Computerized Medical Imaging and Graphics 108 (2023) 102272 Available online 20 July 2023 0895-6111/© 2023 Elsevier Ltd. All rights reserved. Contents lists available at ScienceDirect Computerized Medical Imaging and Graphics journal homepage: www.elsevier.com/locate/compmedimag Cross modality generative learning framework for anatomical transitive Magnetic Resonance Imaging (MRI) from Electrical Impedance Tomography (EIT) image ✩ Zuojun Wang a,1 , Mehmood Nawaz b,1 , Sheheryar Khan c , Peng Xia a , Muhammad Irfan d , Eddie C. Wong e , Russell Chan e , Peng Cao a,∗ a The Department of Diagnostic Radiology, The University of Hong Kong, Hong Kong b The Department of Biomedical Engineering, The Chinese University of Hong Kong, Hong Kong c School of Professional Education and Executive Development, The Hong Kong Polytechnic University, Hong Kong d Faculty of Electrical Engineering, Ghulam Ishaq Khan Institute of Engineering Sciences and Technology, Pakistan e Gense Technologies Ltd., Hong Kong ARTICLE INFO Keywords: Electrical Impedance Tomography Magnetic Resonance Image Generative networks Medical images ABSTRACT This paper presents a cross-modality generative learning framework for transitive magnetic resonance imaging (MRI) from electrical impedance tomography (EIT). The proposed framework is aimed at converting low- resolution EIT images to high-resolution wrist MRI images using a cascaded cycle generative adversarial network (CycleGAN) model. This model comprises three main components: the collection of initial EIT from the medical device, the generation of a high-resolution transitive EIT image from the corresponding MRI image for domain adaptation, and the coalescence of two CycleGAN models for cross-modality generation. The initial EIT image was generated at three different frequencies (70 kHz, 140 kHz, and 200 kHz) using a 16-electrode belt. Wrist T1-weighted images were acquired on a 1.5T MRI. A total of 19 normal volunteers were imaged using both EIT and MRI, which resulted in 713 paired EIT and MRI images. The cascaded CycleGAN, end- to-end CycleGAN, and Pix2Pix models were trained and tested on the same cohort. The proposed method achieved the highest accuracy in bone detection, with 0.97 for the proposed cascaded CycleGAN, 0.68 for end- to-end CycleGAN, and 0.70 for the Pix2Pix model. Visual inspection showed that the proposed method reduced bone-related errors in the MRI-style anatomical reference compared with end-to-end CycleGAN and Pix2Pix. Multifrequency EIT inputs reduced the testing normalized root mean squared error of MRI-style anatomical reference from 67.9% ± 12.7% to 61.4% ± 8.8% compared with that of single-frequency EIT. The mean conductivity values of fat and bone from regularized EIT were 0.0435 ± 0.0379 S/m and 0.0183 ± 0.0154 S/m, respectively, when the anatomical prior was employed. These results demonstrate that the proposed framework is able to generate MRI-style anatomical references from EIT images with a good degree of accuracy. 1. Introduction Electrical impedance tomography (EIT) is a noninvasive imaging technique that visualizes the conductivity distribution of the interior structure when a weak low-frequency current is applied to the surface of the object (Liu et al., 2017; Seppänen et al., 2009; Wang et al., 2004). The boundary voltage measurements are used to generate an internal image of the spatial conductivity distribution. The reconstruction from ✩ This work is supported by the Hong Kong Innovation and Technology Fund (PRP/014/20FX) and the Research Grants Council of the Hong Kong SAR (UGC/FDS24/E18/22). ∗ Corresponding author. E-mail addresses: u3006796@connect.hku.hk (Z. Wang), mehmoodnawaz@cuhk.edu.hk (M. Nawaz), caopeng1@hku.hk (P. Cao). 1 These authors contributed equally to this work. a single set of measurements in a given time is called absolute EIT, which has been proven to be highly sensitive to errors, such as inaccu- racies in the assumed model geometry and nonideal properties of the electrodes and amplifiers (Nissinen et al., 2009). Frequency-difference EIT measures the conductivity at two or more excitation frequencies, and obtains an image of the difference in electrical properties be- tween the frequencies. EIT is a low-cost, highly portable, real-time, and radiation-free imaging device suitable for bedside diagnosis and https://doi.org/10.1016/j.compmedimag.2023.102272 Received 9 May 2023; Received in revised form 4 July 2023; Accepted 8 July 2023