tDKI-Net: A Joint q-t Space Learning Network for Diffusion-Time-Dependent Kurtosis Imaging

• Published in: IEEE journal of biomedical and health informatics Vol. 28; no. 12; pp. 7300 - 7310
• Main Authors: Zheng, Tianshu, Ba, Ruicheng, Huang, Yongquan, Wu, Dan
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.12.2024
• Subjects: <named-content xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" content-type="math" xlink:type="simple"> <inline-formula> <tex-math notation="LaTeX"> q-t</tex-math> </inline-formula> </named-content> space learning; <named-content xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" content-type="math" xlink:type="simple"> <inline-formula> <tex-math notation="LaTeX"> t</tex-math> </inline-formula> </named-content>DKI; Algorithms; Brain - diagnostic imaging; Data models; Dictionaries; Diffusion Magnetic Resonance Imaging - methods; Estimation error; Fitting; Humans; Image Interpretation, Computer-Assisted - methods; Image Processing, Computer-Assisted - methods; Kurtosis; Magnetic resonance imaging; Neural Networks, Computer; quantitative magnetic resonance imaging; time-dependent diffusion magnetic resonance imaging; Training
• ISSN: 2168-2194; 2168-2208; 2168-2208
• DOI: 10.1109/JBHI.2024.3417259

Time-dependent diffusion magnetic resonance imaging (TDDMRI) is useful for the non-invasive characterization of tissue microstructure. These models require densely sampled q-t space data for microstructural fitting, leading to very time-consuming acquisition protocols. To overcome this problem, we present a joint q-t space model- t DKI-Net to estimate diffusion-time dependent kurtosis and the transmembrane exchange, using downsampled q-t space data. The t DKI-Net is composed of several q -Encoders and a t -Encoder, designed based on the extragradient mechanism, each integrated with their respective mapping networks. In the t DKI-Net, two types of encoders along with their mapping networks are employed sequentially to generate kurtosis at individual diffusion times and to fit the transmembrane exchange time (<inline-formula><tex-math notation="LaTeX">\boldsymbol{\tau}_{\boldsymbol{m}}</tex-math></inline-formula>) using the time-dependent kurtosis according to the Kärger's model. Meanwhile, we proposed a three-stage training strategy, including physics-informed self-supervised pretraining, DKI warm-up, and joint training, to match the network structure. Our results demonstrated that the proposed t DKI-Net could effectively accelerate t DKI scans, resulting in lower estimation error compared with other methods. Our proposed three-stage training strategy demonstrated superior results than those training from scratch, e.g., the normalized root mean square error (NRMSE) of <inline-formula><tex-math notation="LaTeX">\boldsymbol{\tau}_{\boldsymbol{m}}</tex-math></inline-formula> decreased by up to 1.4%. We also investigated the training data size effects and found that although we used one-subject training, the network achieved lower NRMSEs for <inline-formula><tex-math notation="LaTeX">\bm {K}_{\text {avg}}</tex-math></inline-formula>, <inline-formula><tex-math notation="LaTeX">\bm {K}_{\bm {0}}</tex-math></inline-formula> and <inline-formula><tex-math notation="LaTeX">\boldsymbol{\tau}_{\boldsymbol{m}}</tex-math></inline-formula> (2.50%, 3.04%, 10.86%) than previous work that used three-subject training (3.8%, 9.5%, 12.1%). t DKI-Net can considerably reduce the scan time by 10.5-fold by joint downsampling the q-t space data without compromising the estimation accuracy.