Renal artery ultrasound report generation method and system, electronic equipment and storage medium

By combining ultrasound case texts and multi-source knowledge to generate renal artery ultrasound reports, the shortcomings of existing technologies in diagnostic accuracy and terminology standardization are addressed, enabling precise diagnosis and standardized report output for renal artery stenosis.

CN122314224APending Publication Date: 2026-06-30BEIJING UNIV OF TECH
0 Cites 0 Cited by

Patent Information

Application Number
CN202610618527.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-05-07
Publication Date
2026-06-30

AI Technical Summary

Technical Problem

Existing methods for generating renal artery ultrasound reports rely on single-image visual information, resulting in insufficient diagnostic accuracy and standardized terminology, making it difficult to meet the clinical needs for accurate diagnosis and standardized report output for renal artery stenosis.

Method used

By performing keyword matching on the clinical examination text of ultrasound cases, a diagnostic factor pool is constructed. This pool is then encoded by combining general medical knowledge from a large model with specialized knowledge from clinical experts to generate a global dual-source knowledge embedding. By fusing visual global features and knowledge fusion residual features, a renal artery ultrasound diagnostic report is generated.

Benefits of technology

It significantly improves the diagnostic accuracy, terminology standardization, and clinical adaptability of reports, meeting the needs for accurate diagnosis and standardized reporting of renal artery stenosis.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a renal artery ultrasound report generation method and system, electronic equipment and a storage medium, and the method comprises the steps: carrying out the keyword matching of a clinical examination text of an ultrasound case, obtaining a renal artery diagnosis index, generating a retrieval prompt word based on the renal artery diagnosis index, obtaining the general medical knowledge of a large model and the specialized knowledge of clinical experts, and carrying out the retrieval of the retrieval prompt word. Global double-source knowledge embedding is obtained; basic embedding is generated for clinical examination text coding of the ultrasound case, and knowledge fusion residual features are obtained according to the basic embedding and global double-source knowledge embedding; feature extraction is performed on a renal artery multi-frame ultrasonic image to be detected to obtain visual global features, and element-by-element additive fusion is performed on the visual global features, global double-source knowledge embedding and knowledge fusion residual features to obtain global unified features; and obtaining a renal artery ultrasonic diagnosis report according to the global unified features. Through the setting, the diagnosis accuracy and term normalization of the report can be improved.
Need to check novelty before this filing date? Find Prior Art