Search

CN-121999954-A - Automatic generation method, system, equipment and storage medium for oral cavity image report

CN121999954ACN 121999954 ACN121999954 ACN 121999954ACN-121999954-A

Abstract

The application provides an automatic generation method of an oral cavity image report, which comprises the steps of receiving and processing oral cavity medical image data, classifying the processed oral cavity medical image data, matching the oral cavity medical image data based on a classification result with prestored oral cavity condition text data, wherein the oral cavity condition text data comprises a disease description text and a diagnosis suggestion text, associating corresponding image information and text information based on the matching result to generate a standardized diagnosis report corresponding to the oral cavity medical image, detecting and correcting potential errors in the standardized diagnosis report according to expert knowledge base data to generate a target report, and the expert knowledge base data comprises morphological characteristics of an anatomical structure, disease characteristics and diagnosis suggestions. The application is helpful for improving the standardized level of the oral medical diagnosis, the report is more visual, and the workload of doctors is reduced.

Inventors

  • YANG HUIFANG
  • LI GANG
  • XU XIAO

Assignees

  • 北京大学口腔医学院

Dates

Publication Date
20260508
Application Date
20241107

Claims (11)

  1. 1. An automatic generation method of an oral image report is characterized by comprising the following steps of, Receiving and processing the oral medical image data; classifying the processed oral medical image data, and matching the oral medical image data with prestored oral disease text data based on the classification result, wherein the oral disease text data comprises a disease description text and a diagnosis suggestion text; Based on the matching result, associating corresponding image information and text information, and generating a standardized diagnosis report corresponding to the oral medical image; And detecting and correcting potential errors in the standardized diagnosis report according to expert knowledge base data to generate a target report, wherein the expert knowledge base data comprises morphological characteristics of an anatomical structure, disease characteristics and diagnosis suggestions.
  2. 2. The method according to claim 1, wherein the processing of the oral medical image data includes denoising and enhancing the image by a preprocessing algorithm, and extracting features of the image data.
  3. 3. The method according to claim 1, wherein the step of classifying the processed oral medical image data and matching the processed oral medical image data based on the classification result and pre-stored oral condition text data comprises the steps of, Multitasking classification is carried out on the processed oral medical image data according to different attributes; And matching the corresponding text in the text data to the current classification of each attribute of the image according to the classification.
  4. 4. The method according to claim 1, wherein the step of generating a standardized diagnostic report corresponding to the oral medical image by associating corresponding image information and text information based on the matching result comprises, According to the matching result, the text description and the corresponding image area are marked and linked through an image-text matching algorithm based on an Attention mechanism and associated with the corresponding image information and the text information; and generating a model based on the text of the transducer, nesting the image information and the text information in a standardized template, and outputting a standardized diagnosis report.
  5. 5. The method for automatically generating an oral cavity image report according to claim 1, wherein the method for generating a target report comprises detecting differences between the standardized diagnostic report and the expert knowledge base information according to disease features and diagnosis advice conclusions corresponding to the expert knowledge base matched with the image data, using the standardized diagnostic report as a sampling information source for a part of the differences within a preset threshold range, using the knowledge base information conclusions as a sampling information source for a part of the differences exceeding the preset threshold, and fusing all the sampling information to generate the target report.
  6. 6. An automatic oral image report generation system, comprising The image data module is used for receiving and processing the oral medical image data; The text data module is used for storing oral disease text data, wherein the oral disease text data comprises disease description text and diagnosis suggestion text; the multi-mode fusion model is used for matching the classification result of classifying the processed oral medical image data with prestored oral disease text data; the image-text generating module is used for associating corresponding image information and text information based on the matching result; the natural language processing module is used for generating a standardized diagnosis report corresponding to the oral medical image; The expert knowledge base module is used for storing and managing expert knowledge base data in the oral cavity field, receiving image characteristics of the oral medical image data and matching corresponding knowledge base information, wherein the expert knowledge base data comprises an anatomical structure, disease characteristics and diagnosis standards; and the automatic error correction and quality control module is used for detecting and correcting potential errors in the standardized diagnosis report according to the expert knowledge base data to generate a target report.
  7. 7. The automated oral image report generating system according to claim 6, wherein the image data comprises one or more of a photograph, CT, MRI, intraoral scan data, ultrasound data, and the image data module processes the image data according to a set weight.
  8. 8. The automated oral image report generating system of claim 6, further comprising a user interaction module for user review and modification of the target report.
  9. 9. The automated oral image report generating system of claim 6, wherein the system comprises a transmission interface for connecting the system to a user electronic health system for transmitting the target report in the system to the user electronic health system.
  10. 10. An electronic device, comprising: processor, and A memory having executable code stored thereon, which when executed by the processor, causes the processor to perform the method of any of claims 1-5.
  11. 11. A storage medium having stored thereon executable code which, when executed by a processor of an electronic device, causes the processor to perform the method of any of claims 1-5.

