CN109002442B - Device and method for searching diagnosis cases based on doctor related attributes - Google Patents
Device and method for searching diagnosis cases based on doctor related attributes Download PDFInfo
- Publication number
- CN109002442B CN109002442B CN201710419900.3A CN201710419900A CN109002442B CN 109002442 B CN109002442 B CN 109002442B CN 201710419900 A CN201710419900 A CN 201710419900A CN 109002442 B CN109002442 B CN 109002442B
- Authority
- CN
- China
- Prior art keywords
- case
- diagnosis
- matrix
- column
- row
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention provides a device and a method for searching diagnosis cases based on doctor related attributes, wherein the device comprises a keyword input part (11), a diagnosis case searching part (12), a first matrix establishing part (13), a relation set establishing part (14), a second matrix establishing part (15), a similarity calculating part (16) and a diagnosis case recommending part (17). According to the device and the method, the searched diagnosis cases can be ordered according to the similarity of the attribute related to the doctor of each diagnosis case and the attribute related to the doctor designated by the user, and the diagnosis cases are recommended according to the order, so that the user can quickly obtain the expected diagnosis cases.
Description
Technical Field
The invention relates to a device and a method for searching diagnosis cases based on doctor related attributes.
Background
Patent document 1 (chinese patent application 201310603059.5) discloses a medical record query method. The method comprises the following steps: when a storage instruction for medical records is received, analyzing the medical records to obtain multiple diagnosis and treatment information and characteristic data corresponding to each diagnosis and treatment information; according to the structural relation among the diagnosis and treatment information, carrying out structural storage on the diagnosis and treatment information, wherein each diagnosis and treatment information and corresponding characteristic data are stored in an associated mode; analyzing the query instruction to determine target diagnosis and treatment information and target keywords, and querying target feature data matched with the target keywords in feature data associated with the target diagnosis and treatment information.
According to the technical scheme of the patent document 1, diagnosis and treatment information in medical records can be stored in a structuring mode, so that a user can inquire about the duration of the disease, the stored diagnosis and treatment information can be inquired directly according to inquiry conditions, full-text retrieval of the medical records is not needed, and inquiry efficiency is improved.
However, after retrieving medical records, the user can sometimes obtain a lot of medical records conforming to the retrieval conditions. In this case, the user can judge which medical record is the desired medical record by reading the retrieved medical records one by one. In this way, the user spends a lot of time screening the medical records, which causes great trouble to the user.
Disclosure of Invention
Problems to be solved by the invention
The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a device and a method capable of retrieving a diagnosis case based on a doctor-related attribute, thereby rapidly obtaining a desired diagnosis case.
Technical means for solving the problems
The invention relates to a device for searching diagnosis cases based on doctor related attributes, which comprises: a keyword input unit for inputting a keyword; a diagnosis case search unit that searches a diagnosis case sharing library using the keyword; a first matrix creation section that creates a first matrix that is a two-dimensional matrix of diagnostic case-diagnostic case sharing library attributes, the first matrix creation section creating the first matrix by: taking all the attributes described by the doctor diagnosis and treatment experience in the diagnosis case sharing library, namely the diagnosis case sharing library attribute as the column name of a matrix, taking the diagnosis case ID searched by the diagnosis case searching part and the reference attribute Id described by the doctor diagnosis and treatment experience designated by a user as row names, taking the reference attribute Id as a first row, and marking the value of a row and column of the first matrix as 1 when the diagnosis case ID or the reference attribute Id represented by a certain row of the first matrix has the diagnosis case sharing library attribute represented by a certain column of the first matrix, otherwise marking the value of the row and column of the first matrix as 0; a relationship set creation unit that creates a relationship set, which is a relationship set of a diagnosis case shared library attribute-a diagnosis case, from the first matrix, the relationship set creation unit creating the relationship set by: the relation set of each diagnosis case sharing library attribute comprises all diagnosis case IDs or reference attribute Ids with column marks of 1 corresponding to the diagnosis case sharing library attribute in the first matrix; a second matrix creation unit that creates a second matrix, which is a two-dimensional matrix of diagnosis cases-diagnosis cases, from the relation set, the second matrix creation unit creating the second matrix by: taking the diagnostic case ID and the reference attribute ID as row names and column names, taking the reference attribute ID as a first row and a first column, taking the number of combinations of different diagnostic case IDs in the relation set of each diagnostic case sharing library attribute, or the number of combinations of the diagnostic case ID and the reference attribute ID as the value of the corresponding row and column, and marking the value of which the row and the column represent the same diagnostic case ID or both represent the reference attribute ID as 0 in the second matrix; a similarity calculation unit that calculates a similarity between each diagnosis case ID and the reference attribute ID based on the second matrix; and a diagnosis case recommending unit that sorts the diagnosis case IDs according to the similarity, and recommends the diagnosis case IDs in the order of arrangement.
