A paired CT and MRI dataset for advanced medical imaging applications

Translating between medical imaging modalities, such as computed tomography (CT) and magnetic resonance imaging (MRI), presents a formidable challenge due to their fundamentally distinct picture characteristics and contrasts. The existence of paired datasets across modalities is essential for the validation and progression of image-to-image translation techniques; nevertheless, such datasets are exceedingly rare. This article presents a novel paired multimodal dataset designed to facilitate the translation of medical imaging modalities from CT to T1- and T2-weighted MRI and vice versa. The dataset was compiled from 18 patients, assuring diverse representation across multiple diseases and demographics, providing a robust resource for algorithm validation and clinical research. Every patient profile comprises co-registered T1, T2, and CT images, meticulously confirmed by proficient radiologists to guarantee precision and dependability. This dataset is designed to function as a benchmark for researchers for developing sophisticated image-to-image translation methods and multimodal medical imaging algorithms. It overcomes a significant deficiency in the availability of paired multimodal datasets, thereby advancing both computational and clinical applications.