Description

Automatic generation method, system, equipment and storage medium for oral cavity image report Technical Field The application relates to the field of intelligent medical image analysis, in particular to an automatic generation method, system, equipment and storage medium for an oral image report. Background The oral medical image is used as an important tool for diagnosing oral diseases and is widely applied to clinical practice. Currently, in the oral diagnosis and treatment process, doctors need to manually read various image data (including MRI, CBCT, intraoral scanning, ultrasound, photographed photographs, etc.), and make diagnosis in combination with clinical symptoms of patients. However, image analysis often depends on the experience of doctors, and not only requires a lot of time, but also is susceptible to human factors. With the rapid development of artificial intelligence and deep learning technologies, systems for automatically generating image diagnosis reports have gradually become key to improving clinical efficiency and diagnosis accuracy. However, applications for automatic report generation in the oral field have challenges such as how to integrate images and text information effectively, how to guarantee report accuracy and clinical relevance. Disclosure of Invention In order to solve the problems, the invention provides an automatic generation system of an oral cavity image report, which utilizes a large model technology, a text generation technology and an image-text generation technology to automatically analyze an oral cavity medical image and generate an accurate and professional diagnosis report through the construction of a bottom expert information base. In a first aspect of the present application, there is provided an automatic oral image report generating method, comprising, Receiving and processing the oral medical image data; classifying the processed oral medical image data, and matching the oral medical image data with prestored oral disease text data based on the classification result, wherein the oral disease text data comprises a disease description text and a diagnosis suggestion text; Based on the matching result, associating corresponding image information and text information, and generating a standardized diagnosis report corresponding to the oral medical image; And detecting and correcting potential errors in the standardized diagnosis report according to expert knowledge base data to generate a target report, wherein the expert knowledge base data comprises morphological characteristics of an anatomical structure, disease characteristics and diagnosis suggestions. In an alternative embodiment, the processing the oral medical image data includes denoising and enhancing the image by a preprocessing algorithm, and extracting features of the image data. In an alternative embodiment, the classifying the processed oral medical image data and matching the processed oral medical image data with pre-stored oral condition text data based on the classification result comprises, Multitasking classification is carried out on the processed oral medical image data according to different attributes; And matching the corresponding text in the text data to the current classification of each attribute of the image according to the classification. In an alternative embodiment, the method includes generating a standardized diagnostic report corresponding to the oral medical image based on the matching result by associating corresponding image information with text information, According to the matching result, the text description and the corresponding image area are marked and linked through an image-text matching algorithm based on Attent ion mechanism and associated with the corresponding image information and text information; and generating a model based on the text of the transducer, nesting the image information and the text information in a standardized template, and outputting a standardized diagnosis report. In an alternative implementation, the method comprises the steps of detecting and correcting potential errors in the standardized diagnosis report according to the expert knowledge base data to generate a target report, detecting conclusion differences of the standardized diagnosis report and the expert knowledge base information according to disease features and diagnosis suggestion conclusions corresponding to the image data matching expert knowledge base, adopting the standardized diagnosis report as a sampling information source for the part of the difference within a preset threshold range, adopting the knowledge base information conclusions as the sampling information source for the part of the difference exceeding the preset threshold, and fusing all the sampling information to generate the target report. In a second aspect of the present application, there is provided an automatic oral image report generating system comprising The image data module is used for