In addition, the method for searching diagnosis cases based on doctor related attributes according to the present invention includes: a keyword input step for inputting keywords; a step of searching for the diagnosis cases, which uses the keywords to search a shared library of diagnosis cases; a first matrix establishing step of establishing a first matrix, which is a two-dimensional matrix of diagnostic case-diagnostic case sharing library attributes, by: taking all the attributes described by the doctor diagnosis and treatment experience in the diagnosis case sharing library, namely the diagnosis case sharing library attribute as the column name of a matrix, taking the diagnosis case ID searched by the diagnosis case searching part and the reference attribute Id described by the doctor diagnosis and treatment experience designated by a user as row names, taking the reference attribute Id as a first row, and marking the value of a row and column of the first matrix as 1 when the diagnosis case ID or the reference attribute Id represented by a certain row of the first matrix has the diagnosis case sharing library attribute represented by a certain column of the first matrix, otherwise marking the value of the row and column of the first matrix as 0; a relationship set establishing step of establishing a relationship set, which is a relationship set of a diagnosis case sharing library attribute-a diagnosis case, according to the first matrix, wherein in the relationship set establishing step, the relationship set is established by the following method: the relation set of each diagnosis case sharing library attribute comprises all diagnosis case IDs or reference attribute Ids with column marks of 1 corresponding to the diagnosis case sharing library attribute in the first matrix; a second matrix establishing step of establishing a second matrix, which is a two-dimensional matrix of a diagnosis case-a diagnosis case, based on the relation set, by: taking the diagnostic case ID and the reference attribute ID as row names and column names, taking the reference attribute ID as a first row and a first column, taking the number of combinations of different diagnostic case IDs in the relation set of each diagnostic case sharing library attribute, or the number of combinations of the diagnostic case ID and the reference attribute ID as the value of the corresponding row and column, and marking the value of which the row and the column represent the same diagnostic case ID or both represent the reference attribute ID as 0 in the second matrix; a similarity calculation step of calculating the similarity between each diagnosis case ID and the reference attribute ID according to the second matrix; and a diagnostic case recommending step of sorting the diagnostic case IDs according to the similarity and recommending the diagnostic case IDs according to the sorting order.
ADVANTAGEOUS EFFECTS OF INVENTION
According to the device and the method, the searched diagnosis cases can be ordered according to the similarity of the attribute related to the doctor of each diagnosis case and the attribute related to the doctor designated by the user, and the diagnosis cases are recommended according to the order, so that the user can quickly obtain the expected diagnosis cases.
Drawings
Fig. 1 is a configuration diagram showing an apparatus according to an embodiment of the present invention.
Fig. 2 is a flowchart showing the retrieval and recommendation of a diagnosis case performed by the apparatus according to the embodiment of the present invention.
Detailed Description
Hereinafter, a specific embodiment of the present invention will be described with reference to the drawings.
Fig. 1 is a configuration diagram showing an apparatus according to an embodiment of the present invention, and fig. 2 is a flowchart showing a process of searching and recommending a diagnosis case by the apparatus according to the embodiment of the present invention.
As shown in fig. 1, the apparatus 1 of the present invention comprises: a keyword input unit 11, a diagnosis case search unit 12, a first matrix creation unit 13, a relationship set creation unit 14, a second matrix creation unit 15, a similarity calculation unit 16, and a diagnosis case recommendation unit 17.
The keyword input unit 11 is used to input keywords (step S1 in fig. 2), and a conventional input device such as a keyboard may be used.
