Volumetric Imitation Generative Adversarial Networks for Anatomical Human Body Modeling

Overview

Journal Bioengineering (Basel)

Date 2024 Feb 23

PMID 38391649

Authors

Jion Kim

Yan Li

Byeong-Seok Shin

Affiliations

Soon will be listed here.

Abstract

Volumetric representation is a technique used to express 3D objects in various fields, such as medical applications. On the other hand, tomography images for reconstructing volumetric data have limited utilization because they contain personal information. Existing GAN-based medical image generation techniques can produce virtual tomographic images for volume reconstruction while preserving the patient's privacy. Nevertheless, these images often do not consider vertical correlations between the adjacent slices, leading to erroneous results in 3D reconstruction. Furthermore, while volume generation techniques have been introduced, they often focus on surface modeling, making it challenging to represent the internal anatomical features accurately. This paper proposes volumetric imitation GAN (VI-GAN), which imitates a human anatomical model to generate volumetric data. The primary goal of this model is to capture the attributes and 3D structure, including the external shape, internal slices, and the relationship between the vertical slices of the human anatomical model. The proposed network consists of a generator for feature extraction and up-sampling based on a 3D U-Net and ResNet structure and a 3D-convolution-based LFFB (local feature fusion block). In addition, a discriminator utilizes 3D convolution to evaluate the authenticity of the generated volume compared to the ground truth. VI-GAN also devises reconstruction loss, including feature and similarity losses, to converge the generated volumetric data into a human anatomical model. In this experiment, the CT data of 234 people were used to assess the reliability of the results. When using volume evaluation metrics to measure similarity, VI-GAN generated a volume that realistically represented the human anatomical model compared to existing volume generation methods.

References

Jiang M, Zhi M, Wei L, Yang X, Zhang J, Li Y . FA-GAN: Fused attentive generative adversarial networks for MRI image super-resolution. Comput Med Imaging Graph. 2021; 92:101969. PMC: 8453331. DOI: 10.1016/j.compmedimag.2021.101969. View

Roschger P, Gupta H, Berzlanovich A, Ittner G, Dempster D, Fratzl P . Constant mineralization density distribution in cancellous human bone. Bone. 2003; 32(3):316-23. DOI: 10.1016/s8756-3282(02)00973-0. View

Taha A, Hanbury A . Metrics for evaluating 3D medical image segmentation: analysis, selection, and tool. BMC Med Imaging. 2015; 15:29. PMC: 4533825. DOI: 10.1186/s12880-015-0068-x. View

Munawar A, Li Z, Nagururu N, Trakimas D, Kazanzides P, Taylor R . Fully immersive virtual reality for skull-base surgery: surgical training and beyond. Int J Comput Assist Radiol Surg. 2023; 19(1):51-59. PMC: 11163985. DOI: 10.1007/s11548-023-02956-5. View

Nakao M, Nakamura M, Mizowaki T, Matsuda T . Statistical deformation reconstruction using multi-organ shape features for pancreatic cancer localization. Med Image Anal. 2020; 67:101829. DOI: 10.1016/j.media.2020.101829. View

Fang C, An J, Bruno A, Cai X, Fan J, Fujimoto J . Consensus recommendations of three-dimensional visualization for diagnosis and management of liver diseases. Hepatol Int. 2020; 14(4):437-453. PMC: 7366600. DOI: 10.1007/s12072-020-10052-y. View

Rister B, Yi D, Shivakumar K, Nobashi T, Rubin D . CT-ORG, a new dataset for multiple organ segmentation in computed tomography. Sci Data. 2020; 7(1):381. PMC: 7658204. DOI: 10.1038/s41597-020-00715-8. View

Xu C, Xu L, Ohorodnyk P, Roth M, Chen B, Li S . Contrast agent-free synthesis and segmentation of ischemic heart disease images using progressive sequential causal GANs. Med Image Anal. 2020; 62:101668. DOI: 10.1016/j.media.2020.101668. View

Han C, Rundo L, Murao K, Noguchi T, Shimahara Y, Milacski Z . MADGAN: unsupervised medical anomaly detection GAN using multiple adjacent brain MRI slice reconstruction. BMC Bioinformatics. 2021; 22(Suppl 2):31. PMC: 8073969. DOI: 10.1186/s12859-020-03936-1. View

10.

Ran M, Hu J, Chen Y, Chen H, Sun H, Zhou J . Denoising of 3D magnetic resonance images using a residual encoder-decoder Wasserstein generative adversarial network. Med Image Anal. 2019; 55:165-180. DOI: 10.1016/j.media.2019.05.001. View

11.

Dai W, Zhang H, Yu J, Xu H, Chen H, Luo S . CT Imaging and Differential Diagnosis of COVID-19. Can Assoc Radiol J. 2020; 71(2):195-200. PMC: 7140975. DOI: 10.1177/0846537120913033. View

12.

Wisse L, Chetelat G, Daugherty A, de Flores R, La Joie R, Mueller S . Hippocampal subfield volumetry from structural isotropic 1 mm MRI scans: A note of caution. Hum Brain Mapp. 2020; 42(2):539-550. PMC: 7775994. DOI: 10.1002/hbm.25234. View

13.

Kavur A, Gezer N, Baris M, Aslan S, Conze P, Groza V . CHAOS Challenge - combined (CT-MR) healthy abdominal organ segmentation. Med Image Anal. 2021; 69:101950. DOI: 10.1016/j.media.2020.101950. View

14.

Zhang L, Shen Y, Li H . VSI: a visual saliency-induced index for perceptual image quality assessment. IEEE Trans Image Process. 2014; 23(10):4270-81. DOI: 10.1109/TIP.2014.2346028. View

15.

Wang Q, Zhang X, Zhang W, Gao M, Huang S, Wang J . Realistic Lung Nodule Synthesis With Multi-Target Co-Guided Adversarial Mechanism. IEEE Trans Med Imaging. 2021; 40(9):2343-2353. DOI: 10.1109/TMI.2021.3077089. View

16.

Masutani E, Bahrami N, Hsiao A . Deep Learning Single-Frame and Multiframe Super-Resolution for Cardiac MRI. Radiology. 2020; 295(3):552-561. PMC: 7263289. DOI: 10.1148/radiol.2020192173. View

17.

Costa P, Galdran A, Meyer M, Niemeijer M, Abramoff M, Mendonca A . End-to-End Adversarial Retinal Image Synthesis. IEEE Trans Med Imaging. 2017; 37(3):781-791. DOI: 10.1109/TMI.2017.2759102. View

18.

Shi W, Xiaoping Liu P, Zheng M . Cutting procedures with improved visual effects and haptic interaction for surgical simulation systems. Comput Methods Programs Biomed. 2019; 184:105270. DOI: 10.1016/j.cmpb.2019.105270. View