Buko Erick O, Parvaze Suhail P, Johnson Casey P
Department of Veterinary Clinical Sciences, University of Minnesota, St. Paul, Minnesota, USA.
Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, USA.
Magn Reson Med. 2025 Jul 17. doi: 10.1002/mrm.30636.
Intravoxel incoherent motion (IVIM) diffusion-weighted imaging of tissue perfusion and diffusion has limited signal in low-perfused tissues, which poses a challenge in measuring IVIM parameters. This work proposes an optimized analytical segmented (opAS) method for extracting IVIM parameters and compares its performance with prior segmented (S), oversegmented (OS), and analytical segmented (AS) fitting methods.
The four fitting methods were compared using numerical simulations and retrospective analysis of IVIM data of a piglet model of bone marrow ischemia. The simulations compared the accuracy and precision of the four methods for different combinations of IVIM perfusion-related parameters (D*, pseudo-diffusion coefficient; and f, pseudo-diffusion fraction) at a fixed molecular-diffusion coefficient (D). In vivo data from n = 11 piglets were analyzed using the four methods to compare their sensitivities in detecting lack of perfusion in ischemic versus perfused pairs of femoral heads.
In simulation, opAS had the lowest errors in estimating D, f, D*, and f × D* across all signal-to-noise ratio (SNR) levels, particularly in challenging parameter combinations of low f and D*. In the piglet model, all four IVIM fitting methods consistently found that D increased and f and f × D* decreased in the ischemic versus perfused femoral heads. However, D* was only consistently decreased in the ischemic versus perfused femoral heads using the AS and opAS methods.
The opAS method more accurately estimates IVIM parameters than the S, OS, and AS approaches, particularly perfusion-related parameters (D* and f) in low-perfused tissues.
组织灌注和扩散的体素内不相干运动（IVIM）扩散加权成像在低灌注组织中的信号有限，这给测量IVIM参数带来了挑战。本研究提出一种优化的分析分割（opAS）方法来提取IVIM参数，并将其性能与先前的分割（S）、过度分割（OS）和分析分割（AS）拟合方法进行比较。
使用数值模拟和对骨髓缺血仔猪模型的IVIM数据进行回顾性分析，对这四种拟合方法进行比较。模拟比较了在固定分子扩散系数（D）下，四种方法对IVIM灌注相关参数（D*，伪扩散系数；f，伪扩散分数）不同组合的准确性和精密度。使用这四种方法对n = 11只仔猪的体内数据进行分析，以比较它们在检测缺血与灌注的股骨头对中灌注缺乏情况时的敏感性。
在模拟中，在所有信噪比（SNR）水平下，opAS在估计D、f、D和f×D时误差最低，尤其是在低f和D的具有挑战性的参数组合中。在仔猪模型中，所有四种IVIM拟合方法一致发现，缺血股骨头与灌注股骨头相比，D增加，f和f×D降低。然而，仅使用AS和opAS方法时，缺血股骨头与灌注股骨头相比，D*才持续降低。
opAS方法比S、OS和AS方法更准确地估计IVIM参数，特别是在低灌注组织中的灌注相关参数（D*和f）。