The diagnosis case search unit 12 searches the diagnosis case sharing library using the keyword (step S2 in fig. 2). In the shared library of diagnostic cases, the description section of each diagnostic case with respect to the doctor's experience of diagnosis includes: the diagnosis case sharing library can be used by the person skilled in the art for searching, such as departments where doctors are located, names of doctors, levels of hospitals where doctors are located, and the like. In general, the description part of diagnosis cases related to doctor diagnosis experience is segmented to obtain multiple attributes of the diagnosis cases, and then the multiple attributes of all the diagnosis cases are combined, and repeated attributes are removed, so that the attribute of the diagnosis case sharing library is formed.
The first matrix creation unit 13 creates a first matrix that is a two-dimensional matrix of the diagnosis case-diagnosis case sharing library attribute (step S3 in fig. 2).
Specifically, all the attributes describing the diagnosis experience of the doctor in the diagnosis case sharing library, namely the diagnosis case sharing library attributes, are taken as column names of a matrix, and are marked as follows:
X 1 ，X 2 ，X 3 ...X i (i=n, N is a positive integer)
The diagnostic case ID and the reference attribute ID searched by the diagnostic case search unit are set as row names, and the reference attribute ID is set as a first row, and the row names are recorded as:
Id，ID 1 ，ID 2 ，ID 3 ...ID j (j=n, N is a positive integer)
Here, the reference attribute Id is an attribute describing the experience of doctor diagnosis and treatment, which is specified by the user as required.
When the diagnostic case ID or the reference attribute ID represented by a row of the first matrix has the diagnostic case-sharing library attribute represented by a column of the first matrix, the value of the column of the first matrix is marked as 1, otherwise marked as 0, so that a first matrix as follows is obtained:
the relationship set creation unit 14 creates a relationship set, which is a relationship set of the diagnosis case shared library attribute and the diagnosis case, based on the first matrix (step S4 in fig. 2).
Specifically, the relation set of each diagnosis case sharing library attribute includes all diagnosis case IDs or the reference attribute IDs with a column mark of 1 corresponding to the diagnosis case sharing library attribute in the first matrix, that is:
X 1 ：{Id，ID 1 ，ID 2 …}
X 2 ：{Id…ID j }
X 3 ：{ID 1 …ID j }
X 4 ：{ID 1 ，ID 2 …}
…
the second matrix creation unit 15 creates a second matrix, which is a two-dimensional matrix of diagnosis cases-diagnosis cases, from the relation set (step S5 in fig. 2).
Specifically, the diagnostic case ID and the reference attribute ID are set as row names and column names, and the reference attribute ID is set as first row and first column, the number of combinations of the diagnostic case IDs, or the number of combinations of the diagnostic case ID and the reference attribute ID, which are different from each other in the relation set of each diagnostic case sharing library attribute, is set as the value of the corresponding row and column, and the values of the same diagnostic case ID or the reference attribute ID, which are both represented by rows and columns, are marked as 0 in the second matrix, so that a second matrix as shown below is obtained:
here, for example, the ID 2 Line ID 3 Column value and the first ID 3 Line ID 2 The values of the columns are all represented in the set of relationships, ID 2 With ID 3 And therefore the same value.
A similarity calculation unit 16 that calculates a similarity between each diagnosis case ID and the reference attribute ID based on the second matrix and the relationship set (step S6 in fig. 2).
In the present embodiment, the diagnostic case ID can be calculated using the following equation one j Similarity to the reference attribute Id
j is a positive integer, j=1, 2,3 …, and let u=id for simplicity of the formula j V=id, at the same time, Z uv For the value of the ith row and the ith column in the second matrix, P is the total number of the ith row, P 'is the total number of the ith column, K and K' are positive integers, Z uk Z is the value of the kth column of the ith row in the second matrix K′v Is the value of the kth' row and the kth column in the second matrix.
The diagnosis case recommending section 17 sorts the diagnosis case IDs according to the similarity, and recommends the diagnosis case IDs in the order of the sorting (step S7 in fig. 2).
In this embodiment, the diagnosis case recommending section sorts the diagnosis case IDs in order of the degree of similarity from large to small, and recommends the diagnosis case IDs in order of the degree of similarity.
