Artificial intelligence recognition method and device for osteosarcoma images based on deep learning

CN116721088BActive Publication Date: 2024-11-29BEIJING JISHUITAN HOSPITAL
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202310735620.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-06-20
Publication Date
2024-11-29
Estimated Expiration
2043-06-20

AI Technical Summary

Technical Problem

The existing osteosarcoma recognition methods rely on the experience of medical staff, consume time and energy, and are inefficient in identification.

Method used

Using a deep learning-based method, by acquiring multimodal osteosarcoma images for fusion processing, using pre-trained feature extraction models and osteosarcoma recognition models, extracting feature vectors corresponding to multiple tags, and osteosarcoma recognition is performed based on these feature vectors. , improve identification accuracy and efficiency.

Benefits of technology

It realizes efficient and high-precision osteosarcoma image recognition, reduces the recognition time and energy of medical staff, and provides strong support for the early diagnosis and treatment of osteosarcoma.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116721088B_ABST
    Figure CN116721088B_ABST
Patent Text Reader

Abstract

The present invention provides an artificial intelligence recognition method and device for osteosarcoma images based on deep learning, including: obtaining a set of osteosarcoma images to be recognized, the osteosarcoma image set includes osteosarcoma images of multiple modalities, and fusing each osteosarcoma image in the osteosarcoma image set to obtain a target osteosarcoma image; extracting features from the target osteosarcoma image through a pre-trained feature extraction model to obtain a first feature vector corresponding to multiple labels; and recognizing osteosarcoma from the target osteosarcoma image based on the first feature vector corresponding to each label through a pre-trained osteosarcoma recognition model to obtain a first osteosarcoma recognition result. The present invention can recognize osteosarcoma images efficiently and accurately, thereby better assisting medical staff and providing strong support for the early diagnosis and treatment of osteosarcoma.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Pneumonia image classification method based on NASNet

    CN113052236A

  • Bone tumor focus analysis system

    CN114757894A