This study investigates a locally low-rank (LLR) denoising algorithm applied to source images from a clinical task-based functional MRI (fMRI) exam before post-processing for improving statistical confidence of task-based activation maps.
Methods
Task-based motor and language fMRI was obtained in eleven healthy volunteers under an IRB approved protocol. LLR denoising was then applied to raw complex-valued image data before fMRI processing. Activation maps generated from conventional non-denoised (control) data were compared with maps derived from LLR-denoised image data. Four board-certified neuroradiologists completed consensus assessment of activation maps; region-specific and aggregate motor and language consensus thresholds were then compared with nonparametric statistical tests. Additional evaluation included retrospective truncation of exam data without and with LLR denoising; a ROI-based analysis tracked t-statistics and temporal SNR (tSNR) as scan durations decreased. A test-retest assessment was performed; retest data were matched with initial test data and compared for one subject.
Results
fMRI activation maps generated from LLR-denoised data predominantly exhibited statistically significant (p = 4.88×10–4 to p = 0.042; one p = 0.062) increases in consensus t-statistic thresholds for motor and language activation maps. Following data truncation, LLR data showed task-specific increases in t-statistics and tSNR respectively exceeding 20 and 50% compared to control. LLR denoising enabled truncation of exam durations while preserving cluster volumes at fixed thresholds. Test-retest showed variable activation with LLR data thresholded higher in matching initial test data.
Conclusion
LLR denoising affords robust increases in t-statistics on fMRI activation maps compared to routine processing, and offers potential for reduced scan duration while preserving map quality.
OgawaSLeeTMKayAR, et al.Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci U S A1990; 87: 9868–9872.
2.
LiégeoisFCrossJHGadianDG, et al.Role of fMRI in decision-making process: Epilepsy surgery for children. J Magn Reson Imaging2006; 23: 933–940.
3.
MahdaviAAzarRShoarMH, et al.Functional MRI in clinical practice: assessment of language and motor for pre-surgical planning. Neuroradiol J2015; 28: 468–473. Epub ahead of print 2015. DOI: 10.1177/1971400915609343
4.
BlackDFVachhaBMianA, et al.American society of functional neuroradiology-recommended fMRI paradigm algorithms for presurgical language assessment. AJNR Am J Neuroradiol2017; 38: E65–E73.
5.
BlackDFLittleJTJohnsonDR. Neuroanatomical considerations in preoperative functional brain mapping. Top Magn Reson Imaging2019; 28: 213–224.
6.
VysotskiSMaduraCSwanB, et al.Preoperative FMRI associated with decreased mortality and morbidity in brain tumor patients. Interdiscip Neurosurg2018; 13: 40–45.
7.
O’ConnorEEZeffiroTA. Why is clinical fMRI in a resting state?Front Neurol2019; 10: 420.
8.
SpechtK. Current challenges in translational and clinical fMRI and future directions. Front Psychiatry2020; 10: 924.
9.
FooTKLaskarisEVermilyeaM, et al.Lightweight, compact, and high-performance 3T MR system for imaging the brain and extremities. Magn Reson Med2018; 80: 2232–2245.
10.
TanETLeeSKWeaversPT, et al.High slew-rate head-only gradient for improving distortion in echo planar imaging: Preliminary experience. J Magn Reson Imaging2016; 44: 653–664.
11.
KangDJoHJInMH, et al.The benefit of high-performance gradients on echo planar imaging for BOLD-based resting-state functional MRI. Phys Med Biol2020; 65: 235024.
12.
TrattnigSSpringerEBognerW, et al.Key clinical benefits of neuroimaging at 7 T. Neuroimage2018; 168: 477–489.
13.
KaramatMIDarvish-MollaSSantos-DiazA. Opportunities and challenges of 7 tesla magnetic resonance imaging: A review. Crit Rev Biomed Eng2016; 44: 73–89.
14.
HoffMNMcKinneyAShellockFG, et al.Safety considerations of 7-T MRI in clinical practice. Radiology2019; 292: 509–518.
15.
LarkmanDJHajnalJV.HerlihyAH, et al.Use of multicoil arrays for separation of signal from multiple slices simultaneously excited. J Magn Reson Imaging2001; 13: 313–317.
16.
NunesRGHajnalJVGolayX, et al.Simultaneous slice excitation and reconstruction for single shot EPI. In: Proc Intl Soc Mag Reson Med2006; 293.
17.
