
- model
- term care
- resident
- long
- analysis method
- 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.)
- Pending
Links
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The present invention discloses a kind of mechanism long-term care hospital stays pattern analysis method based on model, include the following steps: the mechanism for introducing long-term care (LTC): introducing the mechanism of long-term care (LTC) to patient and related personnel, LTC mainly includes the house nursing (RC) and nursing (NC) provided by the community service project and Aged caring institutions institute of local authority's operation.The mechanism long-term care hospital stays pattern analysis method based on model, by Markov model fitting data collection, first stage fitting shows that RC is a kind of state, NC is two states, simultaneously with the survivor of Kaplan-Meier type estimation and the two kinds of genders of RC and NC estimated with Markov model, the structure of Markov model is determined with this, and corresponding data are analyzed and processed, can be very good data required for calculating according to every property parameters of resident by mechanism long-term care hospital stays pattern analysis method and carry out corresponding management measure.
Description
Technical field
The present invention relates to field of medical technology, specially a kind of mechanism long-term care hospital stays mode based on model point
Analysis method.
Background technique
The elderly is often subjected to physically and mentally healthy and number of storage tanks produced per day decline, for example, feed, go to the toilet can with self nursing
It can become difficult, when old man cannot obtain looking after at home, it is necessary to which the help of the mechanism of long-term care (LTC), LTC are provided
It mainly include the house nursing (RC) and nursing provided by the community service project and Aged caring institutions institute of local authority's operation
(NC), in general, RC is made of the board of directors and personal nursing, in poor health but remain to management number of storage tanks produced per day for those
People provides service, and NC needs NHS then for body steadiness but the higher the elderly of body & mind disabled degree provides service
The investment of (national health service) registered nurse.
The analysis of hospital stays data belongs to a statistical branch, referred to as survival analysis, it is usually by the hospital stays
The tool that data influence life span as the different patient characteristics of research.However, the data mining for analyzing Survival data
Method (such as method based on decision rule and artificial neural network) is usually directed to one group of given patient characteristic (after treating
Dead or survival in 3 years) classification existence result prediction.For these methods, generate the bottoms of observation service level data with
Machine process usually implicitly models.Other methods, such as method based on flow model and random process, concentrate on to bottom mistake
The clearly modeling of journey, it is therefore an objective to capture the high-level patterns of residence time.
In the prior art, it is difficult to pass through the mechanism long-term care hospital stays according to every property parameters of resident well
Data required for pattern analysis method calculates simultaneously carry out corresponding management measure, in certain journey while reducing efficiency of service
The expenditure of cost is increased on degree, at the same be easy to appear due to data it is more caused by managerial confusion.
Summary of the invention
(1) the technical issues of solving
In view of the deficiencies of the prior art, the mechanism long-term care hospital stays mode based on model that the present invention provides a kind of
Analysis method solves in the prior art, it is difficult to pass through mechanism long-term care according to every property parameters of resident well
The problem of hospital stays pattern analysis method calculates required data and carries out corresponding management measure.
(2) technical solution
In order to achieve the above object, the present invention is achieved by the following technical programs: a kind of mechanism based on model is long-term
Hospital stays pattern analysis method is nursed, is included the following steps:
S1, the mechanism for introducing long-term care (LTC): introducing the mechanism of long-term care (LTC) to patient and related personnel,
LTC mainly includes the house nursing (RC) provided by the community service project and Aged caring institutions institute of local authority's operation and protects
It manages (NC)；
S2, data are collected: periodically collects the data by its resident looked after collected from local authority, admission date is admitted to hospital
Place, transfer date and place and discharge date, and be stored in database after being recorded as table；
S3, the flowing for simulating the ILTC middle-aged and the old: use state combination flows feelings to capture the resident of each type nursing
Condition, and be recorded in database, the residence time in RC and NC follows coxian distribution, A class probability density function form are as follows:
fA(t)=- ΦT Aexp(QAAt)QAA 1, wherein Φ A is the column vector for entering class A probability by each member condition；
S4, the feature for integrating resident: the P attribute X:x of each type of resident in simultaneously recording step S3 is detected1、x2、
x3...xp(transition rate of attribute i to the j of k-th of resident is written as: qI, j, k=exp (βT ijxk), wherein (i ≠ j))；
The residence time mode state of public funding resident is distinguished in S5, ILTC: by Markov model fitting data collection,
First stage fitting shows that RC is a kind of state, and NC is two states；
The feature of the residence time pattern differentials of S6, NC resident is included in: characteristic being included in mould using two different methods
Type, i.e. model 1: as residence time of the continuous variable in RC, model 2: as previous presence of the binary variable in RC
Time；
The gender differences record of S7, residence time mode: it takes gender as distinguishing characteristics and distinguishes resident and be recorded in data
In library；
S8, the structure for determining Markov model: pass through formula: L (θ)=(nΣi-1)log{l(θ|ci, ti), calculate n
To the log likelihood of c and t, the value of one group of θ is then selected, log-likelihood function is maximized, Markov model is fitted
Into observation data, Markov model is fitted to entire dwell data length.
