PACS (Picture Archiving and Communication System) voice statistical method and system based on large model, and storage medium

Abstract

The invention aims to disclose a PACS (Picture Archiving and Communication System) voice statistical method and system based on a large model and a storage medium, and belongs to the technical field of medical image processing. The method comprises the following steps: S1, collecting a plurality of first natural language characters conforming to a PACS statistical request; s2, compiling a plurality of first SQL statements corresponding to the first natural language characters; s3, performing word segmentation on the first natural language characters to form a plurality of second natural language characters; s4, the large language model generates a plurality of new second SQL statements on the basis of the second natural language characters; s5, establishing a mapping relation between the second natural language characters and the second SQL statements through a large language model; s6, the user says a section of third natural language text through voice, and the voice conversion module converts the third natural language text into the second natural language text; and S7, the PACS system generates the statistical content based on the second SQL statement. The method has the beneficial effect that the statistical content is directly generated according to the voice of the user.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to a large-model-based PACS system speech statistics method, system and storage medium. Background Art

[0002] The medical imaging system (PACS) is a system for hospital image management. Currently, statistical department reports need to use a query system to obtain and generate the desired information from massive amounts of data information. However, the query requirements of different hospitals and departments vary greatly, and fixed interface development is difficult to meet the diverse requirements of various hospitals and departments.

[0003] In view of this, it is necessary to develop a speech statistics method, system and storage medium for PACS system based on large model. Summary of the invention

[0004] The purpose of the present invention is to disclose a PACS system speech statistics method, system and storage medium based on a large model.

[0005] The first object of the present invention is to provide a PACS system speech statistics method based on a large model.

[0006] The second inventive object of the present invention is to provide a PACS system speech statistics system based on a large model.

[0007] A third object of the present invention is to provide a computer-readable storage medium.

[0008] To achieve the above first invention objective, the present invention provides a PACS system speech statistics method based on a large model, comprising the following steps:

[0009] S1: Collecting a number of first natural language characters that meet the statistical request of the PACS system;

[0010] S2: writing a plurality of first SQL statements corresponding to the first natural language text;

[0011] S3: Segmenting the first natural language text, extracting time-related words, department name words, and purpose-related words, and cross-editing the time-related words, department name words, and purpose-related words through a large language model to form a plurality of second natural language texts;

[0012] S4: the large language model generates a plurality of new second SQL statements based on the second natural language text;

[0013] S5: Establishing a mapping relationship between the second natural language text and the second SQL statement through a large language model;

[0014] S6: The user speaks a third natural language text by voice, and the PACS system converts the third natural language text into the second natural language text by a voice conversion module;

[0015] S7: The PACS system automatically calls the second SQL statement according to the second natural language text, and generates statistical content based on the second SQL statement.

[0016] Preferably, in step S6, the speech conversion module extracts time-related vocabulary, department name vocabulary and purpose-related vocabulary from the third natural language text and generates the second natural language text.

[0017] Preferably, the time-related vocabulary is one of this week, this month, last month, first three quarters, this year, and last year;

[0018] The department name is one of radiology, pediatrics, and orthopedics;

[0019] The purpose-type vocabulary is one of the number of examinations, the number of patients received, and the total amount of charges.

[0020] Preferably, after completing step S7, the time-related words, department name words and purpose-related words in the second natural language text are cached and displayed on the screen of the PACS system;

[0021] The PACS system automatically generates statistical content based on the time-related vocabulary, department name vocabulary and purpose-related vocabulary displayed on the screen manually selected by the user.

[0022] Based on the same inventive principle, in order to achieve the above second inventive object, the present invention provides a PACS system speech statistics system based on a large model, including a large language model training module, a speech conversion module and a screen;

[0023] The large language model training module collects a number of first natural language characters that meet the statistical request of the PACS system and a first SQL statement corresponding to the first natural language characters;

[0024] The large language model training module performs word segmentation on the first natural language text, and extracts time-related words, department name words, and purpose-related words, and cross-edits the time-related words, department name words, and purpose-related words through the large language model training module to form a plurality of second natural language texts. The large language model training module generates a second SQL statement based on the second natural language text and establishes a mapping relationship between the two.

[0025] The speech conversion module is used to convert the third natural language text spoken by the user into the second natural language text, and the large language model training module calls the corresponding second SQL statement based on the second natural language text;

[0026] The PACS system generates statistical content based on the second SQL statement and displays it on the screen.

[0027] Preferably, the speech conversion module extracts time-related vocabulary, department name vocabulary and purpose-related vocabulary from the third natural language text and generates the second natural language text.

