Motion-Robust Direct Myelin Imaging in MRI using Self-Gating

Motion-Robust Direct Myelin Imaging in MRI using Self-Gating | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Motion-Robust Direct Myelin Imaging in MRI using Self-Gating Jinil Park, Sam Sedaghat, Kader Karli Oguz, Youngkyoo Jung, Eddie Fu, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7199927/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 09 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted 10 You are reading this latest preprint version Abstract Direct myelin imaging with inversion-recovery ultrashort-echo-time (IR-UTE) is highly motion-sensitive, yet extra hardware or longer scans are impractical. We evaluated whether a superior–inferior (SI) self-navigator with bit-reversed spoke-angles mitigates motion artifacts without extending acquisition. Dual-echo IR-UTE was implemented at 3T. After an adiabatic inversion pulse, 21 spokes were acquired per segment around the white-matter null point, and dual-echo subtraction suppressed residual long-T2 signals. Adding SI-navigators at the end of each segment allows motion detection without lengthening TR. And bit-reversal is used to pseudo-randomize the spoke-angles. Bloch simulations of a 2D synthetic brain removed 10% of spokes to mimic gating. Three volunteers were scanned: (i) sequential-ordering, no-motion; (ii) bit-reversed, no-motion; (iii) bit-reversed with deliberate head nods. The navigator rejected 1,280 of 12,000 spokes (10.7%) during nodding, and the same gating was reconstructed on motion-free data. Simulation showed coherent streaks for sequential ordering but an isotropic point-spread function for bit-reversal. In vivo, gating degraded only the sequential dataset. bit-reversal preserved subcortical and deep-white-matter detail. During intentional nodding, gating with bit-reversal enhanced myelin contrast, outperforming the image reconstructed from the ungated data. The SI-navigator plus bit-reversal enables effective motion gating without hardware or time penalty, supporting routine motion-robust myelin mapping. Biological sciences/Neuroscience Physical sciences/Physics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Myelin is a dense layer of cell membrane that wraps around axons, produced by non-neuronal cells known as oligodendrocytes 1,2 . Structurally, it consists of a multilayered membrane with alternating lipid and protein layers, forming an insulating sheath around nerve fibers in both the white and gray matter of the brain, as well as in the spinal cord. This insulation boosts the speed of action potential conduction by approximately 100 times and reduces the refractory period by more than 30 times 3 . Collectively, these enhancements amplify the brain's capacity for information processing by over 3,000 times, making myelin essential for normal neurological functions such as speech, coordination, and cognition 3–5 . When axons lose their myelin, they also lose saltatory conduction, mechanical protection, and metabolic support, which can lead to structural and functional damage 6–8 . Myelin fundamentally transforms how electrical signals are initiated and propagated in the nervous system, and its disruption plays a key role in the pathology of numerous neurological disorders, including multiple sclerosis (MS) 9 , traumatic brain injury (TBI) 10,11 , epilepsy 12 , and Alzheimer's disease (AD) 13–15 . Direct assessment of myelin integrity in both white and gray matter may be important for diagnosing and evaluating the prognosis of myelin-related disorders. However, the hydrogens ( 1 H, the source of signal in MRI) in myelin have extremely short T2 relaxation times (T2* < ~0.3 ms at 3T) and therefore cannot be directly imaged using conventional MRI sequences, which typically have echo times (TEs) of several milliseconds or longer 16–23 . As a result, conventional sequences provide only an indirect or qualitative assessment of myelin. Another challenge in directly imaging myelin is selectivity, where long T2 water is present in much higher concentrations in the brain and produces a signal that is approximately 10–20 times stronger than that of myelin 24 . Consequently, myelin is invisible on conventional MRI. Alternatively, indirect imaging techniques have been explored, including myelin water fraction (MWF) mapping 25,26 , magnetization transfer (MT) imaging 27 , and quantitative susceptibility mapping (QSM) 28 . Recently, ultrashort echo time (UTE) sequences have been investigated to directly capture the short-lived myelin signal. To suppress long T2 signal and hence improve signal contrast from myelin, adiabatic inversion pulses have been utilized to provide robust and uniform suppression of long T2 water signals 29–31 . Combined with the echo subtraction technique, inversion recovery prepared UTE (IR-UTE) can effectively detect myelin signal directly with high signal contrast in the brain 31,32 . The IR-UTE technique has been validated in both TBI and MS, showing promise in detecting changes in myelin content. For example, Ma et al. demonstrated that IR-UTE enables quantitative myelin imaging in the mouse brain using a preclinical 3T MRI scanner, allowing detection of white matter demyelination following open-field low-intensity blast injury (a model of mild TBI) 33 . Jang et al. reported that the myelin signal detected by IR-UTE significantly correlates with disability in MS patients, as measured by the Expanded Disability Status Scale (EDSS) 34 . These findings suggest that IR-UTE may serve as a potential imaging-based biomarker for diseases involving myelin alterations. Despite promising results, a key limitation of the current IR-UTE technique is its long scan time (~ 10 minutes). This extended duration is primarily due to the inversion recovery imaging scheme, which requires a long repetition time (TR) of approximately 1 second. Additionally, since myelin imaging typically yields low signal intensity, a longer scan time is often necessary to improve the signal-to-noise ratio (SNR). However, prolonged imaging increases the risk of motion-related artifacts, such as blurriness, ghosting, and streaking, caused by intra-scan head movement. These artifacts can significantly degrade the already low-SNR myelin images. The issue is further exacerbated in patients with neurological disorders, who often have impaired motor control and may struggle to remain still during lengthy scans. In MRI, various methods have been developed to prevent or correct motion-related artifacts. Prospective motion gating typically requires specialized hardware, such as a respiratory bellows, an electrocardiogram gating device, or a motion-tracking camera 35,36 . Retrospective motion gating often involves additional data acquisition to capture motion-related information, commonly referred to as a navigator signal 37,38 . In this study, we propose a self-gated IR-UTE technique incorporating bit-reversed view ordering to correct for motion-related artifacts and enhance direct myelin imaging, without requiring additional scan time. Methods IR-UTE with dual-echo subtraction Myelin has a short T2* relaxation time and low proton density (i.e., density of 1 H) 23,24 , resulting in an extremely low “detectable” MR signal. As a result, it is sensitive to noise, aliasing or streak artifacts, ghosting, and image reconstruction errors, any of which can degrade myelin delineation. In addition, the large dynamic-range gap between myelin and neighboring structures increases the risk of quantization errors that compromise signal. To overcome these limitations, a method that combines an IR preparation pulse with UTE dual-echo subtraction has recently been proposed 31 . In this approach, an adiabatic inversion pulse is applied, followed by acquisition of two images at different TEs during the inversion time (TI) (Figures 1A, B). The TI is selected so that the long-T2 white matter (WM L ) is nulled (Figure 1c). Because myelin’s T2* is extremely short, a long adiabatic inversion pulse cannot invert the magnetization; instead, it drives it to saturation (M z ≈ 0) 39 . Immediately thereafter, the saturated myelin recovers rapidly owing to its short T1. At the chosen TI, the first echo captures signals of myelin, long-T2 grey matter (GM L ), and any remaining long-T2 components, whereas the second echo is acquired after the myelin signal has substantially decayed (Figure 1D). Subtracting the two echoes therefore isolates the myelin-specific signal. Because the inversion recovery requires a long TR, acquiring one spoke per IR pulse significantly increases the scan time. To address this problem, the sequence acquires multiple k-space spokes segmented around the white matter null point, increasing acquisition efficiency (Figure 1A). Self-navigating projection This section describes our approach to motion tracking using self-navigating. Due to the oversampled center of k-space, non-Cartesian radial trajectories are known to be intrinsically more robust to motion than Cartesian trajectories. However, the motion can still affect the radial imaging, with unwanted image blurring, streaking, and ghosting. Therefore, identifying motion during scanning and discarding the motion-corrupted data can enhance the image quality. The need becomes even more critical when the subject makes an unexpected, bulk movement (e.g., patients with impaired motor function), after which the head may settle in a different position and appear severe artifacts. It is thus preferable to measure motion during the scan and address it retrospectively. In the proposed approach, we monitor head motion using a superior-inferior (SI) projection navigator. The navigator is inserted at regular intervals to the end of each segment (Figure 2A). Placing the navigator at the end of segment maintains a consistent TI, ensuring stable contrast in motion tracking signal. Widely used in lung imaging, this navigator periodically applies a short readout gradient along the G z axis, acquires a half k-space line, and then performs a 1D Fourier transform to generate a SI projection profile (Figures 2B, C) 38,40 . Continuous comparison of the profiles enables tracking movement of head throughout the IR-UTE acquisition (Figure 2D). Figure 2E shows an example of motion tracking using the self-navigator, estimated from the signal intensity at a fixed pixel location in the 1D profile, defined by the position of maximum signal at the first time point. In this example, the subject intentionally nodded their head every 20 seconds during the scan. When the head nods, the projection profile changes; this variation can be measured and incorporated into the image reconstruction process to reduce motion artifacts by excluding motion-contaminated data. Bit-reversed view ordering In radial trajectories, it is important for retrospective gating to maintain that the remaining spokes are uniformly distributed in k-space after discarding unwanted data. If the spokes are acquired sequentially, the data loss caused by gating will be concentrated in a specific region of k-space, resulting in an undersampling artifact. Various sampling strategies such as golden-angle, phyllotaxis, and interleaved schemes have been proposed to avoid this problem 38,41,42 . Here we adopt a bit-reversed view ordering. By reversing the binary index of each spoke, the original trajectory is rearranged into a pseudo-random sequence. For example, the 52 nd spoke (binary 110100) becomes the 13 th spoke (binary 001011) after the reordering. The main advantage