Preferably, in step s3, used combinations of states is short stay state or extended stationary periods state.
Preferably, in the step s 7, with Kaplan-Meier type estimate and with Markov model estimate RC and two kinds of NC
The survivor of gender.
Preferably, in step s 8, all by using general optimizer, (Matlab is provided excellent all models fittings
Change device) maximize what log-likelihood function was realized.
(3) beneficial effect
The present invention provides a kind of mechanism long-term care hospital stays pattern analysis method based on model.Having following has
Beneficial effect:
The mechanism long-term care hospital stays pattern analysis method based on model, by Markov model fitting data
Collection, first stage fitting show that RC is a kind of state, and NC is two states, while being estimated and being used with Kaplan-Meier type
The survivor of two kinds of genders of RC and NC of Markov model estimation, the structure of Markov model is determined with this, and to corresponding
Data are analyzed and processed, and can be very good to pass through mechanism long-term care hospital stays mode according to every property parameters of resident
Analysis method calculate required for data simultaneously carry out corresponding management measure, improve efficiency of service while to a certain extent
Reduce the expenditure of cost, at the same avoid due to data it is more caused by managerial confusion.
Specific embodiment
Described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Based on the present invention
In embodiment, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
A kind of mechanism long-term care hospital stays pattern analysis method based on model, includes the following steps:
S1, the mechanism for introducing long-term care (LTC): introducing the mechanism of long-term care (LTC) to patient and related personnel,
LTC mainly includes the house nursing (RC) provided by the community service project and Aged caring institutions institute of local authority's operation and protects
It manages (NC)；
S2, data are collected: periodically collects the data by its resident looked after collected from local authority, admission date is admitted to hospital
Place, transfer date and place and discharge date, and be stored in database after being recorded as table；
S3, the flowing for simulating the ILTC middle-aged and the old: use state combination flows feelings to capture the resident of each type nursing
Condition, and be recorded in database, the residence time in RC and NC follows coxian distribution, A class probability density function form are as follows:
fA(t)=- ΦT Aexp(QAAt)QAA 1, wherein Φ A be entered by each member condition class A probability column vector it is (used
Combinations of states is short stay state or extended stationary periods state)；
S4, the feature for integrating resident: the P attribute X:x of each type of resident in simultaneously recording step S3 is detected1、x2、
x3...xp(transition rate of attribute i to the j of k-th of resident is written as: qI, j, k=exp (βT ijxk), wherein (i ≠ j))；
The residence time mode state of public funding resident is distinguished in S5, ILTC: by Markov model fitting data collection,
First stage fitting shows that RC is a kind of state, and NC is two states；
The feature of the residence time pattern differentials of S6, NC resident is included in: characteristic being included in mould using two different methods
Type, i.e. model 1: as residence time of the continuous variable in RC, model 2: as previous presence of the binary variable in RC
Time；
The gender differences record of S7, residence time mode: it takes gender as distinguishing characteristics and distinguishes resident and be recorded in data
(with the survivor of Kaplan-Meier type estimation and the two kinds of genders of RC and NC estimated with Markov model) in library；
S8, the structure for determining Markov model: pass through formula: L (θ)=(nΣi-1)log{l(θ|ci, ti), calculate n
To the log likelihood of c and t, the value of one group of θ is then selected, log-likelihood function is maximized, Markov model is fitted
Into observation data, Markov model is fitted to entire dwell data length, and (all models fittings are all by using logical
Optimizer (optimizer that Matlab is provided) maximizes what log-likelihood function was realized).