[0028] Preferably, the time-related vocabulary is one of this week, this month, last month, first three quarters, this year, and last year;

[0029] The department name is one of radiology, pediatrics, and orthopedics;

[0030] The purpose-type vocabulary is one of the number of examinations, the number of patients received, and the total amount of charges.

[0031] Preferably, the PACS system caches the time-related words, department name words and purpose-related words in the second natural language text and displays them on the screen;

[0032] The PACS system automatically generates statistical content based on the time-related vocabulary, department name vocabulary and purpose-related vocabulary displayed on the screen manually selected by the user.

[0033] Preferably, the screen is a touch screen.

[0034] Based on the same inventive principle, in order to achieve the above-mentioned third invention purpose, the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the large model-based PACS system speech statistics method as described in the first invention is implemented.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] Write corresponding first SQL statements for a number of first natural language characters; segment the first natural language characters, extract time-related words, department name words and purpose-related words and perform cross-editing to form a number of second natural language characters; the large language model generates a number of new second SQL statements based on the second natural language characters and establishes a mapping relationship between the two; the user only needs to speak the statistical requirements (third natural language characters), the voice conversion module converts the third natural language characters into the second natural language characters and inputs them into the PACS system, and the PACS system calls the second SQL statement that has a mapping relationship with the second natural language, thereby realizing the direct generation of statistical content based on the user's voice, making the PACS system more adaptable and more convenient for users to call statistical content. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a flow chart of the speech statistics method of the PACS system based on the large model of the present invention;

[0038] Figure 2 It is a block diagram of the speech statistics system of the PACS system based on the large model of the present invention;

[0039] Figure 3 It is a schematic diagram of the computer medium module of the present invention. DETAILED DESCRIPTION

[0040] The present invention is described in detail below in conjunction with the various embodiments shown in the accompanying drawings, but it should be noted that these embodiments are not limitations of the present invention, and any equivalent transformations or substitutions in functions, methods, or structures made by ordinary technicians in the field based on these embodiments are all within the scope of protection of the present invention.

[0041] The specific implementation process of the present invention is described below through multiple embodiments.

[0042] Embodiment 1:

[0043] Ginseng Figure 1 As shown, this embodiment discloses a specific implementation of a PACS system speech statistics method based on a large model (hereinafter referred to as "method").

[0044] Ginseng Figure 1 As shown, in this embodiment, the PACS system speech statistics method based on the large model includes the following steps:

[0045] S1: Collect a number of first natural language characters that meet the statistical request of the PACS system; specifically, compile a number of first natural language characters according to possible statistical requirements of different hospitals and departments.

[0046] S2: Write a plurality of first SQL statements corresponding to the first natural language characters; specifically, write a corresponding first SQL statement for each first natural language character, and the first SQL statement initiates a statistical request to the PACS system.

[0047] S3: Segment the first natural language text, and extract time-related vocabulary, department name vocabulary and purpose-related vocabulary, and cross-edit the time-related vocabulary, department name vocabulary and purpose-related vocabulary through a large language model to form a number of second natural language texts; specifically, the time-related vocabulary represents the statistical period, and the time-related vocabulary is one of this week, this month, last month, the first three quarters, this year, and last year, and the time-related vocabulary can also be other unlisted statistical periods; the department name vocabulary is one of radiology, pediatrics, and orthopedics, and the department name vocabulary represents the statistical user, and the department name can also be other unlisted statistical users; the purpose-related vocabulary represents the statistical target, and the purpose-related vocabulary is one of the number of examinations, the number of patients received, and the total amount of charges, and the purpose-related vocabulary can also be other unlisted statistical targets; the first natural language text is as comprehensive as possible and covers different statistical targets of different statistical periods of different departments as much as possible. After the large language model cross-edits the time-related vocabulary, department name vocabulary and purpose-related vocabulary, a richer number of second natural language texts are formed, which provides a sufficient foundation for subsequent large language model training.

[0048] S4: The large language model generates a plurality of new second SQL statements based on the second natural language text; specifically, the large language model (LLM) generates a plurality of new second SQL statements based on the cross-edited second natural language text, and the plurality of second SQL statements are stored in the PACS system.

[0049] S5: Establish a mapping relationship between the second natural language text and the second SQL statement through a large language model; specifically, each second natural language text corresponds to a second SQL statement, so that the corresponding second SQL statement can be directly called when a statistical demand is subsequently issued.