When the diagnostic case IDs having the same similarity exist, a weight score of the diagnostic case IDs having the same similarity is calculated according to the following formula two:
score＝(P-1) n (N+1) (formula one)
P is the click rate of the diagnosis case ID; n is the number of columns in the first matrix in which the values of the rows representing the reference attribute Id and the diagnostic case Id in the same column are the same, N is a coefficient,
and further sorting the diagnosis case IDs with the same similarity according to the calculated weight score from large to small, and recommending the diagnosis case IDs by combining the sorting according to the similarity with the sorting according to the weight score.
Here, n is used to adjust the click rate of the diagnostic case ID and the weight of the attribute, and a specific number may be set as needed. In this embodiment, n is 1/2.
Example 1
Assume that: when the user inputs the keyword 'hypertension' to search, the search result number is not 0, and the attribute of the diagnosis case sharing library is as follows:
{ cardiac surgery, advanced physician, three-phase hospital, coastal area of east China, 15 years from doctor }.
And inputting a keyword 'hypertension' into the diagnosis case sharing library to obtain a retrieval result set. The search result set includes four diagnosis cases with ID in turn 1 、ID 2 、ID 3 、ID 4 。
The diagnostic cases have the following properties in turn about the description of the doctor's experience in diagnosis:
ID 1 : { cardiac surgery, advanced physician, coastal area of east China, 15 years from doctor }
ID 2 : { cardiac surgery, advanced physician, three-dimensional hospital })
ID 3 : { advanced physicians, three-dimensional hospitals, coastal areas of China east China, 15 years from doctor }
ID 4 : { cardiac surgery, three-dimensional hospital, 15 years from doctor }
And assume that the user-specified reference attribute Id is { advanced physician, three hospitals, more than 15 years from doctor }.
Establishing a two-dimensional matrix of the attributes of the diagnosis case-diagnosis case sharing library, and taking all the attributes of the diagnosis case sharing library as the column names of the matrix, namely the column names are as follows: cardiac surgery, advanced physicians, trimethyl hospitals, coastal areas of eastern China, and more than 15 years from medical.
Taking the reference attribute Id as a first row of a matrix, taking the search result diagnosis case ID as other rows, wherein the rows of the matrix are Id and ID 1 、ID 2 、ID 3 、ID 4 。
When the diagnosis case or the reference attribute has the attribute listed in the column, the value of the corresponding column is marked as 1, otherwise, the value is marked as 0, and a two-dimensional matrix of the diagnosis case-diagnosis case sharing library attribute is obtained as follows:
subsequently, a relation set of the attribute of the diagnosis case shared library and the diagnosis case is established, namely
Cardiac surgery: { ID 1 ，ID 2 ，ID 4 }
Advanced physicians: { Id, ID 1 ，ID 2 ，ID 3 }
Three hospitals: { Id, ID 2 ，ID 3 ，ID 4 }
Coastal areas of eastern China: { ID1, ID3}
From medical 15 years or more: { Id, ID 1 ，ID 3 ，ID 4 }。
Subsequently, a two-dimensional matrix of diagnostic cases is established, i.e
Then, using equation one, the similarity of each diagnostic case ID to the reference attribute ID, that is,
Comparing the calculated similarity, it can be seen that:
C(ID 3 ，Id)＞C(ID 2 ，Id)＝C(ID 4 ，Id)＞C(ID 1 ，Id)
then according to the ID 3 、ID 2 、ID 4 、ID 1 Or ID 3 、ID 4 、ID 2 、ID 1 Is recommended for diagnosing cases in sequence, and ID 3 Is the most recommended diagnostic case.
Example two
Assume that: the user inputs a keyword 'diabetes' to search, the number of search results is not 0, and the attribute of the diagnosis case sharing library is as follows:
{ endocrinology, leading physician, three-phase Hospital, coastal area of east China }
And inputting a keyword diabetes into the diagnosis case sharing library to obtain a retrieval result. The search result includes three diagnosis cases, which are in turn ID 1 ，ID 2 ，ID 3 。
The attributes described about doctor's experience in three diagnostic cases are in turn:
ID 1 : { endocrinology, trimethyl Hospital }
ID 2 : { coastal area of China east }
ID 3 : { endocrinology department, leading and conception physician }
And assume that the user-specified reference attribute Id is { endocrinology, dominant physician, three-medical, eastern coastal region }.
