Multi Modal MRI

Multimodal MRI leverages data from multiple MRI sequences to improve diagnostic accuracy and treatment planning by integrating complementary information about tissue properties. Current research focuses on optimizing data fusion strategies within deep learning models, including exploring various architectures like UNets and Transformers, and developing methods for handling incomplete or imperfectly registered datasets. These advancements aim to enhance the robustness and accuracy of automated segmentation and classification tasks for various pathologies, ultimately improving clinical workflows and patient care. The field is also actively investigating explainability methods to increase the transparency and trustworthiness of these powerful AI tools in medical applications.

Papers

April 8, 2022

A Learnable Variational Model for Joint Multimodal MRI Reconstruction and Synthesis
Wanyu Bian, Qingchao Zhang, Xiaojing Ye, Yunmei Chen
Critical Synthesis Variational Learning Multi Modal MRI Novel Deep Learning Model

November 8, 2021

Feature-enhanced Generation and Multi-modality Fusion based Deep Neural Network for Brain Tumor Segmentation with Missing MR Modalities
Tongxue Zhou, Stéphane Canu, Pierre Vera, Su Ruan
Deep Neural Network Brain Tumor Segmentation Feature Generation Multi Modal MRI Multi Modality Fusion Modality Missing Brain Image

Multi Modal MRI

Papers

A Learnable Variational Model for Joint Multimodal MRI Reconstruction and Synthesis

Feature-enhanced Generation and Multi-modality Fusion based Deep Neural Network for Brain Tumor Segmentation with Missing MR Modalities