is that it can be applied to any kind of trajectory, providing a pseudo-random view ordering. Simulation A computer simulation was performed to test whether bit‑reversed view ordering can compensate for the k‑space non‑uniformity introduced by retrospective motion gating in IR‑UTE myelin imaging. A 2D high-resolution digital head phantom (654 × 654 pixels) consisting of six tissue components—cerebrospinal fluid (CSF), GM L , WM L , myelin within white matter, skull bone, and subcutaneous fat—was created, as illustrated in Figure 3A. To construct the segmentation map, a sample brain map from the open-source segmentation tool, multiplicative intrinsic component optimization (MICO) 43 , was used, which provided labels for white and gray matters. The remaining three components (CSF, skull, and subcutaneous fat) were manually added. Four demyelinated lesions were placed within the white matter region. Using the segmentation map, a Bloch simulation was performed based on the tissue parameters shown in Figure 3B. All simulations were conducted in MATLAB R2024a (MathWorks, Natick, MA) using in-house code that solves the Bloch equations analytically to model signal evolution. The simulation reproduced an IR‑UTE dual‑echo acquisition, using the following sequence parameters: flip angle = 15°, TR = 1000 ms, inversion time TI = 379.9 ms (null point of WM L ), readout bandwidth = 250 kHz, and field of view = 220 × 220 mm 2 , matrix size = 220 × 220. Two radial echoes were collected per spoke (TE 1 = 0.03 ms, TE 2 = 3 ms). A total of 692 spokes covered 2D circular k-space, delivered in segments of 21 spokes; the spoke‑to‑spoke interval was 6 ms. To assess sampling robustness, k‑space was traversed either sequentially or with bit‑reversed ordering. The data loss due to motion gating was assumed to be 10% of the total spokes. In vivo study Three healthy male volunteers (42, 42, and 38-year-old) were scanned on a 3T clinical MRI system (MAGNETOM Prisma Fit, Siemens Healthineers, Erlangen, Germany) using a 32‑channel head/neck coil. A written informed consent was collected under an institutional‑review‑board approval (University of California, Davis). All procedures were performed in accordance with the Declaration of Helsinki and relevant institutional guidelines and regulations. Three scans were performed using 3D IR-UTE sequence: Sequential view ordering without motion (Scan #1), bit-reversed view ordering without motion(Scan #2), and bit-reversed view ordering with intentional motion (Scan #3). During Scan #3, the participant was instructed to nod their head once with a large movement. Scans #1 and #2 served as motion-free references, acquired using two different view ordering strategies, allowing for a controlled comparison of their effects under retrospective undersampling. In Scan #3, motion-corrupted data were identified and excluded using a self-gating technique. The actual motion profile obtained in Scan #3 was then reused to simulate gating-induced undersampling effects in Scans #1 and #2. The imaging parameters were: flip angle = 10°, TR = 1000 ms, TI = 327 ms (white matter null point tuned at 3T), dual echo readout (TE 1 = 0.03 ms, TE 2 = 3 ms), readout bandwidth = 92 kHz, field of view = 220 × 220 × 176 mm 3 , matrix size = 220 × 220 × 88, 20 spokes per segment, total spokes = 12,000, and spoke to spoke interval = 5 ms. A 50 µs nonselective hard pulse excited each spoke, and a 10.24 ms hyperbolic secant inversion pulse preceded every segment. Results Simulation Bit-reversed view ordering significantly reduced artifacts caused by retrospective motion gating (Fig. 3 C). When 10% of spokes were removed using sequential view ordering, prominent streaks appeared along the spoke directions, accompanied by blurring in myelin-rich regions (red arrows). In contrast, bit-reversed view ordering distributed the data loss more evenly, leading to a noticeable reduction in streaking artifacts. This improvement was also evident in the simulated point spread function (PSF): sequential view ordering resulted in pronounced side-lobes (yellow In vivo study – Self navigator and gating Figure 4 A plots the navigator signal from Scan #3. The changed baseline reflects the changed head position due to intentional head nodding. The data with the navigator signal changed by nodding was excluded using threshold-based gating and then used for reconstruction to effectively suppress motion artifacts. IR-UTE acquired 12,000 spokes and rejected 1,280 (10.67%) of them, which were head position data changed by nodding. The spokes where head position data changed by nodding are removed (black dots) are continuously missing in sequential view ordering (Fig. 4 B), while bit-reversed view ordering scatters the missing segments pseudo randomly throughout the sphere (Fig. 4 C). In vivo study – Myelin imaging For all three volunteers, the self-gating approach significantly improved the quality of myelin imaging in the presence of intentional head motion. Figure 5 contrasts the myelin images from the two motion-free reference data (Scan #1 and Scan #2) obtained with sequential view ordering or bit-reversed view ordering, which were undersampled using a simulated motion-gating trace (obtained from Scan #3). This demonstrates how sequential and bit-reversed view orderings influence the myelin imaging, by comparing the images from ungated (fully sampled) and gated (undersampled) data. With sequential view ordering, the simulated gating produced focal signal loss in myelin (white arrows in Figs. 5 C and 5 G). In the corresponding difference image between ungated and gated myelin images (Fig. 5 D), the error appears as coherent signal change across the brain. In contrast, bit-reversed view ordering improves the myelin image in both subcortical and deep-white-matter regions (white arrows in Figs. 5 J and 5 N). The corresponding difference image exhibits incoherent low-level noise, underscoring the scheme’s superior robustness (Fig. 5 K). Figure 6 shows the myelin imaging with the intentional head motion (Scan #3), where self-gating and bit-reversed view ordering were utilized. Figures 6 A- 6 C were reconstructed from all acquired spokes without motion gating. Although the full data typically allows higher SNR than the undersampled data, the head nods may introduce blurring and ghost artifacts, impairing overall image quality and myelin contrast. Figures 6 D- 6 F show the result from the same data after retrospective motion gating. Spokes coinciding with nodding motion, detected by self-gating, were discarded before reconstruction. Despite the reduced number of data sampling, sharper myelin boundaries and unbiased myelin signal were achieved by omitting the motion-contaminated data (white arrows). Discussion Myelin has extremely short T2s (< 1 ms), due to its highly ordered microstructure and chemical shift with multiple lipid peaks. Therefore myelin cannot be directly imaged with conventional MRI sequences, which typically have TEs of several milliseconds or longer 16–23 . As a result, conventional sequences only provide an indirect assessment of myelin. The MR properties of myelin have been investigated by several groups. Broad line proton spectroscopy studies have demonstrated that myelin is in a liquid-crystalline state 16 . Multi-component analysis of spin echo or free induction decay of white matter samples has shown a broad range of short T2 or T2* values for myelin protons (e.g., ~ 50 µs 44 , 7.5–101 µs 45 , 50-1000 µs 18 , 150–250 µs 24 , etc.). UTE MRI sequences with TEs as short as ~ 8–32 µs, which are 100–1000 times shorter than those of conventional clinical sequences, allow direct detection of signal from myelin 46–48 . Because the myelin signal is significantly lower than that of long T2 tissues in the brain, direct myelin imaging techniques rely on IR preparation and echo subtraction to suppress the long T2 signal. Despite the effectiveness of this suppression, the inherently low SNR of the myelin lipid remains a major challenge. In this context, even minor imaging artifacts can severely impact the detection of myelin. Unlike conventional Cartesian MRI, where motion typically results in ghosting artifacts along the phase-encoding direction, UTE imaging, based on a radial trajectory, produces motion artifacts that radiate in all directions. As a result, strong short T2 signals from surrounding bone or fat, as well as residual long T2 signals, can interfere with the detection of the myelin signal if motion occurs during the scan. In this study, we demonstrated the efficacy of motion-gating using self-navigator and bit-reversed view ordering in IR-UTE for improved myelin imaging. The self-navigator does not impose any additional scan time because it is located at the end of segments (Fig. 2 A), making it clinically feasible. Our findings show that combining SI self-navigator with a bit-reversed radial trajectory markedly improves the motion robustness of IR-UTE myelin imaging. In the motion-free volunteer, both ordering schemes scanned, when the navigator-derived gating trace was applied retrospectively the sequential data developed focal signal decrease, whereas the bit-reversed view ordering data retained cortical detail and myelin contrast (Fig. 5 ). When intentionally nodded, about 10% of the spokes were removed by the navigator. nevertheless, the gated, bit-reversed view ordering reconstruction recovered sharp enhanced grey/white contrast, outperforming the ungated image that had the full data set (Fig. 6 ). These results underscore two practical points: (i) dispersing missing spokes pseudo-randomly with bit-reversal converts coherent streaks into low-energy noise, and (ii) placing the SI navigator at each segment boundary permits effective gating without extending scan time. Because both modifications require only a gradient-table permutation and a single 1D readout, they can be deployed on existing clinical scanners and may enable reliable, motion-tolerant myelin mapping in routine neuroimaging. Bit-reversing is a simple yet powerful method for achieving pseudo-randomized view ordering 49 . In IR-prepared imaging, this approach can improve image quality, particularly when a large number of spokes are acquired per inversion recovery, by redistributing T1 blurring effects in a pseudo-random manner. For example, multiple spokes may be acquired near the nulling point of long T2 white matter, with roughly half acquired before the null point and half after. Spokes acquired before the null point capture negative longitudinal magnetization (M z ), while those acquired after the null point capture positive M z from the long T2 white matter. Each spoke samples a different M z magnitude, governed by the TI and T1 relaxation. Consequently, the encoded k-space can exhibit structured modulation patterns as a function of both T1 recovery and view ordering, which may degrade the PSF. Previous studies have explored variable flip angle excitation to compensate for this kind of signal variation 50,51 . However, determining optimal flip angles based on the long T2 white matter signal is not possible in the direct myelin imaging, as the target signal is ideally zero at the nulling point. As an alternative, bit-reversed view ordering randomizes these modulation patterns, making them more 'noise-like' and thereby reducing coherent artifacts. Further in-depth investigation of this approach will be pursued in future studies. This study has several limitations. First, only three healthy volunteers were included to