Medical record extraction and structuring artificial intelligence device based on AI large model

Abstract

The invention discloses a medical record extraction and structuring artificial intelligence device based on an AI large model. The device comprises a file receiving device, a file analyzing device, data structuring processing, data cleaning and checking and result presenting and exporting. The data value is maximized, and the efficiency and accuracy of medical record data arrangement are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of medical informationization and artificial intelligence technology, and specifically to an artificial intelligence device for medical record extraction and structuring based on an AI big model. The invention aims to automatically and efficiently extract key information from complex and diverse medical record materials through advanced AI algorithms, and provide strong data support for clinical research, medical data analysis, etc. Background Art

[0002] In the current medical system, medical records are important documents that record patients' conditions, treatment processes, and effects. However, the compilation and analysis of medical record information is often time-consuming and labor-intensive, and prone to errors, which seriously restricts the efficient use of medical data. With the rapid development of artificial intelligence technology, especially breakthroughs in natural language processing (NLP) and machine learning, it is possible to automatically extract medical record information. Summary of the invention

[0003] The purpose of the present invention is to provide an artificial intelligence device and method for medical record extraction and structuring based on an AI big model, which can realize automatic parsing and structuring of medical record texts by integrating advanced natural language processing, data cleaning, verification and standardization processing technologies, thereby improving the efficiency and accuracy of medical data processing.

[0004] To achieve the above technical objectives, the technical solution of the present invention is implemented as follows: an artificial intelligence device for medical record extraction and structuring based on an AI large model, comprising the following steps:

[0005] S1: The file receiving device is used to receive the medical record file (PDF) uploaded by the user and perform necessary preprocessing, perform format detection on the file, ensure the file type is compatible and remove irrelevant metadata, and convert the file content into a unified encoding format for subsequent processing;

[0006] S2: extracting plain text content from the preprocessed file through the file parsing device, processing the medical record file in image format using OCR (optical character recognition) technology to ensure accurate extraction of text content, segmenting the text using NLP tools, and then identifying key entities in the medical record, such as the patient's name, gender, height and weight, drug name, examination items, etc., further analyzing the relationship between entities, and constructing a semantic network of medical records;

[0007] S3: In data structured processing, pre-trained language models in the medical field (BioBERT, ClinicalBERT, etc.) are used to improve the accuracy and efficiency of parsing. The parsing process is divided into two stages: preliminary parsing (word segmentation, part-of-speech tagging) and in-depth parsing (entity recognition, relationship extraction), and the required information is gradually extracted; a reasonable data model (JSON, XML, etc.) is designed to store the extracted information in a standardized and structured manner;

[0008] S4: Data cleaning and verification, error handling: Design error handling mechanisms for errors or uncertainties in the parsing and mapping process, automatic correction, and manual intervention; Data cleaning: Clean the mapped data, remove redundant information, merge duplicate data, and improve data quality; Template adjustment: Dynamically adjust and optimize the data structure template based on feedback and demand changes in actual applications;

[0009] S5: Result presentation and export: present the processed structured data in a user-friendly manner, including generating detailed reports, displaying key information through web pages, and clicking on the original medical records for intuitive comparison; providing an export function to export result data as needed.

[0010] Beneficial effects of the present invention: Through the above embodiments, the present invention can realize the extraction and structured processing of medical records, improve the efficiency and accuracy of data collection, and provide strong support for clinical trials and data analysis in the medical industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is an overall implementation flow chart of an artificial intelligence device for medical record extraction and structuring based on an AI big model of the present invention.

Claims

1. An artificial intelligence device for medical record extraction and structuring based on an AI big model, characterized by The following steps are involved: S1: The file receiving device is used to receive the medical record file (PDF) uploaded by the user and perform necessary preprocessing, format verification, unified coding, etc., in preparation for subsequent analysis; S2: extracting plain text content from the preprocessed file through the file parsing device, segmenting the text using NLP tools, and then identifying key entities in the medical record, such as height and weight, drug name, examination items, etc., further analyzing the relationship between entities, and constructing a semantic network of medical records; S3: In data structuring processing, the device converts unstructured text data into a structured data format through machine learning or deep learning algorithms; S4: Data cleaning and verification, error handling, data cleaning, template adjustment and other operations are performed on the data. According to the knowledge base and rules in the medical field, the structured data is verified and the data is structured into a standardized format for subsequent storage; S5: Result presentation and export: The processed structured data is presented in a user-friendly manner and an export function is provided to export the recognition result data as needed.

2. According to claim 1, an artificial intelligence device for medical record extraction and structuring based on an AI big model comprises a file receiving device unit, a file parsing device unit, a data structuring processing unit, a data cleaning and verification unit, and a result presentation and export unit, and each unit is connected by a data bus.

3. According to the artificial intelligence device for medical record extraction and structuring based on AI big model in claim 1, the receiving file device unit is characterized in that: The entrance to the entire device is responsible for receiving medical records uploaded by users and supports PDF file format. It realizes automatic detection and preprocessing of files (format verification, unified encoding) to provide a clean data source for subsequent analysis.

4. According to the artificial intelligence device for medical record extraction and structuring based on AI big model in claim 1, the file parsing device unit is characterized in that: Extract the text content from the medical record file. Through NLP technology, unstructured or semi-structured text can be effectively parsed. Use OCR (optical character recognition) technology to process medical record files in image format to ensure accurate extraction of text content; use natural language processing (NLP) technology to perform preliminary processing on the text, such as word segmentation and part-of-speech tagging.

5. According to the artificial intelligence device for extracting and structuring medical records based on an AI big model as described in claim 1, the data structuring processing unit is characterized in that: The device converts unstructured text data into a structured data format through machine learning or deep learning algorithms. It introduces pre-trained models (BERT, RoBERTa) to improve the accuracy and efficiency of entity recognition and relationship extraction, and uses a well-designed data model (JSON, XML) to store the extracted information in a standardized and structured manner.

6. According to the artificial intelligence device for medical record extraction and structuring based on AI big model in claim 1, the data cleaning and verification unit is characterized by: Perform operations such as error handling, data cleaning, and template adjustment on the data, verify the structured data according to the knowledge base and rules in the medical field, and structure the data into a standardized format to facilitate subsequent storage and exchange.

7. According to the artificial intelligence device for medical record extraction and structuring based on AI big model in claim 1, the result presentation and export unit is characterized in that: Present structured medical record data in a user-friendly manner. Users can easily export recognition result data as needed to maximize the value of the data.

Images