Abstract
Purpose To analyze recent artificial intelligence (AI)-driven magnetic resonance (MR) reconstruction methods aimed at addressing long acquisition times, image artifacts, and workflow inefficiencies.
Method A systematic literature review was conducted using PubMed, Scopus, and Scopus AI, targeting peer‑reviewed studies published between 2015 and 2026. Search terms included “MRI,” “magnetic resonance imaging,” “artificial intelligence,” “deep learning,” and “machine learning.” Eligible studies included methodology, validation, and review papers. Extracted data included algorithm type, study design, key findings, and limitations. Themes were identified through content analysis and generative AI use.
Results Prominent themes identified were the emergence of generative adversarial networks, advanced neural networks forming the backbone of MR reconstruction, and clinical implementation considerations.
Discussion Generative models improved reconstruction quality and high‑frequency detail, but faced challenges related to computational cost and limited clinical validation. Advanced neural network architectures enhanced global feature modeling, interpretability, and robustness across imaging conditions. Clinical implementation of AI-driven MR reconstruction offered substantial promise for improving patient care. However, the transition from research to routine clinical practice requires addressing multifaceted challenges.
Conclusion AI‑based MR reconstruction is advancing rapidly, offering meaningful improvements. Continued research should prioritize standardized frameworks, clinically relevant metrics, and validation across diverse patient populations and imaging protocols.
Graduation Date
Summer 8-7-2026
Document Type
Thesis
Degree Name
Master of Science in Radiologic Science
Program
The School of Health Professions
Faculty Advisor
Kevin R. Clark
Committee Member
Jessyca B. Wagner
Director, Graduate Program
William A. Undie
Dean
Kimberly Hoggatt Krumwiede
Recommended Citation
Jones, J. R. High-Definition Healing: Investigating AI-Driven MR Reconstruction Methods. [thesis]. University of Texas MD Anderson Cancer Center; 2026. Retrieved from https://openworks.mdanderson.org/shp_radsci/11
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