Establishing a two-dimensional matrix of the attributes of the diagnosis case-diagnosis case sharing library, and taking all the attributes of the diagnosis case sharing library as the column names of the matrix, namely the column names are as follows: endocrinology, major physicians, trimethyl hospitals, coastal areas of eastern China.
Taking the reference attribute Id as a first row of a matrix, taking the search result diagnosis case ID as other rows, wherein the row of the matrix is Id, and the ID 1 ，ID 2 ，ID 3 。
When the diagnostic case ID or the reference attribute ID has the listed attributes in the column, the value of the corresponding column is marked as 1, otherwise marked as 0, and a two-dimensional matrix of the diagnostic case-diagnostic case shared library attributes is obtained as follows:
subsequently, a relation set of the attribute of the diagnosis case shared library and the diagnosis case is established, namely
Endocrinology department: { Id, ID 1 ，ID 3 }
The dominant physician: { Id, ID 3 }
Three hospitals: { Id, ID 1 }
Coastal areas of eastern China: { Id, ID 2 }。
Subsequently, a two-dimensional matrix of diagnostic cases is established, i.e
Then, using equation one, the similarity of each diagnostic case ID to the reference attribute ID, that is,
Comparing the above similarity, it can be seen that:
C(ID 1 ，Id)＝C(ID 3 ，Id)＞C(ID 2 ，Id)
here, it may be according to the ID 1 、ID 3 、ID 2 Or ID 3 、ID 1 、ID 2 Although the order of the diagnosis cases is recommended, the weight value of each of the diagnosis case ID1 and the diagnosis case ID3 may be calculated based on the formula II, and the diagnosis case ID may be calculated based on the weight value 1 And diagnosis of case ID 3 Further sequencing is performed.
In the present embodiment, it is assumed that case ID is diagnosed 1 Has a click rate P of 500, and diagnoses case ID 3 The click rate P of (2) is 901. Meanwhile, the two-dimensional matrix of the diagnosis case-diagnosis case sharing library attribute can be used for knowing the diagnosis case ID 1 The number N of the same column of the row with the same value as the row with the reference attribute Id is 2, and the diagnosis case ID 3 The number N of columns in which the value of the row in which the reference attribute Id is located is the same as that of the row in which the reference attribute Id is located is 2.
Thus, diagnosis of case ID 1 The weight score1 of (2) is:
diagnostic case ID 3 The weight value score3 of (2) is:
based on the size of the weight score, for diagnostic case ID with the same similarity 1 And ID 3 Reordered, i.e. recommended order is ID 3 、ID 1 。
In combination with the above-mentioned similarity-based ranking, the ranking is then based on ID 3 、ID 1 、ID 2 Is recommended for diagnosing cases in sequence, and ID 3 Is the most recommended diagnostic case.
Symbol description
The device comprises a device 1, a keyword input part 11, a diagnosis case searching part 12, a first matrix establishing part 13, a relation set establishing part 14, a second matrix establishing part 15, a similarity calculating part 16 and a diagnosis case recommending part 17.