In conclusion the mechanism long-term care hospital stays pattern analysis method based on model is somebody's turn to do, by Markov model
Fitting data collection, first stage fitting show that RC is a kind of state, and NC is two states, while being estimated with Kaplan-Meier type
With the survivor for the two kinds of genders of RC and NC estimated with Markov model, the structure of Markov model is determined with this, and to phase
The data answered are analyzed and processed, and can be very good to pass through the mechanism long-term care hospital stays according to every property parameters of resident
Data required for pattern analysis method calculates simultaneously carry out corresponding management measure, in certain journey while improving efficiency of service
Reduce the expenditure of cost on degree, at the same avoid due to data it is more caused by managerial confusion.
It although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
A variety of variations, modification, replacement can be carried out to these embodiments without departing from the principles and spirit of the present invention by understanding
And modification, the scope of the present invention is defined by the appended.
Claims (4)
1. a kind of mechanism long-term care hospital stays pattern analysis method based on model, which comprises the steps of:
S1, the mechanism for introducing long-term care (LTC): the mechanism of long-term care (LTC), LTC are introduced to patient and related personnel
It mainly include the house nursing (RC) and nursing provided by the community service project and Aged caring institutions institute of local authority's operation
(NC)；
S2, collect data: periodically collected collected from local authority by its look after resident data, admission date, place of being admitted to hospital,
Date and place and discharge date are shifted, and is stored in database after being recorded as table；
S3, the flowing for simulating the ILTC middle-aged and the old: use state combines to capture resident's mobility status of each type nursing, and
It is recorded in database, the residence time in RC and NC follows coxian distribution, A class probability density function form are as follows: fA(t)
=-ΦT Aexp(QAAt)QAA 1, wherein Φ A is the column vector for entering class A probability by each member condition；
S4, the feature for integrating resident: the P attribute X:x of each type of resident in simultaneously recording step S3 is detected1、x2、x3...xp
(transition rate of attribute i to the j of k-th of resident is written as: qI, j, k=exp (βT ijxk), wherein (i ≠ j))；
The residence time mode state of public funding resident is distinguished in S5, ILTC: by Markov model fitting data collection, first
Staged matching shows that RC is a kind of state, and NC is two states；
The feature of the residence time pattern differentials of S6, NC resident is included in: characteristic being included in model using two different methods, i.e.,
Model 1: as residence time of the continuous variable in RC, model 2: as binary variable in RC previously there are the times；
The gender differences record of S7, residence time mode: it takes gender as distinguishing characteristics and distinguishes resident and be recorded in database；
S8, the structure for determining Markov model: pass through formula: L (θ)=(nΣi-1)log{l(θ|ci, ti), calculate n to c and
The log likelihood of t then selects the value of one group of θ, maximizes log-likelihood function, Markov model is fitted to observation
In data, Markov model is fitted to entire dwell data length.
2. a kind of mechanism long-term care hospital stays pattern analysis method based on model according to claim 1, special
Sign is: in step s3, used combinations of states is short stay state or extended stationary periods state.
3. a kind of mechanism long-term care hospital stays pattern analysis method based on model according to claim 1, special
Sign is: in the step s 7, with the good fortune of Kaplan-Meier type estimation and the two kinds of genders of RC and NC estimated with Markov model
The person of depositing.
