Am method, system, device and medium for predicting the efficacy of amd based on feature fusion

By combining 3DResNet and radiomics feature extraction networks with LSTM and Transformer modules, and integrating deep learning, radiomics, and demographic features, the problem of accurate prediction of treatment efficacy for elderly patients with macular degeneration was solved, achieving high-quality efficacy prediction and automated image processing.

CN120932807BActive Publication Date: 2026-03-03SUZHOU UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-09-30
Publication Date
2026-03-03

AI Technical Summary

Technical Problem

Existing technologies struggle to accurately predict the efficacy of anti-vascular endothelial growth factor therapy in patients with age-related macular degeneration under limited data conditions. Furthermore, existing methods typically rely solely on radiomics features or deep learning features, rarely effectively integrating radiomics, deep learning features, and other clinical information, resulting in low prediction accuracy.

Method used

High-quality image features are obtained by using a three-dimensional residual network (3DResNet) and a radiomics feature extraction network. Combined with a pre-interactive long short-term memory (LSTM) network and a Transformer module, deep learning, radiomics and demographic features are integrated. The relationship between image features of patients at different time points is obtained through a time series analysis network to achieve efficacy prediction.

Benefits of technology

It improves the predictive accuracy of anti-vascular endothelial growth factor therapy efficacy in patients with age-related macular degeneration, enhances the automated processing capabilities of optical coherence tomography images, and provides technical support for clinical practice.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120932807B_ABST
    Figure CN120932807B_ABST
Patent Text Reader

Abstract

The application discloses an AMD efficacy prediction method, system, device and medium based on feature fusion, belongs to the technical field of efficacy prediction, and comprises the following steps: acquiring optical coherence tomography image data and demographic characteristics of an age-related macular degeneration patient; determining deep learning features and imaging features according to the optical coherence tomography image data; performing feature selection based on the deep learning features, the imaging features and the demographic characteristics to obtain input features; performing calculation by combining a time series analysis network with the input features to determine a hidden state; and performing feature change trend analysis according to the hidden state to determine an age-related macular degeneration efficacy prediction result. The application proposes a multi-feature fusion long short-term memory network, and realizes efficacy prediction of anti-vascular endothelial growth factor treatment for the age-related macular degeneration patient.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Method for predicting glucocorticoid treatment TAO by using deep learning in combination with radiomics

    CN116597937A

  • Treatment outcome prediction for neovascular age-related macular degeneration using baseline features

    CN117121113A