Claims (8)
1. An apparatus for retrieving a diagnostic case based on a doctor related attribute, comprising:
a keyword input unit for inputting a keyword;
a diagnosis case search unit that searches a diagnosis case sharing library using the keyword;
a first matrix creation section that creates a first matrix that is a two-dimensional matrix of diagnostic case-diagnostic case sharing library attributes, the first matrix creation section creating the first matrix by: taking all the attributes described by the doctor diagnosis and treatment experience in the diagnosis case sharing library, namely the diagnosis case sharing library attribute as the column name of a matrix, taking the diagnosis case ID searched by the diagnosis case searching part and the reference attribute Id described by the doctor diagnosis and treatment experience designated by a user as row names, taking the reference attribute Id as a first row, and marking the value of a row and column of the first matrix as 1 when the diagnosis case ID or the reference attribute Id represented by a certain row of the first matrix has the diagnosis case sharing library attribute represented by a certain column of the first matrix, otherwise marking the value of the row and column of the first matrix as 0;
a relationship set creation unit that creates a relationship set, which is a relationship set of a diagnosis case shared library attribute-a diagnosis case, from the first matrix, the relationship set creation unit creating the relationship set by: the relation set of each diagnosis case sharing library attribute comprises all diagnosis case IDs or reference attribute Ids with column marks of 1 corresponding to the diagnosis case sharing library attribute in the first matrix;
a second matrix creation unit that creates a second matrix, which is a two-dimensional matrix of diagnosis cases-diagnosis cases, from the relation set, the second matrix creation unit creating the second matrix by: taking the diagnostic case ID and the reference attribute ID as row names and column names, taking the reference attribute ID as a first row and a first column, taking the number of combinations of different diagnostic case IDs in the relation set of each diagnostic case sharing library attribute, or the number of combinations of the diagnostic case ID and the reference attribute ID as the value of the corresponding row and column, and marking the value of which the row and the column represent the same diagnostic case ID or both represent the reference attribute ID as 0 in the second matrix;
a similarity calculation unit that calculates a similarity between each diagnosis case ID and the reference attribute ID based on the second matrix; and
and a diagnosis case recommending unit that ranks the diagnosis case IDs according to the similarity, and recommends the diagnosis case IDs in the ranking order.
2. The apparatus of claim 1, wherein,
the similarity calculation section calculates the similarity by:
diagnostic case ID j Similarity to the reference attribute Id
Is that
j is a positive integer, j=1, 2,3 …, u=id j ，v＝Id，Z uv For the value of the ith row and the ith column in the second matrix, P is the total number of the ith row, P 'is the total number of the ith column, K and K' are positive integers, Z uk Z is the value of the kth column of the ith row in the second matrix K′v Is the value of the kth' row and the kth column in the second matrix.
3. The apparatus of claim 1 or 2, wherein,
the diagnosis case recommending section sorts the diagnosis case IDs from large to small according to the similarity, and recommends the diagnosis case IDs in the order of arrangement,
when the diagnostic case IDs having the same degree of similarity exist, a weight score of the diagnostic case IDs having the same degree of similarity is calculated in the following manner:
score＝(P-1) n (N+1)
p is the click rate of the diagnosis case ID; n is the number of columns in the first matrix in which the values of the rows representing the reference attribute Id and the diagnostic case Id in the same column are the same, N is a coefficient,
and further sorting the diagnosis case IDs with the same similarity according to the calculated weight score from large to small, and recommending the diagnosis case IDs by combining the sorting according to the similarity with the sorting according to the weight score.
4. The apparatus of claim 3, wherein the device comprises a plurality of sensors,
and n is 1/2.
5. A method for retrieving a diagnostic case based on a doctor related attribute, comprising:
a keyword input step for inputting keywords;
a step of searching for the diagnosis cases, which uses the keywords to search a shared library of diagnosis cases;
a first matrix establishing step of establishing a first matrix, which is a two-dimensional matrix of diagnostic case-diagnostic case sharing library attributes, by: taking all the attributes described by the doctor diagnosis and treatment experience in the diagnosis case sharing library, namely the diagnosis case sharing library attribute as the column name of a matrix, taking the diagnosis case ID searched in the diagnosis case searching step and the reference attribute Id described by the doctor diagnosis and treatment experience designated by a user as row names, taking the reference attribute Id as a first row, and marking the value of a row and column of the first matrix as 1 when the diagnosis case ID or the reference attribute Id represented by a certain row of the first matrix has the diagnosis case sharing library attribute represented by a certain column of the first matrix, otherwise marking the value of the row and column of the first matrix as 0;
a relationship set establishing step of establishing a relationship set, which is a relationship set of a diagnosis case sharing library attribute-a diagnosis case, according to the first matrix, wherein in the relationship set establishing step, the relationship set is established by the following method: the relation set of each diagnosis case sharing library attribute comprises all diagnosis case IDs or reference attribute Ids with column marks of 1 corresponding to the diagnosis case sharing library attribute in the first matrix;
a second matrix establishing step of establishing a second matrix, which is a two-dimensional matrix of a diagnosis case-a diagnosis case, based on the relation set, by: taking the diagnostic case ID and the reference attribute ID as row names and column names, taking the reference attribute ID as a first row and a first column, taking the number of combinations of different diagnostic case IDs in the relation set of each diagnostic case sharing library attribute, or the number of combinations of the diagnostic case ID and the reference attribute ID as the value of the corresponding row and column, and marking the value of which the row and the column represent the same diagnostic case ID or both represent the reference attribute ID as 0 in the second matrix;
a similarity calculation step of calculating the similarity between each diagnosis case ID and the reference attribute ID according to the second matrix; and
and a diagnosis case recommending step, namely sequencing the diagnosis case IDs according to the similarity, and recommending the diagnosis case IDs according to the sequence.