MoellerSYacoubEOlmanCA, et al.Multiband multislice GE-EPI at 7 tesla, with 16-fold acceleration using partial parallel imaging with application to high spatial and temporal whole-brain FMRI. Magn Reson Med2010; 63: 1144–1153.
18.
SetsompopKGagoskiBAPolimeniJR, et al.Blipped-controlled aliasing in parallel imaging for simultaneous multislice echo planar imaging with reduced g-factor penalty. Magn Reson Med2012; 67: 1210–1224.
MarkITBlackDFDeLoneDR, et al.Higher temporal resolution multiband fMRI provides improved presurgical language maps. Neuroradiology2021; 63: 439–445.
21.
SeidelPLevineSMTahedlM, et al.Temporal Signal-to-Noise Changes in Combined Multislice- and In-Plane-Accelerated Echo-Planar Imaging with a 20- and 64-Channel Coil. Sci Rep2021; 10: 5536.
Aja-FernándezSTristán-VegaAHogeWS. Statistical noise analysis in GRAPPA using a parametrized noncentral Chi approximation model. Magn Reson Med2011; 65: 1195–1206.
24.
GudbjartssonHPatzS. The Rician distribution of noisy MRI data. Magn Reson Med1995; 34: 910–914.
25.
CoxRW. AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages. Comput Biomed Res1996; 29: 162–173.
26.
JoHJLeeJMKimJH, et al.Spatial accuracy of fMRI activation influenced by volume- and surface-based spatial smoothing techniques. Neuroimage2007; 34: 550–564.
27.
JoHJSaadZSSimmonsWK, et al.Mapping sources of correlation in resting state FMRI, with artifact detection and removal. Neuroimage2010; 52: 571–582.
28.
JoHJLeeJMKimJH, et al.Artificial shifting of fMRI activation localized by volume- and surface-based analyses. Neuroimage2008; 40: 1077–1089.
29.
WhiteTO’LearyDMagnottaV, et al.Anatomic and functional variability: The effects of filter size in group fMRI data analysis. Neuroimage2001; 13: 577–588.
30.
YueYLohJMLindquistMA. Adaptive spatial smoothing of fMRI images. Stat Interface2010; 3: 3–13.
31.
PurdonPLWeisskoffRM. Effect of temporal autocorrelation due to physiological noise and stimulus paradigm on voxel-level false-positive rates in fMRI. Hum Brain Mapp1998; 6: 239–249.
32.
SkudlarskiPConstableRTGoreJC. ROC analysis of statistical methods used in functional MRI: Individual subjects. Neuroimage1999; 9: 311–329.
33.
FristonKJJosephsOZarahnE, et al.To smooth or not to smooth? Bias and efficiency in fMRI time-series analysis. Neuroimage2000; 12: 196–208.
34.
PurdonPLSoloVWeisskoffRM, et al.Locally regularized spatiotemporal modeling and model comparison for functional MRI. Neuroimage2001; 14: 912–923.
ChiewMSmithSMKoopmansPJ, et al.k-t FASTER: Acceleration of functional MRI data acquisition using low rank constraints. Magn Reson Med2015; 74: 363–364.
37.
AggarwalPGuptaA. Accelerated fMRI reconstruction using Matrix Completion with Sparse Recovery via Split Bregman. Neurocomputing2016; 216: 319–330.
38.
AggarwalPShrivastavaPKabraT, et al.Optshrink LR + S: accelerated fMRI reconstruction using non-convex optimal singular value shrinkage. Brain Inform2017; 4: 65–83.
39.
WeizmanLMillerKLEldarYC, et al.PEAR: PEriodic and fixed Rank separation for fast fMRI. Med Phys2017; 44: 6166–6182.
40.
ChiewMGraedelNNMillerKL. Recovering task fMRI signals from highly under-sampled data with low-rank and temporal subspace constraints. Neuroimage2018; 174: 97–110.
41.
HaldarJPLiangZP. Spatiotemporal imaging with partially separable functions: A matrix recovery approach. In:Proc 2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, 2010, 716–719.
42.
TrzaskoJDManducaA. Local versus global low-rank promotion in dynamic MRI series reconstruction. In: Proc Intl Soc Mag Reson Med2011; 4371.
CandésEJSing-LongCATrzaskoJD. Unbiased risk estimates for singular value thresholding and spectral estimators. IEEE Trans Signal Process2013; 61: 4643–4657.
45.