validate the proposed approach. Although the volunteers were instructed to make intentional motions during the scan, this may not accurately represent the conditions seen in patient groups with various neurological disorders. We plan to test this sequence in patients with a range of neurological diseases in future studies. Second, we did not compare our method with alternative view ordering schemes, such as the golden angle method 52 . While the golden angle approach has demonstrated effectiveness in radial undersampled and time-resolved imaging, our goal is to distribute motion-contaminated spokes in a pseudo-random manner, which differs from the objectives targeted by golden angle techniques. Third, we only incorporated an SI navigator, which is capable of detecting nodding motion. Head motion in other directions, such as shaking, can be detected by adding additional navigators along the G x or G y axes, without increasing scan time. Conclusion This study demonstrates the efficacy of motion gating using self-navigator and bit-reversed view ordering in IR-UTE for motion-robust myelin imaging. Bit-reversed view ordering is a one-line permutation of the gradient list that pseudo-randomizes spoke positions and keeps k-space nearly uniform even after data are rejected. The proposed approach demands no additional hardware, no extra scan time, but only minor modification on sequence, and hence it can be readily implemented on current clinical 3T systems. Combined with direct myelin imaging technique, the proposed approach could benefit patient groups prone to motion such as children, elderly, or epilepsy cohorts. Declarations Funding Research reported in this publication was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (R01AR078877) and the National Institute on Aging (P30AG072972) of the National Institutes of Health (NIH). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. This research was also supported by the Deutsche Forschungsgemeinschaft (DFG, SE 3272/1-1) and University Heidelberg Project Expanding Internationality. Author contributions This paper was co-authored by Jinil Park, Sam Sedaghat, Kader Karli Oguz, Youngkyoo Jung, Eddie Fu, Nian Wang, Fang Liu and Hyungseok Jang, with me serving as the first author and Dr. Hyungseok Jang as the corresponding author. All authors contributed substantially to the conception, design, and data acquisition or interpretation, as well as to the drafting and revision of this manuscript. Each author has reviewed and approved the final version. Data availability The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request. References Fields, R. D. Myelin—More than Insulation. Science 344 , 264–266 (2014). Jahn, O., Tenzer, S. & Werner, H. B. Myelin Proteomics: Molecular Anatomy of an Insulating Sheath. Mol. Neurobiol. 40 , 55–72 (2009). Lu, P. H. et al. Myelin breakdown mediates age-related slowing in cognitive processing speed in healthy elderly men. Brain Cogn. 81 , 131–138 (2013). Waxman, S. G. Conduction in Myelinated, Unmyelinated, and Demyelinated Fibers. Arch. Neurol. 34 , 585–589 (1977). Felts, P. A., Baker, T. A. & Smith, K. J. Conduction in Segmentally Demyelinated Mammalian Central Axons. J. Neurosci. 17 , 7267–7277 (1997). Lee, Y. et al. Oligodendroglia metabolically support axons and contribute to neurodegeneration. Nature 487 , 443–448 (2012). Chabas, D. et al. The Influence of the Proinflammatory Cytokine, Osteopontin, on Autoimmune Demyelinating Disease. Fünfschilling, U. et al. Glycolytic oligodendrocytes maintain myelin and long-term axonal integrity. Nature 485 , 517–521 (2012). Noseworthy, J. H., Rodriguez, M. & Weinshenker, B. G. Multiple Sclerosis. N. Engl. J. Med. (2000). Flygt, J., Djupsjö, A., Lenne, F. & Marklund, N. Myelin loss and oligodendrocyte pathology in white matter tracts following traumatic brain injury in the rat. Eur. J. Neurosci. 38 , 2153–2165 (2013). Wright, A. D. et al. Myelin Water Fraction Is Transiently Reduced after a Single Mild Traumatic Brain Injury – A Prospective Cohort Study in Collegiate Hockey Players. PLOS ONE 11 , e0150215 (2016). Drenthen, G. S. et al. On the merits of non-invasive myelin imaging in epilepsy, a literature review. J. Neurosci. Methods 338 , 108687 (2020). Kavroulakis, E. et al. Myelin content changes in probable Alzheimer’s disease and mild cognitive impairment: Associations with age and severity of neuropsychiatric impairment. J. Magn. Reson. Imaging 47 , 1359–1372 (2018). Dean, D. C. et al. Association of Amyloid Pathology With Myelin Alteration in Preclinical Alzheimer Disease. JAMA Neurol. 74 , 41 (2017). Haroutunian, V. et al. Myelination, oligodendrocytes, and serious mental illness. Glia 62 , 1856–1877 (2014). Lecar, H., Ehrenstein, G. & Stillman, I. Detection of Molecular Motion in Lyophilized Myelin by Nuclear Magnetic Resonance. Biophys. J. 11 , 140–145 (1971). Waldman, A. et al. MRI of the brain with ultra-short echo-time pulse sequences. Neuroradiology 45 , 887–892 (2003). Horch, R. A., Gore, J. C. & Does, M. D. Origins of the ultrashort‐T2 1 H NMR signals in myelinated nerve: A direct measure of myelin content? Magn. Reson. Med. 66 , 24–31 (2011). Wilhelm, M. J. et al. Direct magnetic resonance detection of myelin and prospects for quantitative imaging of myelin density. Proc. Natl. Acad. Sci. 109 , 9605–9610 (2012). Du, J. et al. Ultrashort echo time (UTE) magnetic resonance imaging of the short T2 components in white matter of the brain using a clinical 3T scanner. NeuroImage 87 , 32–41 (2014). Du, J. et al. Measurement of T1 of the Ultrashort T2* Components in White Matter of the Brain at 3T. PLoS ONE 9 , e103296 (2014). He, Q. et al. Direct magnitude and phase imaging of myelin using ultrashort echo time (UTE) pulse sequences: A feasibility study. Magn. Reson. Imaging 39 , 194–199 (2017). Fan, S., Ma, Y., Chang, E. Y., Bydder, G. M. & Du, J. Inversion recovery ultrashort echo time imaging of ultrashort T2 tissue components in ovine brain at 3 T: a sequential D 2 O exchange study. NMR Biomed. 30 , (2017). Fan, S. et al. Yet more evidence that myelin protons can be directly imaged with UTE sequences on a clinical 3T scanner: Bicomponent analysis of native and deuterated ovine brain specimens. Magn. Reson. Med. 80 , 538–547 (2018). Deoni, S. C. L., Dean, D. C., O’Muircheartaigh, J., Dirks, H. & Jerskey, B. A. Investigating white matter development in infancy and early childhood using myelin water faction and relaxation time mapping. NeuroImage 63 , 1038–1053 (2012). Nam, Y., Lee, J., Hwang, D. & Kim, D.-H. Improved estimation of myelin water fraction using complex model fitting. NeuroImage 116 , 214–221 (2015). Schmierer, K., Scaravilli, F., Altmann, D. R., Barker, G. J. & Miller, D. H. Magnetization transfer ratio and myelin in postmortem multiple sclerosis brain. Ann. Neurol. 56 , 407–415 (2004). Chen, W. et al. Quantitative Susceptibility Mapping of Multiple Sclerosis Lesions at Various Ages. Radiology 271 , 183–192 (2014). Ma, Y.-J. et al. Ultrashort echo time (UTE) magnetic resonance imaging of myelin: technical developments and challenges. Quant. Imaging Med. Surg. 10 , 1186–1203 (2020). Jang, H. et al. Inversion recovery zero echo time (IR-ZTE) imaging for direct myelin detection in human brain: a feasibility study. Quant. Imaging Med. Surg. 10 , 895–906 (2020). Jang, H. et al. Improved volumetric myelin imaging in human brain using 3D dual echo inversion recovery‐prepared UTE with complex echo subtraction. Magn. Reson. Med. 83 , 1168–1177 (2020). Jang, H. et al. Inversion recovery UTE based volumetric myelin imaging in human brain using interleaved hybrid encoding. Magn. Reson. Med. 83 , 950–961 (2020). Ma, Y. et al. UTE MRI for assessing demyelination in an mTBI mouse model: An open-field low-intensity blast study. NeuroImage 310 , 121103 (2025). Jang, H. et al. Inversion Recovery Ultrashort TE MR Imaging of Myelin is Significantly Correlated with Disability in Patients with Multiple Sclerosis. Am. J. Neuroradiol. 42 , 868–874 (2021). Ehman, R., McNamara, M., Pallack, M., Hricak, H. & Higgins, C. Magnetic resonance imaging with respiratory gating: techniques and advantages. Am. J. Roentgenol. 143 , 1175–1182 (1984). Maclaren, J., Aksoy, M., Ooi, M. B., Zahneisen, B. & Bammer, R. Prospective motion correction using coil‐mounted cameras: Cross‐calibration considerations. Magn. Reson. Med. 79 , 1911–1921 (2018). Weick, S. et al. DC‐gated high resolution three‐dimensional lung imaging during free‐breathing. J. Magn. Reson. Imaging 37 , 727–732 (2013). Park, J., Shin, T., Yoon, S. H., Goo, J. M. & Park, J. A radial sampling strategy for uniform k ‐space coverage with retrospective respiratory gating in 3D ultrashort‐echo‐time lung imaging. NMR Biomed. 29 , 576–587 (2016). Larson, P. E. Z., Conolly, S. M., Pauly, J. M. & Nishimura, D. G. Using adiabatic inversion pulses for long‐ T2 suppression in ultrashort echo time (UTE) imaging. Magn. Reson. Med. 58 , 952–961 (2007). Stehning, C., Börnert, P., Nehrke, K., Eggers, H. & Stuber, M. Free‐breathing whole‐heart coronary MRA with 3D radial SSFP and self‐navigated image reconstruction. Magn. Reson. Med. 54 , 476–480 (2005). Piccini, D., Littmann, A., Nielles‐Vallespin, S. & Zenge, M. O. Spiral phyllotaxis: The natural way to construct a 3D radial trajectory in MRI. Magn. Reson. Med. 66 , 1049–1056 (2011). Feng, L. Golden‐Angle Radial MRI: Basics, Advances, and Applications. J. Magn. Reson. Imaging 56 , 45–62 (2022). Li, C., Gore, J. C. & Davatzikos, C. Multiplicative intrinsic component optimization (MICO) for MRI bias field estimation and tissue segmentation. Magn. Reson. Imaging 32 , 913–923 (2014). MacKay, A. et al. Insights into brain microstructure from the T2 distribution. Magn. Reson. Imaging 24 , 515–525 (2006). Weiger, M. et al. Advances in MRI of the myelin bilayer. NeuroImage 217 , 116888 (2020). Ma, Y. et al. Making the invisible visible—ultrashort echo time magnetic resonance imaging: Technical developments and applications. Appl. Phys. Rev. 9 , (2022). Cheng, K. Y. et al. Clinical application of ultrashort echo time (UTE) and zero echo time (ZTE) magnetic resonance (MR) imaging in the evaluation of osteoarthritis. Skeletal Radiol. 52 , 2149–2157 (2023). Chang, E. Y., Du, J. & Chung, C. B. UTE imaging in the musculoskeletal system. J. Magn. Reson. Imaging 41 , 870–883 (2015). Chan, R. W., Ramsay, E. A., Cheung, E. Y. & Plewes, D. B. The influence of radial undersampling schemes on compressed sensing reconstruction in breast MRI. Magn. Reson. Med. 67 , 363–377 (2012). Han, M., Chiba, K., Banerjee, S., Carballido-Gamio, J. & Krug, R. Variable flip angle three-dimensional fast spin-echo sequence combined with outer volume suppression for imaging trabecular bone structure of the proximal femur: Reduced-FOV 3D FSE Trabecular Bone MRI. J. Magn. Reson. Imaging 41 , 1300–1310 (2015). Stöcker, T. & Shah, N. J. MP‐SAGE: A new MP‐RAGE sequence with enhanced SNR and CNR for brain imaging utilizing square‐spiral phase encoding and variable flip angles. Magn. Reson. Med. 56 , 824–834 (2006). Feng, L. et al. Golden‐angle radial sparse parallel MRI: Combination of compressed sensing, parallel imaging, and golden‐angle radial sampling for fast and flexible dynamic volumetric MRI. Magn. Reson. Med. 72 , 707–717 (2014). