文献阅读2

03-04

Date:2022.12.11--03
Title:Multimodal medical image fusion using convolutional neural network and extreme learning machine Link:Frontiers | Multimodal medical image fusion using convolutional neural network and extreme learning machine (frontiersin.org)
Framework
Methodology	A novel fusion method on the multimodal medical images exploiting convolutional neural network (CNN) and extreme learning machine (ELM) is proposed. (CELM)
Results
Contributions	• A novel method based on CNN and ELM is proposed to deal with the fusion issue of multimodal medical images. • The traditional CNN model is integrated with ELM to be a modified version called convolutional extreme learning machine (CELM) which has not only much better performance, but also much faster running speed. • Experimental results demonstrate that the proposed method has obvious superiorities over the current typical ones in terms of both gray image fusion and color image fusion, which is beneficial to obviously enhancing the precision of disease detection and diagnosis directly.
Conclusion	CELM combines the advantages of both CNN and ELM. Compared with other typical fusion methods, the proposed one has obvious superiorities in terms of both subjective visual quality and objective metric values.
Notes	Speed up the experiment with ELM. ELM can get weight matrix. Their writing skills is good, especially in the methond part.

Date:2022.12.12--04
Title:A multistage multimodal deep learning model for disease severity assessment and early warnings of high-risk patients of COVID-19 Link:A multistage multimodal deep learning model for disease severity assessment and early warnings of high-risk patients of COVID-19 - PMC (nih.gov)​
Framework
Methodology	Sequential stage-wise learning, weight paramenters,Multistage multimodal deep learning model.
Results
Contributions	A multistage multimodal deep learning (MMDL) model to (1) first assess the patient’s current condition (i.e., the mild and severe symptoms), then (2) give early warnings to patients with mild symptoms who are at high risk to develop severe illness. Build a sequential stage-wise learning architecture. Design a two-layer multimodal feature extractor.The 1st-hierarchy aims to perform intra-modal feature learning and extraction, while the 2nd-hierarchy attempts to perform cross-modal feature fusion.
Conclusion	In this paper, we have conceived and implemented a multistage, multimodal deep learning (MMDL) model to assess the disease severity and forecast the disease progression of patients with COVID-19. In summary, the novelty of MMDL embodies sequential stage-wise learning with multimodal inputs. MMDL shows the advantage of studying whole courses of the disease compared to single-stage learning.
Notes	Sequential stage-wise learning with weight paramenters.