TY - GEN
T1 - DiffGEPCI
T2 - 21st IEEE International Symposium on Biomedical Imaging, ISBI 2024
AU - Hu, Yuyang
AU - Kothapalli, Satya V.V.N.
AU - Gan, Weijie
AU - Sukstanskii, Alexander L.
AU - Wu, Gregory F.
AU - Goyal, Manu
AU - Yablonskiy, Dmitriy A.
AU - Kamilov, Ulugbek S.
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - We introduce a new framework called DiffGEPCI for cross-modality generation in magnetic resonance imaging (MRI) using a 2.5D conditional diffusion model. DiffGEPCI can synthesize high-quality Fluid Attenuated Inversion Recovery (FLAIR) and Magnetization Prepared-Rapid Gradient Echo (MPRAGE) images, without acquiring corresponding measurements, by leveraging multi-Gradient-Recalled Echo (mGRE) MRI signals as conditional inputs. DiffGEPCI operates in a two-step fashion: it initially estimates a 3D volume slice-byslice using the axial plane and subsequently applies a refinement algorithm (referred to as 2.5D) to enhance the quality of the coronal and sagittal planes. Experimental validation on real mGRE data shows that DiffGEPCI achieves excellent performance, surpassing generative adversarial networks (GANs) and traditional diffusion models.
AB - We introduce a new framework called DiffGEPCI for cross-modality generation in magnetic resonance imaging (MRI) using a 2.5D conditional diffusion model. DiffGEPCI can synthesize high-quality Fluid Attenuated Inversion Recovery (FLAIR) and Magnetization Prepared-Rapid Gradient Echo (MPRAGE) images, without acquiring corresponding measurements, by leveraging multi-Gradient-Recalled Echo (mGRE) MRI signals as conditional inputs. DiffGEPCI operates in a two-step fashion: it initially estimates a 3D volume slice-byslice using the axial plane and subsequently applies a refinement algorithm (referred to as 2.5D) to enhance the quality of the coronal and sagittal planes. Experimental validation on real mGRE data shows that DiffGEPCI achieves excellent performance, surpassing generative adversarial networks (GANs) and traditional diffusion models.
KW - diffusion models
KW - medical image synthesis
UR - http://www.scopus.com/inward/record.url?scp=85203364709&partnerID=8YFLogxK
U2 - 10.1109/ISBI56570.2024.10635694
DO - 10.1109/ISBI56570.2024.10635694
M3 - Conference contribution
AN - SCOPUS:85203364709
T3 - Proceedings - International Symposium on Biomedical Imaging
BT - IEEE International Symposium on Biomedical Imaging, ISBI 2024 - Conference Proceedings
PB - IEEE Computer Society
Y2 - 27 May 2024 through 30 May 2024
ER -