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 09 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 30 Sep, 2025 Reviews received at journal 24 Sep, 2025 Reviews received at journal 11 Sep, 2025 Reviewers agreed at journal 04 Sep, 2025 Reviewers agreed at journal 21 Aug, 2025 Reviewers invited by journal 18 Aug, 2025 Editor assigned by journal 12 Aug, 2025 Editor invited by journal 05 Aug, 2025 Submission checks completed at journal 01 Aug, 2025 First submitted to journal 01 Aug, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7199927","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":504666729,"identity":"f82b4d7c-87ab-47d4-8d94-8f58c3bd878d","order_by":0,"name":"Jinil Park","email":"","orcid":"","institution":"University of California. Davis","correspondingAuthor":false,"prefix":"","firstName":"Jinil","middleName":"","lastName":"Park","suffix":""},{"id":504666730,"identity":"6fc1039e-5d68-4fc7-97f8-77842a2c19c9","order_by":1,"name":"Sam Sedaghat","email":"","orcid":"","institution":"University Hospital Heidelberg","correspondingAuthor":false,"prefix":"","firstName":"Sam","middleName":"","lastName":"Sedaghat","suffix":""},{"id":504666731,"identity":"2c2086a1-8aa1-4b2c-8a9c-89feabf603d5","order_by":2,"name":"Kader Karli Oguz","email":"","orcid":"","institution":"University of California. Davis","correspondingAuthor":false,"prefix":"","firstName":"Kader","middleName":"Karli","lastName":"Oguz","suffix":""},{"id":504666732,"identity":"e0872d7d-b7d3-47ae-9082-fb6e42cce3b4","order_by":3,"name":"Youngkyoo Jung","email":"","orcid":"","institution":"University of California. Davis","correspondingAuthor":false,"prefix":"","firstName":"Youngkyoo","middleName":"","lastName":"Jung","suffix":""},{"id":504666733,"identity":"90999488-e939-44d8-bf6f-45cb1f35db91","order_by":4,"name":"Eddie Fu","email":"","orcid":"","institution":"University of California. Davis","correspondingAuthor":false,"prefix":"","firstName":"Eddie","middleName":"","lastName":"Fu","suffix":""},{"id":504666734,"identity":"98c0011c-c844-46d7-95f5-68ee553ba6fa","order_by":5,"name":"Nian Wang","email":"","orcid":"","institution":"University of Texas Southwestern Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Nian","middleName":"","lastName":"Wang","suffix":""},{"id":504666735,"identity":"db835119-d975-4b1f-b2a2-1c62f8345b37","order_by":6,"name":"Fang Liu","email":"","orcid":"","institution":"Massachusetts General Hospital","correspondingAuthor":false,"prefix":"","firstName":"Fang","middleName":"","lastName":"Liu","suffix":""},{"id":504666736,"identity":"334939a5-4fa7-421a-96ab-f05d4b3bf124","order_by":7,"name":"Hyungseok Jang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABEUlEQVRIiWNgGAWjYBACAwaGBCBlAySYG5jBQgfAiIGxAb+WNCDB2NhMrBYQOIyqhQGfFnOJhIefC36dz+NnP9j+uIDhsDzf8TOGB38w2MhuOIBdi+WMhGTpmX23iyV7EhubZzAcNpx5JsfgMA9DmjEuLQY3EhKkeXtuJ244ANTCA3ShwYG0hMNApybi0ZL8m7fnXOL+8w+hWs4/SwA67D8+LWnSPD8OJG6QgNlyI/nAAR6GA7i1nHmQZs3bkJw448bDxtk8BumGM288PnCYxyDZeCYuLcdzkm/z/LFL7O9PPvCZp8Janu98YvPHHxV2sn04tDAI5CQwMLbBTcBgYAH8x4GG/cGjYBSMglEwCkYBAFhpcJFGTTeDAAAAAElFTkSuQmCC","orcid":"","institution":"University of California. Davis","correspondingAuthor":true,"prefix":"","firstName":"Hyungseok","middleName":"","lastName":"Jang","suffix":""}],"badges":[],"createdAt":"2025-07-23 22:38:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7199927/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7199927/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-026-39238-4","type":"published","date":"2026-02-09T15:58:28+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":89990368,"identity":"f209cf8d-7ff0-4789-91c3-193ca2c027cd","added_by":"auto","created_at":"2025-08-27 07:11:38","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":962081,"visible":true,"origin":"","legend":"\u003cp\u003eSequence diagram and concept of the dual-echo IR-UTE technique used for direct myelin imaging. (A) Segmented IR-UTE sequence diagram. Non-selective adiabatic inversion pulse is followed by a TI delay chosen to null long-T2 white matter water (WM\u003csub\u003eL\u003c/sub\u003e). Each TR then acquires 21 radial spokes. (B) Dual-echo UTE read-out module. Spoke uses a hard pulse excitation and collects two ultrashort echoes. Centre-out gradients capture the first ultrashort echo (TE\u003csub\u003e1 \u003c/sub\u003e\u0026lt;\u0026lt; T2*), then rewind and re-ramp to sample a second echo (TE\u003csub\u003e2\u003c/sub\u003e). (C) Longitudinal-magnetization (M\u003csub\u003ez\u003c/sub\u003e) recovery during inversion and TI delay. Myelin (red) is driven to saturation (M\u003csub\u003ez\u003c/sub\u003e ≈ 0) by the long adiabatic pulse and recovers fastest because of its short T1. Long-T2 white matter water (WM\u003csub\u003eL\u003c/sub\u003e, cyan) crosses zero at TI (dashed line), whereas long-T2 grey matter (GM\u003csub\u003eL\u003c/sub\u003e, blue) remains negative. (D) Transverse-magnetization (M\u003csub\u003exy\u003c/sub\u003e) behavior during the dual-echo read-out at the TI tuned to null point of WM\u003csub\u003eL\u003c/sub\u003e. Myelin signal (red) decays rapidly between TE\u003csub\u003e1\u003c/sub\u003e and TE\u003csub\u003e2\u003c/sub\u003e, while GM\u003csub\u003eL\u003c/sub\u003e signal (blue) changes negligibly. Subtracting the two echoes therefore cancels long-T2 components and yields a myelin-specific image (green arrow).\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7199927/v1/06251ee636512cefc39f3b26.png"},{"id":89988278,"identity":"4d4c1dae-5cf1-48d1-8902-9ea88fe62ade","added_by":"auto","created_at":"2025-08-27 07:03:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1098700,"visible":true,"origin":"","legend":"\u003cp\u003eSuperior–inferior (SI) self-navigator for motion gating in segmented IR-UTE. (A) Sequence diagram of the SI self-navigator used in the segmented IR-UTE sequence. Each IR segment consists of 21 image spokes and a single SI-navigator spoke (gray) at the end. (B) Enlarged view of the navigator spoke: a center-out read-out is applied along k\u003csub\u003ez\u003c/sub\u003e. (C) Head outline before (solid) and after (dashed) an intentional nod. (D) SI-projection profiles, showing profiles altered by intentional nods. (E) Navigator amplitude over time during the nodding experiment. The intentional head nods every 20 s are well represented (dashed line).\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7199927/v1/368593aed4ad2d66704dcf99.png"},{"id":89988281,"identity":"e50ccb2b-dffe-41e3-93ee-b4d8f2abf1d2","added_by":"auto","created_at":"2025-08-27 07:03:38","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":5329967,"visible":true,"origin":"","legend":"\u003cp\u003eSimulation of motion-gating with sequential versus bit-reversed view ordering. (A) Two-dimensional digital brain phantom comprising cerebrospinal fluid (CSF), cortical grey matter (GM\u003csub\u003eL\u003c/sub\u003e), deep white matter with myelin (WM\u003csub\u003eL\u003c/sub\u003e/ Myelin), skull, subcutaneous fat, and a synthetic demyelinated lesion (L1–L4). (B) Proton-density (PD), T1, and T2 values assigned to each tissue class. (C) Dual-echo IR-UTE images after deleting 10 % of spokes to mimic retrospective gating. Sequential ordering produces coherent streaks and blurring in the myelin-weighted image (red arrows) and pronounced PSF side-lobes (yellow arrows), whereas bit-reversed view ordering disperses the missing spokes, yielding minimal artifacts and an isotropic PSF.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7199927/v1/fb10c4a6397c36af9b242967.png"},{"id":89988287,"identity":"c51c68f7-b9f5-4bf4-82ab-8fa249ea9122","added_by":"auto","created_at":"2025-08-27 07:03:38","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":3598189,"visible":true,"origin":"","legend":"\u003cp\u003eSelf-navigator gating signal (A) and the distribution of motion-gated spokes for sequential and bit-reversed view ordering, respectively (B, C). (A) Superior–inferior navigator signal obtained from the scan with intentional motion (Scan #3). Baseline shifts (black arrows) indicate large head nods. The motionless data are selected by the threshold gate and used for image reconstruction (black line, 10.7% of the 12,000 rejected spokes). (B) Endpoints of the gated(red) and rejected (black) spokes plotted on a unit sphere for sequential view ordering. Gating removes a contiguous sector, creating a large k-space part. (C) Corresponding distribution for bit-reversed view ordering. The same number of rejected spokes is pseudo-randomly dispersed, maintaining quasi-uniform k-space coverage.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7199927/v1/a0814c0aa74d28474ca6c837.png"},{"id":89988286,"identity":"9b3d3b69-9d9f-4f4d-aedc-d778ae04cb24","added_by":"auto","created_at":"2025-08-27 07:03:38","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":16510269,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of retrospective motion gating on sequential versus bit-reversed IR-UTE imaging (42-year-old male). Sequential view ordering case is shown in the upper two rows (Scan #1, A-G) and bit-reversed ordering case is shown in the lower two rows (Scan #2, H-N). Within each data, the first row shows the ungated (fully data) reconstruction, and the second row shows the gated (undersampled) reconstruction obtained after discarding 1,280 spokes (10.7 %) identified by the navigator trace from Scan #3. Sequential view ordering produces focal signal alteration (G, white arrow), whereas the bit-reversed view ordering yields only low-level, incoherent noise (N, white arrows), which is also confirmed by difference images (between gated and ungated) shown in D and K.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7199927/v1/6e86982a15c3a9a64ca3954d.png"},{"id":89988308,"identity":"12bfa19c-91ef-4889-8a6d-a0df2706394f","added_by":"auto","created_at":"2025-08-27 07:03:39","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":6921375,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of self-gating in the in vivo experiment with intentional head motion (Scan #3 with a 42-year-old male). Top row (A–C) shows the results from the full 12,000 spokes, without motion gating. Bottom row (D–F) shows the results after discarding the 1,280 spokes selected by the SI navigator (10.7 %). White arrows highlight artifacts caused by the head nods, which are conspicuous in the ungated images (C) but largely suppressed after gating (F).\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-7199927/v1/9ff3afa640264112b8337d58.png"},{"id":102785285,"identity":"52ccb06f-9a75-4017-9374-681d0e325b83","added_by":"auto","created_at":"2026-02-16 16:04:21","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":38849142,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7199927/v1/628d3e2c-7bd2-408d-9bf4-569dd0544430.