4. a kind of mechanism long-term care hospital stays pattern analysis method based on model according to claim 1, special
Sign is: in step s 8, all models fittings all by using general optimizer (optimizer that Matlab is provided) most
What bigization log-likelihood function was realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910144639.XA CN109887605A (en) | 2019-02-27 | 2019-02-27 | A kind of mechanism long-term care hospital stays pattern analysis method based on model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910144639.XA CN109887605A (en) | 2019-02-27 | 2019-02-27 | A kind of mechanism long-term care hospital stays pattern analysis method based on model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109887605A true CN109887605A (en) | 2019-06-14 |
Family
ID=66929573
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910144639.XA Pending CN109887605A (en) | 2019-02-27 | 2019-02-27 | A kind of mechanism long-term care hospital stays pattern analysis method based on model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109887605A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113393939A (en) * | 2021-04-26 | 2021-09-14 | 上海米健信息技术有限公司 | Intensive care unit patient hospitalization day prediction method and system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107180283A (en) * | 2017-07-05 | 2017-09-19 | 山东大学 | A kind of behavior prediction system and method for being in hospital again combined based on optimal characteristics |
-
2019
- 2019-02-27 CN CN201910144639.XA patent/CN109887605A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107180283A (en) * | 2017-07-05 | 2017-09-19 | 山东大学 | A kind of behavior prediction system and method for being in hospital again combined based on optimal characteristics |
Non-Patent Citations (1)
| Title |
|---|
| HAIFENG XIE等: "A Model-Based Approach to the Analysis of Patterns of Length of Stay in Institutional Long-Term Care", 《IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN BIOMEDICINE》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113393939A (en) * | 2021-04-26 | 2021-09-14 | 上海米健信息技术有限公司 | Intensive care unit patient hospitalization day prediction method and system |
| CN113393939B (en) * | 2021-04-26 | 2024-05-28 | 上海米健信息技术有限公司 | Method and system for predicting number of hospitalization days of intensive care unit patient |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shibata et al. | Detecting emerging research fronts based on topological measures in citation networks of scientific publications | |
| Moore et al. | Characterizing social environment's association with neurocognition using census and crime data linked to the Philadelphia Neurodevelopmental Cohort | |
| CN107247881A (en) | A kind of multi-modal intelligent analysis method and system | |
| CN109841282A (en) | A kind of Chinese medicine health control cloud system and its building method based on cloud computing | |
| CN111081379B (en) | Disease probability decision method and system thereof | |
| CN104462858A (en) | Health warning method based on multi-order hidden Markov model | |
| Sprint et al. | Behavioral differences between subject groups identified using smart homes and change point detection | |
| Elbayoudi et al. | The human behaviour indicator: A measure of behavioural evolution | |
| Nandy et al. | Intelligent health monitoring system for detection of symptomatic/asymptomatic COVID-19 patient | |
| CN107833633A (en) | A kind of method that hypertensive patient's follow-up is recommended | |
| CN106446560A (en) | Hyperlipidemia prediction method and prediction system based on incremental neural network model | |
| CN106446552A (en) | Prediction method and prediction system for sleep disorder based on incremental neural network model | |
| Yin et al. | Unsupervised daily routine and activity discovery in smart homes | |
| CN109887605A (en) | A kind of mechanism long-term care hospital stays pattern analysis method based on model | |
| Hajihashemi et al. | Detecting daily routines of older adults using sensor time series clustering | |
| Chaari et al. | Comparative survey of multigraph integration methods for holistic brain connectivity mapping | |
| Shahid et al. | Recognizing long-term sleep behaviour change using clustering for elderly in smart homes | |
| JP7365747B1 (en) | Disease treatment process abnormality identification system based on hierarchical neural network | |
| CN106295238A (en) | A kind of hypertensive nephropathy Forecasting Methodology based on increment type neural network model and prognoses system | |
| Thangarasu et al. | Prediction of hidden knowledge from clinical database using data mining techniques | |
| Howedi et al. | A multi-scale fuzzy entropy measure for anomaly detection in activities of daily living | |
| Kim et al. | Self health diagnosis system for Korean traditional medicine with enhanced ART2 | |
| CN113517044A (en) | Clinical data processing method and system for evaluating citicoline based on pharmacokinetics | |
| CN106202986A (en) | A kind of tonsillitis Forecasting Methodology based on increment type neural network model and prognoses system | |
| Shen et al. | LGSleepNet: An Automatic Sleep Staging Model Based on Local and Global Representation Learning |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190614 |