ZhangTPaulyJMLevesqueIR. Accelerating parameter mapping with a locally low rank constraint. Magn Reson Med2015; 73: 655–661.
46.
MoellerSWeingärtnerSAkçakayaM. Multi-scale locally low-rank noise reduction for high-resolution dynamic quantitative cardiac MRI. In: Conf Proc IEEE Eng Med Biol Soc2017: 1473–1476.
47.
LugauerFNickelDWetzlJ, et al.Accelerating multi-echo water-fat MRI with a joint locally low-rank and spatial sparsity-promoting reconstruction. Magn Reson Mater Phy2017; 30: 189–202.
48.
MiaoXLingalaSGGuoY, et al.Accelerated cardiac cine MRI using locally low rank and finite difference constraints. Magn Reson Imaging2016; 34: 707–714.
49.
YamanBWeingärtnerSKargasN, et al.Locally low-rank tensor regularization for high-resolution quantitative dynamic MRI. In: 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), 2017. 1–5.
50.
TamirJIUeckerMChenW, et al.T2 shuffling: Sharp, multicontrast, volumetric fast spin-echo imaging. Magn Reson Med2017; 77: 180–195.
51.
Lima da CruzGBustinAJaubertO, et al.Sparsity and locally low rank regularization for MR fingerprinting. Magn Reson Med2019; 81: 3530–3543.
52.
MilshteynEvon MorzeCReedGD, et al.Using a local low rank plus sparse reconstruction to accelerate dynamic hyperpolarized 13C imaging using the bSSFP sequence. J Magn Reson2018; 290: 46–59.
53.
Caballero-GaudesCReynoldsRC. Methods for cleaning the BOLD fMRI signal. Neuroimage2017; 154: 128–149.
MeyerNKCampeauNGBlackDF, et al.Locally low-rank denoising preserves statistical confidence in task-based functional activation under scan duration reduction. In: Proc Intl Soc Mag Reson Med2021; 2703.
56.
MeyerNKCampeauNGBlackDF, et al.Initial clinical evaluation of locally low-rank denoising on motor areas for task-based presurgical functional MRI. In: Proc Intl Soc Mag Reson Med2021; 2704.
57.
InMHShuYTrzaskoJD, et al.Reducing PNS with minimal performance penalties via simple pulse sequence modifications on a high-performance compact 3T scanner. Phys Med Biol2020; 65: 15NT02.
58.
Brant-ZawadzkiMGillanGDNitzWR, MP RAGE: a three-dimensional, T1-weighted, gradient-echo sequence--initial experience in the brain. Radiology1992; 182: 769–775.
59.
Prism Clinical Imaging, Elm Grove, WI. RRID:SCR_016977.
60.
CaiJFCandèsEJShenZ. A singular value thresholding algorithm for matrix completion. SIAM J Optim2010; 20: 1956–1982.
KerbyDS. The simple difference formula: An approach to teaching nonparametric correlation Comprehensive Psychology. 2014. 3: 1.
64.
DiceLR. Measures of the Amount of Ecologic Association Between Species. Ecology1945; 26: 297–302.
65.
MorrisonMAChurchillNWCusimanoMD, et al.Reliability of task-based fMRI for preoperative planning: A test-retest study in brain tumor patients and healthy controls. PLoS One2016; 11: e0149547.
66.
BenjaminCFWalshawPDHaleK, et al.Presurgical language fMRI: Mapping of six critical regions. Hum Brain Mapp2017; 38: 4239–4255.
67.
MurphyKBodurkaJBandettiniPA. How long to scan? The relationship between fMRI temporal signal to noise ratio and necessary scan duration. Neuroimage2007; 34: 565–574.
68.
ElliottMLKnodtARIrelandD, et al.What Is the Test-Retest Reliability of Common Task-Functional MRI Measures? New Empirical Evidence and a Meta-Analysis. Psychol Sci2020; 31: 792–806.
69.
StippichCBlatowMGarcia AlzamoraM. Task-Based Presurgical Functional MRI in Patients with Brain Tumors. In: Clinical Functional MRI: Presurgical Functional Neuroimaging. Springer, 2015, pp. 89–141.
70.
PetrellaJRShahLMHarrisKM, et al.Preoperative functional MR imaging localization of language and motor areas: Effect on therapeutic decision making in patients with potentially resectable brain tumors. Radiology2006; 240: 793–802.
71.
