Integration of multimodal imaging data with machine learning for improved diagnosis and prognosis in neuroimaging DOI Creative Commons
Saurabh Bhattacharya, Sashikanta Prusty,

Sanjay P. Pande

et al.

Frontiers in Human Neuroscience, Journal Year: 2025, Volume and Issue: 19

Published: March 21, 2025

Introduction Combining many types of imaging data—especially structural MRI (sMRI) and functional (fMRI)—may greatly assist in the diagnosis treatment brain disorders like Alzheimer’s. Current approaches are less helpful for forecasting, however, as they do not always blend spatial temporal patterns from different sources properly. This work presents a novel mixed deep learning (DL) method combining data using CNN, GRU, attention techniques. introduces hybrid Dynamic Cross-Modality Attention Module to help more efficiently data. Through working around issues with current multimodal fusion techniques, our approach increases accuracy readability diagnoses. Methods Utilizing CNNs models dynamics fMRI connection measures utilizing GRUs, proposed extracts characteristics sMRI. Strong integration is made possible by including an mechanism give diagnostically important features top priority. Training evaluation model took place Human Connectome Project (HCP) dataset behavioral data, fMRI, Measures include accuracy, recall, precision F1-score used evaluate performance. Results It was correct 96.79% time combined structure. Regarding identification disorders, successful than existing ones. Discussion These findings indicate that strategy makes sense complimentary information several kinds photos. detail helped one choose which aspects concentrate on, thereby enhancing diagnostic accuracy. Conclusion The offers fresh benchmark neuroimaging analysis has great potential use real-world assessment prediction. Researchers will investigate future applications this technique new picture clinical

Language: Английский

Integration of multimodal imaging data with machine learning for improved diagnosis and prognosis in neuroimaging DOI Creative Commons
Saurabh Bhattacharya, Sashikanta Prusty,

Sanjay P. Pande

et al.

Frontiers in Human Neuroscience, Journal Year: 2025, Volume and Issue: 19

Published: March 21, 2025

Introduction Combining many types of imaging data—especially structural MRI (sMRI) and functional (fMRI)—may greatly assist in the diagnosis treatment brain disorders like Alzheimer’s. Current approaches are less helpful for forecasting, however, as they do not always blend spatial temporal patterns from different sources properly. This work presents a novel mixed deep learning (DL) method combining data using CNN, GRU, attention techniques. introduces hybrid Dynamic Cross-Modality Attention Module to help more efficiently data. Through working around issues with current multimodal fusion techniques, our approach increases accuracy readability diagnoses. Methods Utilizing CNNs models dynamics fMRI connection measures utilizing GRUs, proposed extracts characteristics sMRI. Strong integration is made possible by including an mechanism give diagnostically important features top priority. Training evaluation model took place Human Connectome Project (HCP) dataset behavioral data, fMRI, Measures include accuracy, recall, precision F1-score used evaluate performance. Results It was correct 96.79% time combined structure. Regarding identification disorders, successful than existing ones. Discussion These findings indicate that strategy makes sense complimentary information several kinds photos. detail helped one choose which aspects concentrate on, thereby enhancing diagnostic accuracy. Conclusion The offers fresh benchmark neuroimaging analysis has great potential use real-world assessment prediction. Researchers will investigate future applications this technique new picture clinical

Language: Английский

Citations

0