文献阅读1

01-02

Date:2022.12.09--01
Title: Multimodal Attention-based Deep Learning for Alzheimer’s Disease Diagnosis Link:Multimodal attention-based deep learning for Alzheimer's disease diagnosis - PubMed (nih.gov)
Framework
Methodology	Three nerworks for different features. Self Attention+Cross-modal Attention
Results
Contributions	We offer three distinct contributions: integrating multimodal inputs, multi-task classification, and cross-modal attention for capturing interactions.
Conclusion	The performance of MADDi was superior to that of existing multimodal machine learning methods and was shown to be consistently high regardless of chance initialization.
Notes

Date:2022.12.10--02
Title: Deep learning for the diagnosis of suspicious thyroid nodules based on multimodal ultrasound images Link:Frontiers | Deep learning for the diagnosis of suspicious thyroid nodules based on multimodal ultrasound images (frontiersin.org)
Framework
Methodology	ResNet-50+convolutional bottleneck attention module+spatial attention +channel attention, fivefold cross-validation,Statistical analysis
Results
Contributions	ResNet-50+convolutional bottleneck attention module(extract feature);spatial attention +channel attention(fusion)
Conclusion	The DL models based on multimodal US images showed exceptional performance in the differential diagnosis of suspicious TNs, effectively increased the diagnostic efficacy of TN evaluations by junior radiologists, and provided an objective assessment for the clinical and surgical management phases that follow.
Notes