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Motion-Robust Direct Myelin Imaging in MRI using Self-Gating","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMyelin is a dense layer of cell membrane that wraps around axons, produced by non-neuronal cells known as oligodendrocytes \u003csup\u003e1,2\u003c/sup\u003e. Structurally, it consists of a multilayered membrane with alternating lipid and protein layers, forming an insulating sheath around nerve fibers in both the white and gray matter of the brain, as well as in the spinal cord. This insulation boosts the speed of action potential conduction by approximately 100 times and reduces the refractory period by more than 30 times \u003csup\u003e3\u003c/sup\u003e. Collectively, these enhancements amplify the brain\u0026apos;s capacity for information processing by over 3,000 times, making myelin essential for normal neurological functions such as speech, coordination, and cognition \u003csup\u003e3\u0026ndash;5\u003c/sup\u003e. When axons lose their myelin, they also lose saltatory conduction, mechanical protection, and metabolic support, which can lead to structural and functional damage \u003csup\u003e6\u0026ndash;8\u003c/sup\u003e. Myelin fundamentally transforms how electrical signals are initiated and propagated in the nervous system, and its disruption plays a key role in the pathology of numerous neurological disorders, including multiple sclerosis (MS) \u003csup\u003e9\u003c/sup\u003e, traumatic brain injury (TBI) \u003csup\u003e10,11\u003c/sup\u003e, epilepsy \u003csup\u003e12\u003c/sup\u003e, and Alzheimer\u0026apos;s disease (AD) \u003csup\u003e13\u0026ndash;15\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eDirect assessment of myelin integrity in both white and gray matter may be important for diagnosing and evaluating the prognosis of myelin-related disorders. However, the hydrogens (\u003csup\u003e1\u003c/sup\u003eH, the source of signal in MRI) in myelin have extremely short T2 relaxation times (T2* \u0026lt; ~0.3 ms at 3T) and therefore cannot be directly imaged using conventional MRI sequences, which typically have echo times (TEs) of several milliseconds or longer \u003csup\u003e16\u0026ndash;23\u003c/sup\u003e. As a result, conventional sequences provide only an indirect or qualitative assessment of myelin. Another challenge in directly imaging myelin is selectivity, where long T2 water is present in much higher concentrations in the brain and produces a signal that is approximately 10\u0026ndash;20 times stronger than that of myelin \u003csup\u003e24\u003c/sup\u003e. Consequently, myelin is invisible on conventional MRI. Alternatively, indirect imaging techniques have been explored, including myelin water fraction (MWF) mapping \u003csup\u003e25,26\u003c/sup\u003e, magnetization transfer (MT) imaging \u003csup\u003e27\u003c/sup\u003e, and quantitative susceptibility mapping (QSM) \u003csup\u003e28\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eRecently, ultrashort echo time (UTE) sequences have been investigated to directly capture the short-lived myelin signal. To suppress long T2 signal and hence improve signal contrast from myelin, adiabatic inversion pulses have been utilized to provide robust and uniform suppression of long T2 water signals \u003csup\u003e29\u0026ndash;31\u003c/sup\u003e. Combined with the echo subtraction technique, inversion recovery prepared UTE (IR-UTE) can effectively detect myelin signal directly with high signal contrast in the brain \u003csup\u003e31,32\u003c/sup\u003e. The IR-UTE technique has been validated in both TBI and MS, showing promise in detecting changes in myelin content. For example, Ma et al. demonstrated that IR-UTE enables quantitative myelin imaging in the mouse brain using a preclinical 3T MRI scanner, allowing detection of white matter demyelination following open-field low-intensity blast injury (a model of mild TBI) \u003csup\u003e33\u003c/sup\u003e. Jang et al. reported that the myelin signal detected by IR-UTE significantly correlates with disability in MS patients, as measured by the Expanded Disability Status Scale (EDSS) \u003csup\u003e34\u003c/sup\u003e. These findings suggest that IR-UTE may serve as a potential imaging-based biomarker for diseases involving myelin alterations.\u003c/p\u003e\n\u003cp\u003eDespite promising results, a key limitation of the current IR-UTE technique is its long scan time (~\u0026thinsp;10 minutes). This extended duration is primarily due to the inversion recovery imaging scheme, which requires a long repetition time (TR) of approximately 1 second. Additionally, since myelin imaging typically yields low signal intensity, a longer scan time is often necessary to improve the signal-to-noise ratio (SNR). However, prolonged imaging increases the risk of motion-related artifacts, such as blurriness, ghosting, and streaking, caused by intra-scan head movement. These artifacts can significantly degrade the already low-SNR myelin images. The issue is further exacerbated in patients with neurological disorders, who often have impaired motor control and may struggle to remain still during lengthy scans.\u003c/p\u003e\n\u003cp\u003eIn MRI, various methods have been developed to prevent or correct motion-related artifacts. Prospective motion gating typically requires specialized hardware, such as a respiratory bellows, an electrocardiogram gating device, or a motion-tracking camera\u003csup\u003e35,36\u003c/sup\u003e. Retrospective motion gating often involves additional data acquisition to capture motion-related information, commonly referred to as a navigator signal \u003csup\u003e37,38\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn this study, we propose a self-gated IR-UTE technique incorporating bit-reversed view ordering to correct for motion-related artifacts and enhance direct myelin imaging, without requiring additional scan time.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eIR-UTE with dual-echo subtraction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMyelin has a short T2* relaxation time and low proton density (i.e., density of \u003csup\u003e1\u003c/sup\u003eH) \u003csup\u003e23,24\u003c/sup\u003e, resulting in an extremely low \u0026ldquo;detectable\u0026rdquo; MR signal. As a result, it is sensitive to noise, aliasing or streak artifacts, ghosting, and image reconstruction errors, any of which can degrade myelin delineation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn addition, the large dynamic-range gap between myelin and neighboring structures increases the risk of quantization errors that compromise signal.\u003c/p\u003e\n\u003cp\u003eTo overcome these limitations, a method that combines an IR preparation pulse with UTE dual-echo subtraction has recently been proposed \u003csup\u003e31\u003c/sup\u003e. In this approach, an adiabatic inversion pulse is applied, followed by acquisition of two images at different TEs during the inversion time (TI) (Figures 1A, B). The TI is selected so that the long-T2 white matter (WM\u003csub\u003eL\u003c/sub\u003e) is nulled (Figure 1c). Because myelin\u0026rsquo;s T2* is extremely short, a long adiabatic inversion pulse cannot invert the magnetization; instead, it drives it to saturation (M\u003csub\u003ez\u003c/sub\u003e \u0026asymp; 0) \u003csup\u003e39\u003c/sup\u003e. Immediately thereafter, the saturated myelin recovers rapidly owing to its short T1. At the chosen TI, the first echo captures signals of myelin, long-T2 grey matter (GM\u003csub\u003eL\u003c/sub\u003e), and any remaining long-T2 components, whereas the second echo is acquired after the myelin signal has substantially decayed (Figure 1D). Subtracting the two echoes therefore isolates the myelin-specific signal. Because the inversion recovery requires a long TR, acquiring one spoke per IR pulse significantly increases the scan time. To address this problem, the sequence acquires multiple k-space spokes segmented around the white matter null point, increasing acquisition efficiency (Figure 1A).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSelf-navigating projection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis section describes our approach to motion tracking using self-navigating. Due to the oversampled center of k-space, non-Cartesian radial trajectories are known to be intrinsically more robust to motion than Cartesian trajectories. However, the motion can still affect the radial imaging, with unwanted image blurring, streaking, and ghosting. Therefore, identifying motion during scanning and discarding the motion-corrupted data can enhance the image quality. The need becomes even more critical when the subject makes an unexpected, bulk movement (e.g., patients with impaired motor function), after which the head may settle in a different position and appear severe artifacts. It is thus preferable to measure motion during the scan and address it retrospectively.\u003c/p\u003e\n\u003cp\u003eIn the proposed approach, we monitor head motion using a superior-inferior (SI) projection navigator. The navigator is inserted at regular intervals to the end of each segment (Figure 2A). Placing the navigator at the end of segment maintains a consistent TI, ensuring stable contrast in motion tracking signal. Widely used in lung imaging, this navigator periodically applies a short readout gradient along the G\u003csub\u003ez\u003c/sub\u003e axis, acquires a half k-space line, and then performs a 1D Fourier transform to generate a SI projection profile (Figures 2B, C) \u003csup\u003e38,40\u003c/sup\u003e. Continuous comparison of the profiles enables tracking movement of head throughout the IR-UTE acquisition (Figure 2D). Figure 2E shows an example of motion tracking using the self-navigator, estimated from the signal intensity at a fixed pixel location in the 1D profile, defined by the position of maximum signal at the first time point. In this example, the subject intentionally nodded their head every 20 seconds during the scan. When the head nods, the projection profile changes; this variation can be measured and incorporated into the image reconstruction process to reduce motion artifacts by excluding motion-contaminated data.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBit-reversed view ordering\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn radial trajectories, it is important for retrospective gating to maintain that the remaining spokes are uniformly distributed in k-space after discarding unwanted data. If the spokes are acquired sequentially, the data loss caused by gating will be concentrated in a specific region of k-space, resulting in an undersampling artifact. Various sampling strategies such as golden-angle, phyllotaxis, and interleaved schemes have been proposed to avoid this problem \u003csup\u003e38,41,42\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eHere we adopt a bit-reversed view ordering. By reversing the binary index of each spoke, the original trajectory is rearranged into a pseudo-random sequence. For example, the 52\u003csup\u003end\u003c/sup\u003e spoke (binary 110100) becomes the 13\u003csup\u003eth\u003c/sup\u003e spoke (binary 001011) after the reordering. The main advantage is that it can be applied to any kind of trajectory, providing a pseudo-random view ordering.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSimulation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA computer simulation was performed to test whether bit‑reversed view ordering can compensate for the k‑space non‑uniformity introduced by retrospective motion gating in IR‑UTE myelin imaging.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA 2D high-resolution digital head phantom (654 \u0026times; 654 pixels) consisting of six tissue components\u0026mdash;cerebrospinal fluid (CSF), GM\u003csub\u003eL\u003c/sub\u003e, WM\u003csub\u003eL\u003c/sub\u003e, myelin within white matter, skull bone, and subcutaneous fat\u0026mdash;was created, as illustrated in Figure 3A. To construct the segmentation map, a sample brain map from the open-source segmentation tool, multiplicative intrinsic component optimization (MICO) \u003csup\u003e43\u003c/sup\u003e, was used, which provided labels for white and gray matters. The remaining three components (CSF, skull, and subcutaneous fat) were manually added. Four demyelinated lesions were placed within the white matter region. Using the segmentation map, a Bloch simulation was performed based on the tissue parameters shown in Figure 3B. All simulations were conducted in MATLAB R2024a (MathWorks, Natick, MA) using in-house code that solves the Bloch equations analytically to model signal evolution.