RuttenGJMRamseyNFVan RijenPC, et al.Development of a functional magnetic resonance imaging protocol for intraoperative localization of critical temporoparietal language areas. Ann Neurol2002; 51: 350–360.
72.
BeisteinerRRobinsonSWurnigM, et al.Clinical fMRI: Evidence for a 7T benefit over 3T. Neuroimage2011; 57: 1015–1021.
73.
NadkarniTNAndreoliMJNairVA, et al.Usage of fMRI for pre-surgical planning in brain tumor and vascular lesion patients: Task and statistical threshold effects on language lateralization. Neuroimage Clin2015; 7: 415–423.
74.
VansteenselMJHermesDAarnoutseEJ, et al.Brain-computer interfacing based on cognitive control. Ann Neurol2010; 67: 809–816.
75.
SieroJCWHermesDHoogduinH, et al.BOLD matches neuronal activity at the mm scale: A combined 7T fMRI and ECoG study in human sensorimotor cortex. Neuroimage2014; 101: 177–184.
76.
VoyvodicJT. Activation mapping as a percentage of local excitation: fMRI stability within scans, between scans and across field strengths. Magn Reson Imaging2006; 24: 1249–1261.
77.
VizioliLMoellerSDowdleL, et al.Lowering the thermal noise barrier in functional brain mapping with magnetic resonance imaging. Nat Commun2021; 12: 1–15.
78.
Ades-AronBLemberskiyGVeraartJ, et al.Improved Task-based Functional MRI Language Mapping in Patients with Brain Tumors through Marchenko-Pastur Principal Component Analysis Denoising. Radiology2021; 298: 365–373.
79.
TrzaskoJDManducaA. A unified tensor regression framework for calibrationless dynamic, multi-channel MRI reconstruction. In: Proc Intl Soc Mag Reson Med2013; 603.
80.
GuoSFesslerJANollDC. High-Resolution Oscillating Steady-State fMRI using Patch-Tensor Low-Rank Reconstruction. IEEE Trans Med Imaging2020; 39: 4357–4368.
81.
HoogenraadFGCReichenbachJRHaackeEM, et al.In vivo measurement of changes in venous blood-oxygenation with high resolution functional MRI at 0.95 Tesla by measuring changes in susceptibility and velocity. Magn Reson Med1998; 39: 97–107.
82.
RoweDB. Modeling both the magnitude and phase of complex-valued fMRI data. Neuroimage2005; 25: 1310–1324.
83.
ArjaSKFengZChenZ, et al.Changes in fMRI magnitude data and phase data observed in block-design and event-related tasks. Neuroimage2010; 49: 3149–3160.
84.
ChenZLiuJCalhounVD. Susceptibility-based functional brain mapping by 3D deconvolution of an MR-phase activation map. J Neurosci Methods2013; 216: 33–42.
85.
ChenZCalhounV. T2∗ phase imaging and processing for brain functional magnetic susceptibility (χ) mapping. Biomed Phys Eng Express2016; 2: 025015.
86.
ChenZCaprihanADamarajuE, et al.Functional brain connectivity in resting-state fMRI using phase and magnitude data. J Neurosci Methods2018; 293: 299–309.
87.
ChenZFuZCalhounV. Phase fMRI Reveals More Sparseness and Balance of Rest Brain Functional Connectivity Than Magnitude fMRI. Front Neurosci2019; 13: 204.
88.
NanFYNowakRD, Generalized likelihood ratio detection for fMRI using complex data. IEEE Trans Med Imaging1999; 18: 320–329.
89.
CalhounVDAdaliTPearlsonGD, et al.Independent component analysis of fMRI data in the complex domain. Magn Reson Med2002; 48: 180–192.
90.
RodriguezPAdaliTLiH, et al.Phase correction and denoising for ICA of complex FMRI data. In: 2010 IEEE International Conference on Acoustics, Speech and Signal Processing, 2010. 497–500.
YuMCLinQHKuangLD, et al.ICA of full complex-valued fMRI data using phase information of spatial maps. J Neurosci Methods2015; 249: 75–91.
93.
YuCHPradoROmbaoH, et al.A bayesian variable selection approach yields improved detection of brain activation from complex-valued fMRI. J Am Stat Assoc2018; 113. Epub ahead of print 2018. DOI: 10.1080/01621459.2018.1476244
94.
AdrianDWMaitraRRoweDB. Complex-valued time series modeling for improved activation detection in fMRI studies. Ann Appl Stat2018; 12: 1451–1478.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
