Validation of Deep Learning techniques for quality augmentation in diffusion MRI for clinical studies

Aja-Fernández, Santiago, Martín-Martín, Carmen, Planchuelo Gomez, Alvaro, Faiyaz, Abrar, Uddin, Md Nasir, Schifitto, Giovanni, Tiwari, Abhishek, Shigwan, Saurabh J., Kumar Singh, Rajeev, Zheng, Tianshu, Cao, Zuozhen, Wu, Dan, Blumberg, Stefano B., Sen, Snigdha, Goodwin-Allcock, Tobias, Slator, Paddy J. ORCID: https://orcid.org/0000-0001-6967-989X, Yigit Avci, Mehmet, Li, Zihan, Bilgic, Berkin, Tian, Qiyuan, Wang, Xinyi, Tang, Zihao, Cabezas, Mariano, Rauland, Amelie, Merhof, Dorit, Manzano Maria, Renata, Campos, Vinícius Paraníba, Santini, Tales, da Costa Vieira, Marcelo Andrade, HashemizadehKolowri, SeyyedKazem, DiBella, Edward, Peng, Chenxu, Shen, Zhimin, Chen, Zan, Ullah, Irfan, Mani, Merry, Abdolmotalleby, Hesam, Eckstrom, Samuel, Baete, Steven H., Filipiak, Patryk, Dong, Tanxin, Fan, Qiuyun, de Luis-García, Rodrigo, Tristán-Vega, Antonio and Pieciak, Tomasz 2023. Validation of Deep Learning techniques for quality augmentation in diffusion MRI for clinical studies. NeuroImage: Clinical 39 , 103483. 10.1016/j.nicl.2023.103483

Preview

PDF - Published Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (2MB) | Preview

Abstract

The objective of this study is to evaluate the efficacy of deep learning (DL) techniques in improving the quality of diffusion MRI (dMRI) data in clinical applications. The study aims to determine whether the use of artificial intelligence (AI) methods in medical images may result in the loss of critical clinical information and/or the appearance of false information. To assess this, the focus was on the angular resolution of dMRI and a clinical trial was conducted on migraine, specifically between episodic and chronic migraine patients. The number of gradient directions had an impact on white matter analysis results, with statistically significant differences between groups being drastically reduced when using 21 gradient directions instead of the original 61. Fourteen teams from different institutions were tasked to use DL to enhance three diffusion metrics (FA, AD and MD) calculated from data acquired with 21 gradient directions and a b-value of 1000 s/mm2. The goal was to produce results that were comparable to those calculated from 61 gradient directions. The results were evaluated using both standard image quality metrics and Tract-Based Spatial Statistics (TBSS) to compare episodic and chronic migraine patients. The study results suggest that while most DL techniques improved the ability to detect statistical differences between groups, they also led to an increase in false positive. The results showed that there was a constant growth rate of false positives linearly proportional to the new true positives, which highlights the risk of generalization of AI-based tasks when assessing diverse clinical cohorts and training using data from a single group. The methods also showed divergent performance when replicating the original distribution of the data and some exhibited significant bias. In conclusion, extreme caution should be exercised when using AI methods for harmonization or synthesis in clinical studies when processing heterogeneous data in clinical studies, as important information may be altered, even when global metrics such as structural similarity or peak signal-to-noise ratio appear to suggest otherwise.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Computer Science & Informatics Psychology
Publisher:	Elsevier
ISSN:	2213-1582
Date of First Compliant Deposit:	7 August 2023
Date of Acceptance:	25 July 2023
Last Modified:	17 Apr 2024 14:41
URI:	https://orca.cardiff.ac.uk/id/eprint/161499

Actions (repository staff only)

Edit Item