一种基于决策树结构对比学习的抑郁症诊断分析方法
By employing a contrastive learning method based on decision tree structure, combined with MRI and SNP information, the problem of insufficient utilization of multimodal data in existing technologies has been solved, thereby improving the accuracy and interpretability of depression diagnosis, especially in the analysis of disease-related features.
CN119153035BActive Publication Date: 2026-07-17NANJING FORESTRY UNIV
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NANJING FORESTRY UNIV
- Filing Date
- 2024-08-13
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing technologies struggle to effectively utilize multimodal data, particularly imaging and genetic information, in the diagnosis of depression, resulting in insufficient diagnostic accuracy and interpretability.
Method used
A contrastive learning method based on decision tree structure is adopted, which combines brain magnetic resonance imaging (MRI) and single nucleotide polymorphism (SNP) information. The classification network is trained by fusing multimodal data, and feature mining and diagnostic analysis are performed using multimodal data.
Benefits of technology
It improves the accuracy and interpretability of depression diagnosis, enabling the analysis of disease-related brain regions and risk genes, thus enhancing the accuracy and stability of diagnosis.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN119153035B_ABST
Abstract
本发明公开了一种基于决策树结构对比学习的抑郁症诊断分析方法，包括：步骤1：首先对拥有的数据进行预处理；步骤2：选取其中一个模态的特征来描述模型的编码过程；步骤3：将选择后的特征输入编码器；步骤4：将编码所得到的yi输入投影编码器之中进行特征降维；步骤5：计算分类损失；步骤6：根据表征构建对比学习损失；步骤7：重复所述步骤2至6，得到四种模态数据各自的编码器与表征y与z；步骤8：根据Lall的梯度方向，对上述各编码器中能够调整的权重按梯度下降方向进行迭代更新，完成模型的学习过程。本发明利用多种模态的数据对MDD进行诊断，并结合拥有白盒特性的决策树结构来对MDD相关的风险基因与脑区进行分析，有效提升了MDD的诊断精度。
Need to check novelty before this filing date? Find Prior Art