Dimensions overall score 7.0

GitHub Code Pulse

No public code linked for this paper yet.

• DuSCN-FusionNet achieves a mean balanced accuracy of 80.59% for ADHD classification.result 95%
• DuSCN-FusionNet achieves an AUC of 0.778 for ADHD classification.result 95%
• DuSCN-FusionNet achieves a mean balanced accuracy of 80.59% for ADHD classification.result 95%
• DuSCN-FusionNet achieves an Area Under the Curve (AUC) of 0.778 for ADHD classification.result 95%
• DuSCN-FusionNet utilizes dual-channel Structural Covariance Networks (SCNs) to capture inter-regional morphological relationships.method 95%
• The framework constructs intensity-based and heterogeneity-based SCNs using ROI-wise mean intensity and intra-regional variability descriptors.method 95%
• Auxiliary ROI-wise variability features and global statistical descriptors are integrated via late-stage fusion.method 95%
• Grad-CAM is adapted to the SCN domain to derive ROI-level importance scores for interpretability.method 95%
• The model is evaluated using stratified 10-fold cross-validation with a 5-seed ensemble strategy.method 95%
• DuSCN-FusionNet achieves precision, recall, and F1-scores of 81.66%, 80.59%, and 80.27%, respectively.result 95%
• DuSCN-FusionNet utilizes dual-channel Structural Covariance Networks (SCNs) to capture inter-regional morphological relationships.method 95%
• The framework constructs intensity-based and heterogeneity-based SCNs using ROI-wise mean intensity and intra-regional variability descriptors.method 95%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card