6. The method of claim 5, wherein,
in the similarity calculation step, the similarity is calculated by:
diagnostic case ID j Similarity to the reference attribute Id
Is that
j is a positive integer, j=1, 2,3 …, u=id j ，v＝Id，Z uv For the value of the ith row and the ith column in the second matrix, P is the total number of the ith row, P 'is the total number of the ith column, K and K' are positive integers, Z uk Z is the value of the kth column of the ith row in the second matrix K′v Is the value of the kth' row and the kth column in the second matrix.
7. The method of claim 5 or 6, wherein,
in the diagnostic case recommending step, the diagnostic case IDs are sorted according to the degree of similarity from large to small, and the diagnostic case IDs are recommended according to the sorting order,
when the diagnostic case IDs having the same similarity exist, a weight score of the same diagnostic case ID is calculated in the following manner:
score＝(P-1) n (N+1)
p is the click rate of the diagnosis case ID; n is the number of columns in the first matrix in which the values of the rows representing the reference attribute Id and the diagnostic case Id in the same column are the same, N is a coefficient,
and further sorting the diagnosis case IDs with the same similarity according to the calculated weight score from large to small, and recommending the diagnosis case IDs by combining the sorting according to the similarity with the sorting according to the weight score.
8. The method of claim 7, wherein,
and n is 1/2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710419900.3A CN109002442B (en) | 2017-06-06 | 2017-06-06 | Device and method for searching diagnosis cases based on doctor related attributes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710419900.3A CN109002442B (en) | 2017-06-06 | 2017-06-06 | Device and method for searching diagnosis cases based on doctor related attributes |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109002442A CN109002442A (en) | 2018-12-14 |
| CN109002442B true CN109002442B (en) | 2023-04-25 |
Family
ID=64572953
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710419900.3A Active CN109002442B (en) | 2017-06-06 | 2017-06-06 | Device and method for searching diagnosis cases based on doctor related attributes |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109002442B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111599482A (en) * | 2020-05-14 | 2020-08-28 | 青岛海信医疗设备股份有限公司 | Electronic case recommendation method and server |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5761496A (en) * | 1993-12-14 | 1998-06-02 | Kabushiki Kaisha Toshiba | Similar information retrieval system and its method |
| WO2012104949A1 (en) * | 2011-01-31 | 2012-08-09 | パナソニック株式会社 | Disease case study search device and disease case study search method |
| CN103064941A (en) * | 2012-12-25 | 2013-04-24 | 深圳先进技术研究院 | Image retrieval method and device |
| CN104036109A (en) * | 2014-03-14 | 2014-09-10 | 上海大图医疗科技有限公司 | Image based system and method for case retrieving, sketching and treatment planning |
| CN104881463A (en) * | 2015-05-22 | 2015-09-02 | 清华大学深圳研究生院 | Reference medical record search method and device based on structural medical record database |
| CN105184103A (en) * | 2015-10-15 | 2015-12-23 | 清华大学深圳研究生院 | Virtual medical expert based on medical record database |
| CN105786983A (en) * | 2016-02-15 | 2016-07-20 | 云南电网有限责任公司 | Employee individualized-learning recommendation method based on learning map and collaborative filtering |
| CN106415532A (en) * | 2014-06-04 | 2017-02-15 | 株式会社日立制作所 | Medical care data search system |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1645367A (en) * | 2004-12-27 | 2005-07-27 | 上海医元网数码科技发展有限公司 | System for medicinal seeking |
| WO2012111288A1 (en) * | 2011-02-14 | 2012-08-23 | パナソニック株式会社 | Similar case retrieval device and similar case retrieval method |
| CN103605760A (en) * | 2013-11-25 | 2014-02-26 | 方正国际软件有限公司 | Medical record query method and medical record query system |