\u003c/p\u003e\n\u003cp\u003eThe simulation reproduced an IR‑UTE dual‑echo acquisition, using the following sequence parameters: flip angle = 15\u0026deg;, TR = 1000 ms, inversion time TI = 379.9 ms (null point of WM\u003csub\u003eL\u003c/sub\u003e), readout bandwidth = 250 kHz, and field of view = 220 \u0026times; 220 mm\u003csup\u003e2\u003c/sup\u003e, matrix size = 220 \u0026times; 220. Two radial echoes were collected per spoke (TE\u003csub\u003e1\u003c/sub\u003e = 0.03 ms, TE\u003csub\u003e2\u003c/sub\u003e = 3 ms). A total of 692 spokes covered 2D circular k-space, delivered in segments of 21 spokes; the spoke‑to‑spoke interval was 6 ms. To assess sampling robustness, k‑space was traversed either sequentially or with bit‑reversed ordering. The data loss due to motion gating was assumed to be 10% of the total spokes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIn vivo study\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThree healthy male volunteers (42, 42, and 38-year-old) were scanned on a 3T clinical MRI system (MAGNETOM Prisma Fit, Siemens Healthineers, Erlangen, Germany) using a 32‑channel head/neck coil. A written informed consent was collected under an institutional‑review‑board approval (University of California, Davis). All procedures were performed in accordance with the Declaration of Helsinki and relevant institutional guidelines and regulations. Three scans were performed using 3D IR-UTE sequence: Sequential view ordering without motion (Scan #1), bit-reversed view ordering without motion(Scan #2), and bit-reversed view ordering with intentional motion (Scan #3). During Scan #3, the participant was instructed to nod their head once with a large movement. Scans #1 and #2 served as motion-free references, acquired using two different view ordering strategies, allowing for a controlled comparison of their effects under retrospective undersampling. In Scan #3, motion-corrupted data were identified and excluded using a self-gating technique. The actual motion \u0026nbsp;profile obtained in Scan #3 was then reused to simulate gating-induced undersampling effects in Scans #1 and #2.\u003c/p\u003e\n\u003cp\u003eThe imaging parameters were: flip angle = 10\u0026deg;, TR = 1000 ms, TI = 327 ms (white matter null point tuned at 3T), dual echo readout (TE\u003csub\u003e1\u003c/sub\u003e = 0.03 ms, TE\u003csub\u003e2\u003c/sub\u003e = 3 ms), readout bandwidth = 92 kHz, field of view = 220 \u0026times; 220 \u0026times; 176 mm\u003csup\u003e3\u003c/sup\u003e, matrix size = 220 \u0026times; 220 \u0026times; 88, 20 spokes per segment, total spokes = 12,000, and spoke to spoke interval = 5 ms. A 50 \u0026micro;s nonselective hard pulse excited each spoke, and a 10.24 ms hyperbolic secant inversion pulse preceded every segment.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cb\u003eSimulation\u003c/b\u003e\u003c/p\u003e\u003cp\u003eBit-reversed view ordering significantly reduced artifacts caused by retrospective motion gating (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). When 10% of spokes were removed using sequential view ordering, prominent streaks appeared along the spoke directions, accompanied by blurring in myelin-rich regions (red arrows). In contrast, bit-reversed view ordering distributed the data loss more evenly, leading to a noticeable reduction in streaking artifacts. This improvement was also evident in the simulated point spread function (PSF): sequential view ordering resulted in pronounced side-lobes (yellow\u003c/p\u003e\u003cp\u003e\u003cb\u003eIn vivo study \u0026ndash; Self navigator and gating\u003c/b\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e4\u003c/span\u003eA plots the navigator signal from Scan #3. The changed baseline reflects the changed head position due to intentional head nodding. The data with the navigator signal changed by nodding was excluded using threshold-based gating and then used for reconstruction to effectively suppress motion artifacts. IR-UTE acquired 12,000 spokes and rejected 1,280 (10.67%) of them, which were head position data changed by nodding. The spokes where head position data changed by nodding are removed (black dots) are continuously missing in sequential view ordering (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e4\u003c/span\u003eB), while bit-reversed view ordering scatters the missing segments pseudo randomly throughout the sphere (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e4\u003c/span\u003eC).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eIn vivo study \u0026ndash; Myelin imaging\u003c/b\u003e\u003c/p\u003e\u003cp\u003eFor all three volunteers, the self-gating approach significantly improved the quality of myelin imaging in the presence of intentional head motion.\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e5\u003c/span\u003e contrasts the myelin images from the two motion-free reference data (Scan #1 and Scan #2) obtained with sequential view ordering or bit-reversed view ordering, which were undersampled using a simulated motion-gating trace (obtained from Scan #3). This demonstrates how sequential and bit-reversed view orderings influence the myelin imaging, by comparing the images from ungated (fully sampled) and gated (undersampled) data. With sequential view ordering, the simulated gating produced focal signal loss in myelin (white arrows in Figs.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e5\u003c/span\u003eC and \u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e5\u003c/span\u003eG). In the corresponding difference image between ungated and gated myelin images (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e5\u003c/span\u003eD), the error appears as coherent signal change across the brain. In contrast, bit-reversed view ordering improves the myelin image in both subcortical and deep-white-matter regions (white arrows in Figs.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e5\u003c/span\u003eJ and \u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e5\u003c/span\u003eN). The corresponding difference image exhibits incoherent low-level noise, underscoring the scheme\u0026rsquo;s superior robustness (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e5\u003c/span\u003eK).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e6\u003c/span\u003e shows the myelin imaging with the intentional head motion (Scan #3), where self-gating and bit-reversed view ordering were utilized. Figures\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e6\u003c/span\u003eA-\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e6\u003c/span\u003eC were reconstructed from all acquired spokes without motion gating. Although the full data typically allows higher SNR than the undersampled data, the head nods may introduce blurring and ghost artifacts, impairing overall image quality and myelin contrast. Figures\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e6\u003c/span\u003eD-\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e6\u003c/span\u003eF show the result from the same data after retrospective motion gating. Spokes coinciding with nodding motion, detected by self-gating, were discarded before reconstruction. Despite the reduced number of data sampling, sharper myelin boundaries and unbiased myelin signal were achieved by omitting the motion-contaminated data (white arrows).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eMyelin has extremely short T2s (\u0026lt;\u0026thinsp;1 ms), due to its highly ordered microstructure and chemical shift with multiple lipid peaks. Therefore myelin cannot be directly imaged with conventional MRI sequences, which typically have TEs of several milliseconds or longer \u003csup\u003e16\u0026ndash;23\u003c/sup\u003e. As a result, conventional sequences only provide an indirect assessment of myelin. The MR properties of myelin have been investigated by several groups. Broad line proton spectroscopy studies have demonstrated that myelin is in a liquid-crystalline state \u003csup\u003e16\u003c/sup\u003e. Multi-component analysis of spin echo or free induction decay of white matter samples has shown a broad range of short T2 or T2* values for myelin protons (e.g., ~\u0026thinsp;50 \u0026micro;s \u003csup\u003e44\u003c/sup\u003e, 7.5\u0026ndash;101 \u0026micro;s \u003csup\u003e45\u003c/sup\u003e, 50-1000 \u0026micro;s \u003csup\u003e18\u003c/sup\u003e, 150\u0026ndash;250 \u0026micro;s \u003csup\u003e24\u003c/sup\u003e, etc.). UTE MRI sequences with TEs as short as ~\u0026thinsp;8\u0026ndash;32 \u0026micro;s, which are 100\u0026ndash;1000 times shorter than those of conventional clinical sequences, allow direct detection of signal from myelin \u003csup\u003e46\u0026ndash;48\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eBecause the myelin signal is significantly lower than that of long T2 tissues in the brain, direct myelin imaging techniques rely on IR preparation and echo subtraction to suppress the long T2 signal. Despite the effectiveness of this suppression, the inherently low SNR of the myelin lipid remains a major challenge. In this context, even minor imaging artifacts can severely impact the detection of myelin. Unlike conventional Cartesian MRI, where motion typically results in ghosting artifacts along the phase-encoding direction, UTE imaging, based on a radial trajectory, produces motion artifacts that radiate in all directions. As a result, strong short T2 signals from surrounding bone or fat, as well as residual long T2 signals, can interfere with the detection of the myelin signal if motion occurs during the scan.\u003c/p\u003e\u003cp\u003eIn this study, we demonstrated the efficacy of motion-gating using self-navigator and bit-reversed view ordering in IR-UTE for improved myelin imaging. The self-navigator does not impose any additional scan time because it is located at the end of segments (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e2\u003c/span\u003eA), making it clinically feasible. Our findings show that combining SI self-navigator with a bit-reversed radial trajectory markedly improves the motion robustness of IR-UTE myelin imaging. In the motion-free volunteer, both ordering schemes scanned, when the navigator-derived gating trace was applied retrospectively the sequential data developed focal signal decrease, whereas the bit-reversed view ordering data retained cortical detail and myelin contrast (Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e5\u003c/span\u003e). When intentionally nodded, about 10% of the spokes were removed by the navigator. nevertheless, the gated, bit-reversed view ordering reconstruction recovered sharp enhanced grey/white contrast, outperforming the ungated image that had the full data set (Fig.\u0026nbsp;\u003cspan refid=\"Fig12\" class=\"InternalRef\"\u003e6\u003c/span\u003e). These results underscore two practical points: (i) dispersing missing spokes pseudo-randomly with bit-reversal converts coherent streaks into low-energy noise, and (ii) placing the SI navigator at each segment boundary permits effective gating without extending scan time. Because both modifications require only a gradient-table permutation and a single 1D readout, they can be deployed on existing clinical scanners and may enable reliable, motion-tolerant myelin mapping in routine neuroimaging.