[0050] S6: The user speaks a paragraph of the third natural language through voice, and the PACS system converts the third natural language into the second natural language through the voice conversion module; specifically, when the statistical user has a statistical demand, he speaks a paragraph of the third natural language through voice, and the third natural language must clearly include time-related vocabulary, department name vocabulary and purpose-type vocabulary, such as counting the number of patients received by the pediatric department last week. The voice conversion module converts the third natural language into the second natural language. In step S6, the voice conversion module extracts the time-related vocabulary, department name vocabulary and purpose-type vocabulary from the third natural language and generates the second natural language. The voice conversion module will filter out vocabulary that is not related to the time-related vocabulary, department name vocabulary and purpose-type vocabulary, such as modal particles: ah, ba.

[0051] S7: The PACS system automatically calls the second SQL statement according to the second natural language text, and generates statistical content based on the second SQL statement. Specifically, a statistical request is sent to the PACS system through the second SQL statement corresponding to the second natural language text, so that the statistical requirements stated by the user are displayed in the form of a statistical report; after completing step S7, the time-related words, department name words and purpose-related words in the second natural language text are cached and displayed on the screen of the PACS system. The PACS system automatically generates statistical content based on the time-related words, department name words and purpose-related words manually selected by the user to be displayed on the screen. The statistical content is displayed in the form of bar charts, line charts, trend charts and other icons. After the user frequently uses the PACS system for statistics, the screen can cache commonly used time-related words, department name words and purpose-related words, which can further improve the statistical efficiency of the user and further improve the versatility of each hospital and department.

[0052] Embodiment 2:

[0053] Ginseng Figure 2 As shown, this embodiment discloses a specific implementation of a PACS system speech statistics system (hereinafter referred to as "system") based on a large model.

[0054] Ginseng Figure 2 As shown, in this embodiment, the PACS system speech statistics system based on the big model includes a big language model training module, a speech conversion module and a screen; the big language model training module collects a number of first natural language characters that meet the statistical request of the PACS system and a first SQL statement corresponding to the first natural language characters, and the first natural language characters fully reflect the possible statistical needs of different hospitals and different departments.

[0055] The large language model training module performs word segmentation on the first natural language text, and extracts time-related vocabulary, department name vocabulary and purpose-related vocabulary. The large language model training module cross-edits the time-related vocabulary, department name vocabulary and purpose-related vocabulary to form a number of second natural language texts. The large language model training module generates a second SQL statement based on the second natural language text and establishes a mapping relationship between the two. The time-related vocabulary represents a statistical period. The time-related vocabulary is one of this week, this month, last month, the first three quarters, this year and last year. The time-related vocabulary can also be other unlisted statistical periods. The department name The vocabulary is one of radiology, pediatrics, and orthopedics. The department name vocabulary represents the statistical user, and the department name can also be other unlisted statistical users; the purpose-type vocabulary represents the statistical target, and the purpose-type vocabulary is one of the number of examinations, the number of patients received, and the total amount of charges. The purpose-type vocabulary can also be other unlisted statistical targets; the first natural language text is as comprehensive as possible and covers different statistical targets of different statistical periods of different departments as much as possible. After the large language model cross-edits the time vocabulary, department name vocabulary and purpose-type vocabulary, it forms a richer number of second natural language texts, which provides a sufficient foundation for subsequent large language model training.

[0056] The speech conversion module is used to convert the third natural language text spoken by the user into the second natural language text, and the large language model training module calls the corresponding second SQL statement based on the second natural language text; when the statistical user has statistical needs, a third natural language text is spoken through voice, and the third natural language text must clearly include time-related vocabulary, department name vocabulary and purpose-type vocabulary, such as counting the number of patients received by the pediatric department last week, and the speech conversion module converts the third natural language text into the second natural language text; the speech conversion module extracts the time-related vocabulary, department name vocabulary and purpose-type vocabulary in the third natural language text and generates the second natural language text, and the speech conversion module will filter out vocabulary that is not related to the time-related vocabulary, department name vocabulary and purpose-type vocabulary, such as modal particles: ah, ba.

[0057] The PACS system generates statistical content based on the second SQL statement and displays it on the screen, which is preferably a touch screen; a statistical request is issued to the PACS system through a second SQL statement corresponding to the second natural language text, so that the statistical requirements stated by the user are presented in the form of a statistical report, and the statistical report is displayed in the form of icons such as a bar chart, a line chart, and a trend chart; after each statistical completion, the time-related vocabulary, department name vocabulary, and purpose-related vocabulary in the second natural language text are cached and displayed on the screen of the PACS system, and the PACS system automatically generates statistical content based on the time-related vocabulary, department name vocabulary, and purpose-related vocabulary displayed on the screen manually selected by the user. After the user frequently uses the PACS system for statistics, the screen can cache commonly used time-related vocabulary, department name vocabulary, and purpose-related vocabulary, which can further improve the user's statistical efficiency and further improve the versatility of its various departments in various hospitals.

