Breast mass ultrasound image classification method and system based on deep learning

Through a deep learning-based breast mass ultrasound image classification method, combined with image quality scores and equipment interference characteristics, and dynamic calibration of confidence, the consistency and reliability problems of image classification in existing technologies are solved, and efficient and accurate breast mass ultrasound image classification is achieved.

CN120451685BActive Publication Date: 2025-09-09THE FIRST AFFILIATED HOSPITAL OF FUJIAN MEDICAL UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202510942905.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-07-09
Publication Date
2025-09-09
Estimated Expiration
2045-07-09

AI Technical Summary

Technical Problem

Existing breast mass ultrasound image classification methods have limited resolution, severe noise interference, and image quality that is significantly affected by equipment parameters and operator techniques, resulting in strong subjectivity in manual interpretation and insufficient diagnostic consistency. In addition, existing deep learning methods find it difficult to fully explore the potential correlations between multimodal images, affecting the generalization ability and reliability of the model.

Method used

A deep learning-based breast mass ultrasound image classification method is adopted. The image type, physical basic features, pathological sensitivity features and equipment interference features are extracted through a pre-trained unified feature extraction model. Combined with the image quality score and equipment interference features, a two-level cascade classification network and a cross-modal knowledge enhancer are used to dynamically calibrate the confidence level to achieve image classification.

Benefits of technology

It significantly improves the diagnostic efficiency and accuracy of ultrasound images of breast tumors, overcomes the problem of feature distortion caused by low-quality images, improves the recognition rate of malignant lesions, and provides high-precision and high-reliability intelligent decision-making support.

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

Abstract

The present invention provides a method and system for breast mass ultrasound image classification based on deep learning, which relates to the technical field of ultrasound image classification. The present invention realizes integrated processing of multiple ultrasound image types through a unified feature extraction model, thereby improving diagnostic efficiency. At the same time, the image quality score generated based on physical basic features quantifies the impact of probe pressure unevenness, equipment noise interference and motion artifacts on imaging quality, and dynamically adjusts the knowledge enhancement strength accordingly so that pathological sensitive features can be accurately compensated with the support of medical knowledge graphs. A two-level cascade classification network establishes a hierarchical decision-making mechanism to initiate subtype deep analysis of malignant lesions under the premise of ensuring rapid diagnosis, thereby improving the recognition rate. Further, through the image reliability factor reflecting the credibility of equipment acquisition, a model cognitive factor based on the self-consistency of decision logic is combined to form a dual-factor dynamic calibration system to provide support for efficient and accurate classification of breast mass ultrasound images.
Need to check novelty before this filing date? Find Prior Art