\u003c/p\u003e\u003cp\u003eBit-reversing is a simple yet powerful method for achieving pseudo-randomized view ordering\u003csup\u003e49\u003c/sup\u003e. In IR-prepared imaging, this approach can improve image quality, particularly when a large number of spokes are acquired per inversion recovery, by redistributing T1 blurring effects in a pseudo-random manner. For example, multiple spokes may be acquired near the nulling point of long T2 white matter, with roughly half acquired before the null point and half after. Spokes acquired before the null point capture negative longitudinal magnetization (M\u003csub\u003ez\u003c/sub\u003e), while those acquired after the null point capture positive M\u003csub\u003ez\u003c/sub\u003e from the long T2 white matter. Each spoke samples a different M\u003csub\u003ez\u003c/sub\u003e magnitude, governed by the TI and T1 relaxation. Consequently, the encoded k-space can exhibit structured modulation patterns as a function of both T1 recovery and view ordering, which may degrade the PSF. Previous studies have explored variable flip angle excitation to compensate for this kind of signal variation \u003csup\u003e50,51\u003c/sup\u003e. However, determining optimal flip angles based on the long T2 white matter signal is not possible in the direct myelin imaging, as the target signal is ideally zero at the nulling point. As an alternative, bit-reversed view ordering randomizes these modulation patterns, making them more 'noise-like' and thereby reducing coherent artifacts. Further in-depth investigation of this approach will be pursued in future studies.\u003c/p\u003e\u003cp\u003eThis study has several limitations. First, only three healthy volunteers were included to validate the proposed approach. Although the volunteers were instructed to make intentional motions during the scan, this may not accurately represent the conditions seen in patient groups with various neurological disorders. We plan to test this sequence in patients with a range of neurological diseases in future studies. Second, we did not compare our method with alternative view ordering schemes, such as the golden angle method \u003csup\u003e52\u003c/sup\u003e. While the golden angle approach has demonstrated effectiveness in radial undersampled and time-resolved imaging, our goal is to distribute motion-contaminated spokes in a pseudo-random manner, which differs from the objectives targeted by golden angle techniques. Third, we only incorporated an SI navigator, which is capable of detecting nodding motion. Head motion in other directions, such as shaking, can be detected by adding additional navigators along the G\u003csub\u003ex\u003c/sub\u003e or G\u003csub\u003ey\u003c/sub\u003e axes, without increasing scan time.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study demonstrates the efficacy of motion gating using self-navigator and bit-reversed view ordering in IR-UTE for motion-robust myelin imaging. Bit-reversed view ordering is a one-line permutation of the gradient list that pseudo-randomizes spoke positions and keeps k-space nearly uniform even after data are rejected. The proposed approach demands no additional hardware, no extra scan time, but only minor modification on sequence, and hence it can be readily implemented on current clinical 3T systems. Combined with direct myelin imaging technique, the proposed approach could benefit patient groups prone to motion such as children, elderly, or epilepsy cohorts.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eResearch reported in this publication was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (R01AR078877) and the National Institute on Aging (P30AG072972) of the National Institutes of Health (NIH). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. This research was also supported by the Deutsche Forschungsgemeinschaft (DFG, SE 3272/1-1) and University Heidelberg Project Expanding Internationality.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis paper was co-authored by Jinil Park, Sam Sedaghat, Kader Karli Oguz, Youngkyoo Jung, Eddie Fu, Nian Wang, Fang Liu and Hyungseok Jang, with me serving as the first author and Dr. Hyungseok Jang as the corresponding author. All authors contributed substantially to the conception, design, and data acquisition or interpretation, as well as to the drafting and revision of this manuscript. Each author has reviewed and approved the final version.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eFields, R. D. Myelin\u0026mdash;More than Insulation. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e344\u003c/strong\u003e, 264\u0026ndash;266 (2014).\u003c/li\u003e\n\u003cli\u003eJahn, O., Tenzer, S. \u0026amp; Werner, H. B. Myelin Proteomics: Molecular Anatomy of an Insulating Sheath. \u003cem\u003eMol. Neurobiol.\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 55\u0026ndash;72 (2009).\u003c/li\u003e\n\u003cli\u003eLu, P. H. \u003cem\u003eet al.\u003c/em\u003e Myelin breakdown mediates age-related slowing in cognitive processing speed in healthy elderly men. \u003cem\u003eBrain Cogn.\u003c/em\u003e \u003cstrong\u003e81\u003c/strong\u003e, 131\u0026ndash;138 (2013).\u003c/li\u003e\n\u003cli\u003eWaxman, S. G. Conduction in Myelinated, Unmyelinated, and Demyelinated Fibers. \u003cem\u003eArch. Neurol.\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e, 585\u0026ndash;589 (1977).\u003c/li\u003e\n\u003cli\u003eFelts, P. A., Baker, T. A. \u0026amp; Smith, K. J. Conduction in Segmentally Demyelinated Mammalian Central Axons. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 7267\u0026ndash;7277 (1997).\u003c/li\u003e\n\u003cli\u003eLee, Y. \u003cem\u003eet al.\u003c/em\u003e Oligodendroglia metabolically support axons and contribute to neurodegeneration. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e487\u003c/strong\u003e, 443\u0026ndash;448 (2012).\u003c/li\u003e\n\u003cli\u003eChabas, D. \u003cem\u003eet al.\u003c/em\u003e The Influence of the Proinflammatory Cytokine, Osteopontin, on Autoimmune Demyelinating Disease.\u003c/li\u003e\n\u003cli\u003eF\u0026uuml;nfschilling, U. \u003cem\u003eet al.\u003c/em\u003e Glycolytic oligodendrocytes maintain myelin and long-term axonal integrity. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e485\u003c/strong\u003e, 517\u0026ndash;521 (2012).\u003c/li\u003e\n\u003cli\u003eNoseworthy, J. H., Rodriguez, M. \u0026amp; Weinshenker, B. G. Multiple Sclerosis. \u003cem\u003eN. Engl. J. Med.\u003c/em\u003e (2000).\u003c/li\u003e\n\u003cli\u003eFlygt, J., Djupsj\u0026ouml;, A., Lenne, F. \u0026amp; Marklund, N. Myelin loss and oligodendrocyte pathology in white matter tracts following traumatic brain injury in the rat. \u003cem\u003eEur. J. Neurosci.\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 2153\u0026ndash;2165 (2013).\u003c/li\u003e\n\u003cli\u003eWright, A. D. \u003cem\u003eet al.\u003c/em\u003e Myelin Water Fraction Is Transiently Reduced after a Single Mild Traumatic Brain Injury \u0026ndash; A Prospective Cohort Study in Collegiate Hockey Players. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, e0150215 (2016).\u003c/li\u003e\n\u003cli\u003eDrenthen, G. S. \u003cem\u003eet al.\u003c/em\u003e On the merits of non-invasive myelin imaging in epilepsy, a literature review. \u003cem\u003eJ. Neurosci. Methods\u003c/em\u003e \u003cstrong\u003e338\u003c/strong\u003e, 108687 (2020).\u003c/li\u003e\n\u003cli\u003eKavroulakis, E. \u003cem\u003eet al.\u003c/em\u003e Myelin content changes in probable Alzheimer\u0026rsquo;s disease and mild cognitive impairment: Associations with age and severity of neuropsychiatric impairment. \u003cem\u003eJ. Magn. Reson. Imaging\u003c/em\u003e \u003cstrong\u003e47\u003c/strong\u003e, 1359\u0026ndash;1372 (2018).\u003c/li\u003e\n\u003cli\u003eDean, D. C. \u003cem\u003eet al.\u003c/em\u003e Association of Amyloid Pathology With Myelin Alteration in Preclinical Alzheimer Disease. \u003cem\u003eJAMA Neurol.\u003c/em\u003e \u003cstrong\u003e74\u003c/strong\u003e, 41 (2017).\u003c/li\u003e\n\u003cli\u003eHaroutunian, V. \u003cem\u003eet al.\u003c/em\u003e Myelination, oligodendrocytes, and serious mental illness. \u003cem\u003eGlia\u003c/em\u003e \u003cstrong\u003e62\u003c/strong\u003e, 1856\u0026ndash;1877 (2014).\u003c/li\u003e\n\u003cli\u003eLecar, H., Ehrenstein, G. \u0026amp; Stillman, I. Detection of Molecular Motion in Lyophilized Myelin by Nuclear Magnetic Resonance. \u003cem\u003eBiophys. J.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 140\u0026ndash;145 (1971).\u003c/li\u003e\n\u003cli\u003eWaldman, A. \u003cem\u003eet al.\u003c/em\u003e MRI of the brain with ultra-short echo-time pulse sequences. \u003cem\u003eNeuroradiology\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e, 887\u0026ndash;892 (2003).\u003c/li\u003e\n\u003cli\u003eHorch, R. A., Gore, J. C. \u0026amp; Does, M. D. Origins of the ultrashort‐T2\u003csup\u003e1\u003c/sup\u003eH NMR signals in myelinated nerve: A direct measure of myelin content? \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e \u003cstrong\u003e66\u003c/strong\u003e, 24\u0026ndash;31 (2011).\u003c/li\u003e\n\u003cli\u003eWilhelm, M. J. \u003cem\u003eet al.\u003c/em\u003e Direct magnetic resonance detection of myelin and prospects for quantitative imaging of myelin density. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e \u003cstrong\u003e109\u003c/strong\u003e, 9605\u0026ndash;9610 (2012).\u003c/li\u003e\n\u003cli\u003eDu, J. \u003cem\u003eet al.\u003c/em\u003e Ultrashort echo time (UTE) magnetic resonance imaging of the short T2 components in white matter of the brain using a clinical 3T scanner. \u003cem\u003eNeuroImage\u003c/em\u003e \u003cstrong\u003e87\u003c/strong\u003e, 32\u0026ndash;41 (2014).\u003c/li\u003e\n\u003cli\u003eDu, J. \u003cem\u003eet al.\u003c/em\u003e Measurement of T1 of the Ultrashort T2* Components in White Matter of the Brain at 3T. \u003cem\u003ePLoS ONE\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, e103296 (2014).\u003c/li\u003e\n\u003cli\u003eHe, Q. \u003cem\u003eet al.\u003c/em\u003e Direct magnitude and phase imaging of myelin using ultrashort echo time (UTE) pulse sequences: A feasibility study. \u003cem\u003eMagn. Reson. Imaging\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 194\u0026ndash;199 (2017).\u003c/li\u003e\n\u003cli\u003eFan, S., Ma, Y., Chang, E. Y., Bydder, G. M. \u0026amp; Du, J. Inversion recovery ultrashort echo time imaging of ultrashort T2 tissue components in ovine brain at 3 T: a sequential D\u003csub\u003e2\u003c/sub\u003eO exchange study. \u003cem\u003eNMR Biomed.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, (2017).\u003c/li\u003e\n\u003cli\u003eFan, S. \u003cem\u003eet al.\u003c/em\u003e Yet more evidence that myelin protons can be directly imaged with UTE sequences on a clinical 3T scanner: Bicomponent analysis of native and deuterated ovine brain specimens. \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e \u003cstrong\u003e80\u003c/strong\u003e, 538\u0026ndash;547 (2018).\u003c/li\u003e\n\u003cli\u003eDeoni, S. C. L., Dean, D. C., O\u0026rsquo;Muircheartaigh, J., Dirks, H. \u0026amp; Jerskey, B. A. Investigating white matter development in infancy and early childhood using myelin water faction and relaxation time mapping. \u003cem\u003eNeuroImage\u003c/em\u003e \u003cstrong\u003e63\u003c/strong\u003e, 1038\u0026ndash;1053 (2012).\u003c/li\u003e\n\u003cli\u003eNam, Y., Lee, J., Hwang, D. \u0026amp; Kim, D.-H. Improved estimation of myelin water fraction using complex model fitting. \u003cem\u003eNeuroImage\u003c/em\u003e \u003cstrong\u003e116\u003c/strong\u003e, 214\u0026ndash;221 (2015).