[0058] Embodiment three:

[0059] Ginseng Figure 3 As shown, a computer-readable storage medium stores a computer program, and when the program is executed by a processor, the large model-based PACS system speech statistics method as described in Example 1 is implemented.

[0060] The computer-readable storage medium disclosed in this embodiment has the same technical solutions as those in Embodiment 1. Please refer to Embodiment 1 and will not be described in detail here.

[0061] The various illustrative logic blocks or units described in the embodiments of the present invention can be implemented or operated by a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field programmable gate array or other programmable logic device, a discrete gate or transistor logic, a discrete hardware component, or any combination of the above. The general-purpose processor can be a microprocessor, and optionally, the general-purpose processor can also be any conventional processor, controller, microcontroller or state machine. The processor can also be implemented by a combination of computing devices, such as a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a digital signal processor core, or any other similar configuration.

[0062] The series of detailed descriptions listed above are only specific descriptions of feasible implementation methods of the present invention. They are not intended to limit the scope of protection of the present invention. Any equivalent implementation methods or changes that do not deviate from the technical spirit of the present invention should be included in the scope of protection of the present invention.

[0063] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0064] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A speech statistics method for PACS system based on a large model, characterized in that: The following steps are involved: S1: Collecting a number of first natural language characters that meet the statistical request of the PACS system; S2: writing a plurality of first SQL statements corresponding to the first natural language text; S3: Segmenting the first natural language text, extracting time-related words, department name words, and purpose-related words, and cross-editing the time-related words, department name words, and purpose-related words through a large language model to form a plurality of second natural language texts; S4: the large language model generates a plurality of new second SQL statements based on the second natural language text; S5: Establishing a mapping relationship between the second natural language text and the second SQL statement through a large language model; S6: The user speaks a third natural language text by voice, and the PACS system converts the third natural language text into the second natural language text by a voice conversion module; S7: The PACS system automatically calls the second SQL statement according to the second natural language text, and generates statistical content based on the second SQL statement.

2. The method for PACS system speech statistics based on a large model as claimed in claim 1, characterized in that: In step S6, the speech conversion module extracts time-related words, department name words and purpose-related words from the third natural language text and generates the second natural language text.

3. The method for PACS system speech statistics based on a large model as claimed in claim 1 or 2, characterized in that: The time-related vocabulary is one of this week, this month, last month, first three quarters, this year, and last year; The department name is one of radiology, pediatrics, and orthopedics; The purpose-type vocabulary is one of the number of examinations, the number of patients received, and the total amount of charges.

4. The method for PACS system speech statistics based on a large model as claimed in claim 3, characterized in that: After completing step S7, the time-related words, department name words and purpose-related words in the second natural language text are cached and displayed on the screen of the PACS system; The PACS system automatically generates statistical content based on the time-related vocabulary, department name vocabulary and purpose-related vocabulary displayed on the screen manually selected by the user.

5. The PACS system speech statistics system based on the large model is characterized by: Includes large language model training module, speech conversion module and screen; The large language model training module collects a number of first natural language characters that meet the statistical request of the PACS system and a first SQL statement corresponding to the first natural language characters; The large language model training module performs word segmentation on the first natural language text, and extracts time-related words, department name words, and purpose-related words, and cross-edits the time-related words, department name words, and purpose-related words through the large language model training module to form a plurality of second natural language texts. The large language model training module generates a second SQL statement based on the second natural language text and establishes a mapping relationship between the two. The speech conversion module is used to convert the third natural language text spoken by the user into the second natural language text, and the large language model training module calls the corresponding second SQL statement based on the second natural language text; The PACS system generates statistical content based on the second SQL statement and displays it on the screen.

6. The PACS system speech statistics system based on a large model as claimed in claim 5, characterized in that: The speech conversion module extracts time-related words, department name words and purpose-related words from the third natural language text and generates the second natural language text.

7. The PACS system speech statistics system based on a large model as described in claim 5 or 6, characterized in that: The time-related vocabulary is one of this week, this month, last month, first three quarters, this year, and last year; The department name is one of radiology, pediatrics, and orthopedics; The purpose-type vocabulary is one of the number of examinations, the number of patients received, and the total amount of charges.

8. The PACS system speech statistics system based on a large model as claimed in claim 7, characterized in that: The PACS system caches the time words, department name words and purpose words in the second natural language text and displays them on the screen; The PACS system automatically generates statistical content based on the time-related vocabulary, department name vocabulary and purpose-related vocabulary displayed on the screen manually selected by the user.

9. The PACS system speech statistics system based on a large model as claimed in claim 8, characterized in that: The screen is a touch screen.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the large model-based PACS system speech statistics method as described in any one of claims 1 to 4 is implemented.

Images