Uncertainty in denoising of MRSI using low-rank methods.
Uncertainty in denoising of MRSI using low-rank methods.
复制标题
DOI：
10.1002/mrm.29018
复制
发表时间：
2022-03
影响因子：
3.3
通讯作者：
Chiew M
中科院分区：
文献类型：
--
作者：
Clarke WT;Chiew M
MRSI数据的低秩去噪导致光谱SNR的明显增加。然而，尚不清楚这是否转化为光谱拟合后代谢物浓度的不确定性较低。去噪后的真实不确定度的估计对于光谱学中的下游分析是期望的。在这项工作中，不确定性减少低秩去噪方法的基础上时空分离性和线性可预测性MRSI进行评估。提出了一种新的去噪后代谢物浓度不确定度的估计方法。自动秩阈值选择方法也评估在模拟低信噪比制度。使用质子MRSI数据的Monte Carlo模拟和在5个受试者中重复体内采集的再现性进行去噪方法的评估。在模拟和体内数据中，基于时空的去噪被证明可以降低浓度的不确定性，但线性预测去噪增加了不确定性。去噪后拟合算法提供的不确定性估计始终低估了实际代谢物的不确定性。然而，所提出的不确定性估计，基于分析表达式的条目明智的方差去噪后，是更准确的。它还表明，自动秩阈值选择使用Marchenko-Pastur分布可以偏置数据在低信噪比条件下。提出了一种替代的软阈值函数。基于时空可分性的低秩去噪方法确实降低了MRS（I）数据中的不确定性。然而，由于存在非均匀方差，因此需要进行彻底的评估，因为从残余基线噪声测量的SNR的评估是不够的。在低信噪比情况下，选择合适的秩阈值方法也很重要。
Low-rank denoising of MRSI data results in an apparent increase in spectral SNR. However, it is not clear if this translates to a lower uncertainty in metabolite concentrations after spectroscopic fitting. Estimation of the true uncertainty after denoising is desirable for downstream analysis in spectroscopy. In this work, the uncertainty reduction from low-rank denoising methods based on spatiotemporal separability and linear predictability in MRSI are assessed. A new method for estimating metabolite concentration uncertainty after denoising is proposed. Automatic rank threshold selection methods are also assessed in simulated low SNR regimes. Assessment of denoising methods is conducted using Monte Carlo simulation of proton MRSI data and by reproducibility of repeated in vivo acquisitions in 5 subjects. In simulated and in vivo data, spatiotemporal based denoising is shown to reduce the concentration uncertainty, but linear prediction denoising increases uncertainty. Uncertainty estimates provided by fitting algorithms after denoising consistently underestimate actual metabolite uncertainty. However, the proposed uncertainty estimation, based on an analytical expression for entry-wise variance after denoising, is more accurate. It is also shown automated rank threshold selection using Marchenko-Pastur distribution can bias the data in low SNR conditions. An alternative soft-thresholding function is proposed. Low-rank denoising methods based on spatiotemporal separability do reduce uncertainty in MRS(I) data. However, thorough assessment is needed as assessment by SNR measured from residual baseline noise is insufficient given the presence of non-uniform variance. It is also important to select the right rank thresholding method in low SNR cases.
登录
查看更多内容
影响因子：
4.6
作者：
Steel A;Chiew M;Jezzard P;Voets NL;Plaha P;Thomas MA;Stagg CJ;Emir UE
通讯作者：
Emir UE
DOI：
10.1109/29.1488
发表时间：
1988-01-01
期刊：
IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING
影响因子：
--
作者：
CADZOW, JA
通讯作者：
CADZOW, JA
影响因子：
3.3
作者：
Lam, Fan;Liang, Zhi-Pei
通讯作者：
Liang, Zhi-Pei
影响因子：
2.9
作者：
Chiew, Mark;Jiang, Wenwen;Emir, Uzay E.
通讯作者：
Emir, Uzay E.
影响因子：
4.6
作者：
Brender, Jeffrey R.;Kishimoto, Shun;Krishna, Murali C.
通讯作者：
Krishna, Murali C.