| CN106776606A (en) * | 2015-11-20 | 2017-05-31 | 株式会社日立制作所 | Retrieval device and search method based on electronic health record database |
| CN105893597B (en) * | 2016-04-20 | 2022-05-31 | 上海家好科技有限公司 | Similar medical record retrieval method and system |
-
2017
- 2017-06-06 CN CN201710419900.3A patent/CN109002442B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5761496A (en) * | 1993-12-14 | 1998-06-02 | Kabushiki Kaisha Toshiba | Similar information retrieval system and its method |
| WO2012104949A1 (en) * | 2011-01-31 | 2012-08-09 | パナソニック株式会社 | Disease case study search device and disease case study search method |
| CN103064941A (en) * | 2012-12-25 | 2013-04-24 | 深圳先进技术研究院 | Image retrieval method and device |
| CN104036109A (en) * | 2014-03-14 | 2014-09-10 | 上海大图医疗科技有限公司 | Image based system and method for case retrieving, sketching and treatment planning |
| CN106415532A (en) * | 2014-06-04 | 2017-02-15 | 株式会社日立制作所 | Medical care data search system |
| CN104881463A (en) * | 2015-05-22 | 2015-09-02 | 清华大学深圳研究生院 | Reference medical record search method and device based on structural medical record database |
| CN105184103A (en) * | 2015-10-15 | 2015-12-23 | 清华大学深圳研究生院 | Virtual medical expert based on medical record database |
| CN105786983A (en) * | 2016-02-15 | 2016-07-20 | 云南电网有限责任公司 | Employee individualized-learning recommendation method based on learning map and collaborative filtering |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109002442A (en) | 2018-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8046363B2 (en) | System and method for clustering documents | |
| US8238663B2 (en) | Similar image search apparatus and computer readable medium | |
| US8046368B2 (en) | Document retrieval system and document retrieval method | |
| JP5173700B2 (en) | Data search apparatus, control method therefor, and data search system | |
| RU2541198C2 (en) | Systems for maintaining clinical decision making with external context | |
| CN109074858B (en) | Hospital matching of de-identified healthcare databases without distinct quasi-identifiers | |
| US7933452B2 (en) | System and methods of image retrieval | |
| US20090287663A1 (en) | Disease name input support program, method and apparatus | |
| JP5939141B2 (en) | Program and medical record retrieval device | |
| US20070082353A1 (en) | Genetic marker selection program for genetic diagnosis, apparatus and system for executing the same, and genetic diagnosis system | |
| KR19980086683A (en) | Recording medium recording similar search device and similar search program | |
| JP2019512795A (en) | Relevance feedback to improve the performance of classification models that classify patients with similar profiles together | |
| CN110299209B (en) | Similar medical record searching method, device and equipment and readable storage medium | |
| CN101441658A (en) | Search method and system facing to radiation image in PACS database based on content | |
| US20220020457A1 (en) | Medical data evaluation utilization system and medical data evaluation utilization method | |
| CN111415760B (en) | Doctor recommendation method, doctor recommendation system, computer equipment and storage medium | |
| de Herrera et al. | Comparing fusion techniques for the ImageCLEF 2013 medical case retrieval task | |
| US20040044547A1 (en) | Database for retrieving medical studies | |
| CN109002442B (en) | Device and method for searching diagnosis cases based on doctor related attributes | |
| CN111223533B (en) | Medical data retrieval method and system | |
| US20040167800A1 (en) | Methods and systems for searching, displaying, and managing medical teaching cases in a medical teaching case database | |
| CN107862043A (en) | Check the search method and device of information | |
| CN111984694A (en) | Orthopedics search engine system | |
| JP2004348271A (en) | Clinical trial data outputting device, clinical trial data outputting method, and clinical trial data outputting program | |
| CN110476215A (en) | Signature-hash for multisequencing file |
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 |