Enhancing CT Image synthesis from multi-modal MRI data based on a multi-task neural network framework
This work addresses medical imaging challenges for healthcare applications, but it is incremental as it builds on existing Transformer U-Net architectures with multi-task extensions.
The study tackled the problem of synthesizing CT images from multi-modal MRI data by decomposing it into subtasks like skull segmentation and Hounsfield unit prediction, using a multi-task neural network framework based on an enhanced Transformer U-Net, and validated it on a public brain image dataset with comparisons between T1-weighted and T2-Flair images.
Image segmentation, real-value prediction, and cross-modal translation are critical challenges in medical imaging. In this study, we propose a versatile multi-task neural network framework, based on an enhanced Transformer U-Net architecture, capable of simultaneously, selectively, and adaptively addressing these medical image tasks. Validation is performed on a public repository of human brain MR and CT images. We decompose the traditional problem of synthesizing CT images into distinct subtasks, which include skull segmentation, Hounsfield unit (HU) value prediction, and image sequential reconstruction. To enhance the framework's versatility in handling multi-modal data, we expand the model with multiple image channels. Comparisons between synthesized CT images derived from T1-weighted and T2-Flair images were conducted, evaluating the model's capability to integrate multi-modal information from both morphological and pixel value perspectives.