Rapid and robust quantitative cartilage assessment for the clinical setting: deep learning-enhanced accelerated T2...

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Biomedical Engineering
• Radiology

Background:

• Clinical T2 mapping for cartilage assessment faces challenges including poor reproducibility and long scan times.
• Current methods require complex image analysis, limiting widespread adoption in clinical practice.

Purpose of the Study:

• To develop an accelerated deep learning (DL)-enhanced cartilage T2 mapping sequence (DL CartiGram).
• To assess the repeatability and reproducibility of DL CartiGram.
• To evaluate the accuracy of DL CartiGram compared to conventional T2 mapping using a semi-automatic pipeline.

Main Methods:

• DL CartiGram was implemented using a modified 2D Multi-Echo Spin-Echo sequence with parallel imaging and DL-based reconstruction at 3T.
• Phantom studies assessed intra-site repeatability and inter-site reproducibility.
• In vivo studies on 43 patients compared DL CartiGram with conventional T2 mapping, using DL knee segmentation and the DOSMA framework for T2 value extraction.

Main Results:

• Phantom tests demonstrated high intra-site repeatability (CV ≤ 2.52%) and inter-site reproducibility (CV = 2.74%, CCC = 99%).
• DL CartiGram reduced scan time by 40% in vivo.
• In vivo T2 measurements showed high accuracy (CV = 0.97%) with no significant differences compared to conventional T2 mapping (p=0.1).

Conclusions:

• DL CartiGram significantly accelerates cartilage T2 mapping while maintaining excellent repeatability and reproducibility.
• The combination of DL CartiGram and a semi-automatic post-processing pipeline offers a promising solution for quantitative T2 cartilage biomarker assessment in clinical settings.