A fat fraction phantom for establishing new convolutional neural network to determine the pancreatic fat deposition

John Zhiyong Yang; Rinki Murphy; Jun Lu

doi:10.1016/j.heliyon.2022.e12478

A fat fraction phantom for establishing new convolutional neural network to determine the pancreatic fat deposition

Heliyon. 2022 Dec 21;8(12):e12478. doi: 10.1016/j.heliyon.2022.e12478. eCollection 2022 Dec.

Authors

John Zhiyong Yang¹, Rinki Murphy^{2

3

4

5

6}, Jun Lu^{1

6

7

8}

Affiliations

¹ Auckland Bioengineering Institute, The University of Auckland, Auckland, New Zealand.
² School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand.
³ Department of Surgery, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand.
⁴ Auckland Diabetes Centre, Auckland District Health Board, Auckland, New Zealand.
⁵ Whitiora Diabetes Department, Counties Manukau District Health Board, Auckland, New Zealand.
⁶ Maurice Wilkins Centre for Biodiscovery, Auckland, New Zealand.
⁷ College of Food Engineering and Nutrition Sciences, Shanxi Normal University, Xi'an, 710119, Shanxi Province, China.
⁸ Key Laboratory of Forest Plant Ecology, Ministry of Education, Northeast Forestry University, Harbin 150040, China.

Abstract

The determination of fat fraction based on Magnetic Resonance Imaging (MRI) requires extremely accurate data reconstruction for the assessment of pancreatic fat accumulation in medical diagnostics and biological research. In this study, the signal model of the oil and water emulsion was created with a 3.0 T field strength. We examined the quantification of the fat fraction from phantom and the intrapancreatic fat fraction using the techniques of magnetic resonance spectroscopy (MRS) and Iterative Decomposition with Echo Asymmetry and Least-Squares estimate (IDEAL) in magnetic resonance imaging (MRI). Additionally, we contrasted expert manual pancreatic fat assessment with MRS and IDEAL pancreatic fat fraction quantification. There was a strong connection between the true fat volume fraction and the fat fraction from IDEAL and MRS (R² = 0.99 and 0.99, respectively). For both phantom and in vivo measurements, Pearson's correlation and linear regression analysis were used. The findings of the in vivo assessment revealed a variable correlation between the pancreatic fat fraction MRI readings. We also used MR-opsy for manual pancreatic fat fraction segmentation since it read pancreatic fat fractions more accurately than IDEAL and MRS, which aided in the development of machine learning's ability to assess pancreatic fat automatically.

Keywords: Automatic measurement; Intra-pancreatic fat; MRI; Machine learning; Phantom.