\u003c/li\u003e\n\u003cli\u003eSchmierer, K., Scaravilli, F., Altmann, D. R., Barker, G. J. \u0026amp; Miller, D. H. Magnetization transfer ratio and myelin in postmortem multiple sclerosis brain. \u003cem\u003eAnn. Neurol.\u003c/em\u003e \u003cstrong\u003e56\u003c/strong\u003e, 407\u0026ndash;415 (2004).\u003c/li\u003e\n\u003cli\u003eChen, W. \u003cem\u003eet al.\u003c/em\u003e Quantitative Susceptibility Mapping of Multiple Sclerosis Lesions at Various Ages. \u003cem\u003eRadiology\u003c/em\u003e \u003cstrong\u003e271\u003c/strong\u003e, 183\u0026ndash;192 (2014).\u003c/li\u003e\n\u003cli\u003eMa, Y.-J. \u003cem\u003eet al.\u003c/em\u003e Ultrashort echo time (UTE) magnetic resonance imaging of myelin: technical developments and challenges. \u003cem\u003eQuant. Imaging Med. Surg.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 1186\u0026ndash;1203 (2020).\u003c/li\u003e\n\u003cli\u003eJang, H. \u003cem\u003eet al.\u003c/em\u003e Inversion recovery zero echo time (IR-ZTE) imaging for direct myelin detection in human brain: a feasibility study. \u003cem\u003eQuant. Imaging Med. Surg.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 895\u0026ndash;906 (2020).\u003c/li\u003e\n\u003cli\u003eJang, H. \u003cem\u003eet al.\u003c/em\u003e Improved volumetric myelin imaging in human brain using 3D dual echo inversion recovery‐prepared UTE with complex echo subtraction. \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e \u003cstrong\u003e83\u003c/strong\u003e, 1168\u0026ndash;1177 (2020).\u003c/li\u003e\n\u003cli\u003eJang, H. \u003cem\u003eet al.\u003c/em\u003e Inversion recovery UTE based volumetric myelin imaging in human brain using interleaved hybrid encoding. \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e \u003cstrong\u003e83\u003c/strong\u003e, 950\u0026ndash;961 (2020).\u003c/li\u003e\n\u003cli\u003eMa, Y. \u003cem\u003eet al.\u003c/em\u003e UTE MRI for assessing demyelination in an mTBI mouse model: An open-field low-intensity blast study. \u003cem\u003eNeuroImage\u003c/em\u003e \u003cstrong\u003e310\u003c/strong\u003e, 121103 (2025).\u003c/li\u003e\n\u003cli\u003eJang, H. \u003cem\u003eet al.\u003c/em\u003e Inversion Recovery Ultrashort TE MR Imaging of Myelin is Significantly Correlated with Disability in Patients with Multiple Sclerosis. \u003cem\u003eAm. J. Neuroradiol.\u003c/em\u003e \u003cstrong\u003e42\u003c/strong\u003e, 868\u0026ndash;874 (2021).\u003c/li\u003e\n\u003cli\u003eEhman, R., McNamara, M., Pallack, M., Hricak, H. \u0026amp; Higgins, C. Magnetic resonance imaging with respiratory gating: techniques and advantages. \u003cem\u003eAm. J. Roentgenol.\u003c/em\u003e \u003cstrong\u003e143\u003c/strong\u003e, 1175\u0026ndash;1182 (1984).\u003c/li\u003e\n\u003cli\u003eMaclaren, J., Aksoy, M., Ooi, M. B., Zahneisen, B. \u0026amp; Bammer, R. Prospective motion correction using coil‐mounted cameras: Cross‐calibration considerations. \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e \u003cstrong\u003e79\u003c/strong\u003e, 1911\u0026ndash;1921 (2018).\u003c/li\u003e\n\u003cli\u003eWeick, S. \u003cem\u003eet al.\u003c/em\u003e DC‐gated high resolution three‐dimensional lung imaging during free‐breathing. \u003cem\u003eJ. Magn. Reson. Imaging\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 727\u0026ndash;732 (2013).\u003c/li\u003e\n\u003cli\u003ePark, J., Shin, T., Yoon, S. H., Goo, J. M. \u0026amp; Park, J. A radial sampling strategy for uniform \u003cem\u003ek\u003c/em\u003e ‐space coverage with retrospective respiratory gating in 3D ultrashort‐echo‐time lung imaging. \u003cem\u003eNMR Biomed.\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 576\u0026ndash;587 (2016).\u003c/li\u003e\n\u003cli\u003eLarson, P. E. Z., Conolly, S. M., Pauly, J. M. \u0026amp; Nishimura, D. G. Using adiabatic inversion pulses for long‐ T2 suppression in ultrashort echo time (UTE) imaging. \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e \u003cstrong\u003e58\u003c/strong\u003e, 952\u0026ndash;961 (2007).\u003c/li\u003e\n\u003cli\u003eStehning, C., B\u0026ouml;rnert, P., Nehrke, K., Eggers, H. \u0026amp; Stuber, M. Free‐breathing whole‐heart coronary MRA with 3D radial SSFP and self‐navigated image reconstruction. \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e \u003cstrong\u003e54\u003c/strong\u003e, 476\u0026ndash;480 (2005).\u003c/li\u003e\n\u003cli\u003ePiccini, D., Littmann, A., Nielles‐Vallespin, S. \u0026amp; Zenge, M. O. Spiral phyllotaxis: The natural way to construct a 3D radial trajectory in MRI. \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e \u003cstrong\u003e66\u003c/strong\u003e, 1049\u0026ndash;1056 (2011).\u003c/li\u003e\n\u003cli\u003eFeng, L. Golden‐Angle Radial MRI: Basics, Advances, and Applications. \u003cem\u003eJ. Magn. Reson. Imaging\u003c/em\u003e \u003cstrong\u003e56\u003c/strong\u003e, 45\u0026ndash;62 (2022).\u003c/li\u003e\n\u003cli\u003eLi, C., Gore, J. C. \u0026amp; Davatzikos, C. Multiplicative intrinsic component optimization (MICO) for MRI bias field estimation and tissue segmentation. \u003cem\u003eMagn. Reson. Imaging\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 913\u0026ndash;923 (2014).\u003c/li\u003e\n\u003cli\u003eMacKay, A. \u003cem\u003eet al.\u003c/em\u003e Insights into brain microstructure from the T2 distribution. \u003cem\u003eMagn. Reson. Imaging\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 515\u0026ndash;525 (2006).\u003c/li\u003e\n\u003cli\u003eWeiger, M. \u003cem\u003eet al.\u003c/em\u003e Advances in MRI of the myelin bilayer. \u003cem\u003eNeuroImage\u003c/em\u003e \u003cstrong\u003e217\u003c/strong\u003e, 116888 (2020).\u003c/li\u003e\n\u003cli\u003eMa, Y. \u003cem\u003eet al.\u003c/em\u003e Making the invisible visible\u0026mdash;ultrashort echo time magnetic resonance imaging: Technical developments and applications. \u003cem\u003eAppl. Phys. Rev.\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, (2022).\u003c/li\u003e\n\u003cli\u003eCheng, K. Y. \u003cem\u003eet al.\u003c/em\u003e Clinical application of ultrashort echo time (UTE) and zero echo time (ZTE) magnetic resonance (MR) imaging in the evaluation of osteoarthritis. \u003cem\u003eSkeletal Radiol.\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 2149\u0026ndash;2157 (2023).\u003c/li\u003e\n\u003cli\u003eChang, E. Y., Du, J. \u0026amp; Chung, C. B. UTE imaging in the musculoskeletal system. \u003cem\u003eJ. Magn. Reson. Imaging\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 870\u0026ndash;883 (2015).\u003c/li\u003e\n\u003cli\u003eChan, R. W., Ramsay, E. A., Cheung, E. Y. \u0026amp; Plewes, D. B. The influence of radial undersampling schemes on compressed sensing reconstruction in breast MRI. \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e \u003cstrong\u003e67\u003c/strong\u003e, 363\u0026ndash;377 (2012).\u003c/li\u003e\n\u003cli\u003eHan, M., Chiba, K., Banerjee, S., Carballido-Gamio, J. \u0026amp; Krug, R. Variable flip angle three-dimensional fast spin-echo sequence combined with outer volume suppression for imaging trabecular bone structure of the proximal femur: Reduced-FOV 3D FSE Trabecular Bone MRI. \u003cem\u003eJ. Magn. Reson. Imaging\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 1300\u0026ndash;1310 (2015).\u003c/li\u003e\n\u003cli\u003eSt\u0026ouml;cker, T. \u0026amp; Shah, N. J. MP‐SAGE: A new MP‐RAGE sequence with enhanced SNR and CNR for brain imaging utilizing square‐spiral phase encoding and variable flip angles. \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e \u003cstrong\u003e56\u003c/strong\u003e, 824\u0026ndash;834 (2006).\u003c/li\u003e\n\u003cli\u003eFeng, L. \u003cem\u003eet al.\u003c/em\u003e Golden‐angle radial sparse parallel MRI: Combination of compressed sensing, parallel imaging, and golden‐angle radial sampling for fast and flexible dynamic volumetric MRI. \u003cem\u003eMagn. Reson. Med.\u003c/em\u003e\u003cstrong\u003e72\u003c/strong\u003e, 707\u0026ndash;717 (2014).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7199927/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7199927/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDirect myelin imaging with inversion-recovery ultrashort-echo-time (IR-UTE) is highly motion-sensitive, yet extra hardware or longer scans are impractical. We evaluated whether a superior\u0026ndash;inferior (SI) self-navigator with bit-reversed spoke-angles mitigates motion artifacts without extending acquisition. Dual-echo IR-UTE was implemented at 3T. After an adiabatic inversion pulse, 21 spokes were acquired per segment around the white-matter null point, and dual-echo subtraction suppressed residual long-T2 signals. Adding SI-navigators at the end of each segment allows motion detection without lengthening TR. And bit-reversal is used to pseudo-randomize the spoke-angles. Bloch simulations of a 2D synthetic brain removed 10% of spokes to mimic gating. Three volunteers were scanned: (i) sequential-ordering, no-motion; (ii) bit-reversed, no-motion; (iii) bit-reversed with deliberate head nods. The navigator rejected 1,280 of 12,000 spokes (10.7%) during nodding, and the same gating was reconstructed on motion-free data. Simulation showed coherent streaks for sequential ordering but an isotropic point-spread function for bit-reversal. In vivo, gating degraded only the sequential dataset. bit-reversal preserved subcortical and deep-white-matter detail. During intentional nodding, gating with bit-reversal enhanced myelin contrast, outperforming the image reconstructed from the ungated data. The SI-navigator plus bit-reversal enables effective motion gating without hardware or time penalty, supporting routine motion-robust myelin mapping.\u003c/p\u003e","manuscriptTitle":"Motion-Robust Direct Myelin Imaging in MRI using Self-Gating","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-27 07:03:32","doi":"10.21203/rs.3.rs-7199927/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-09-30T11:14:56+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-24T21:16:48+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-11T18:05:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"56344066354037382814007606645725220445","date":"2025-09-04T11:00:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"249260902220316475100848567518535362439","date":"2025-08-21T14:06:48+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-18T19:12:25+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-12T17:28:40+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-08-05T06:37:57+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-01T23:30:23+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-08-01T23:27:27+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a77e2922-eb6b-4e6d-a998-551f80ece932","owner":[],"postedDate":"August 27th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":53613454,"name":"Biological sciences/Neuroscience"},{"id":53613455,"name":"Physical sciences/Physics"}],"tags":[],"updatedAt":"2026-02-16T16:01:23+00:00","versionOfRecord":{"articleIdentity":"rs-7199927","link":"https://doi.org/10.1038/s41598-026-39238-4","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2026-02-09 15:58:28","publishedOnDateReadable":"February 9th, 2026"},"versionCreatedAt":"2025-08-27 07:03:32","video":"","vorDoi":"10.1038/s41598-026-39238-4","vorDoiUrl":"https://doi.org/10.1038/s41598-026-39238-4","workflowStages":[]},"version":"v1","identity":"rs-7199927","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7199927","identity":"rs-7199927","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}