Real-time AI integration for MR to detect artifacts and guide pulse sequence adaptations

Gudmundson, A. T.; Shams, Z.; Gad, A.; Wang, S.; Simicic, D.; Murali-Manohar, S.; Simegn, G. L.; Özdemir, I.; Davies-Jenkins, C. W.; Yedavalli, V.; Oeltzschner, G.; Demirel, O. B.; Sulam, J.; schär, M.; Ganji, S.; Edden, R. A. E.

doi:10.64898/2026.05.04.722724

bioRxiv · 10.64898/2026.05.04.722724

Real-time AI integration for MR to detect artifacts and guide pulse sequence adaptations

Gudmundson, A. T.·Shams, Z.·Gad, A.·Wang, S.·Simicic, D.·Murali-Manohar, S.·Simegn, G. L.·Özdemir, I.·Davies-Jenkins, C. W.·Yedavalli, V.·Oeltzschner, G.·Demirel, O. B.·Sulam, J.·schär, M.·Ganji, S.·Edden, R. A. E.

View original source ↗Export metadata ↓

Abstract

This work presents a first-of-its-kind artificial intelligence (AI-)integrated MR pulse sequence that detects out-of-voxel (OOV) artifacts in real-time (within-TR) and responds prospectively by updating the crusher gradient scheme. Per Excitation Real-time Execution & Guided Responses with Integrated Neural-network Evaluation (PEREGRINE), allows for deployment of deep learning models and pulse sequence updates. In this study, PEREGRINE operated a time-domain (TD) and frequency-domain (FD) convolutional autoencoder that detect OOV artifacts. Scans without (AI-off) and with (AI-on) updates were collected from the medial prefrontal cortex of healthy volunteers using a MEGA-edited MRS experiment. The degree of OOV contamination (OOV Score) was quantified per transient based upon the prevalence of OOV signals in the TD and FD data. OOV Scores above a user-defined threshold triggered an update of the crusher gradient scheme, iterating through 48 permutations (6 axis transpositions x 8 polarity flips). Within each 2-second TR, PEREGRINE successfully provided single-transient OOV Scores and updated gradients accordingly. No difference was observed between the OOV Scores from the full ("Full" condition) AI-on and AI-off sessions due to the AI-on scan cycling over better and worse gradient permutations relative to the AI-off scan. However, the AI-on scan had significantly lower OOV Scores than the AI-off scan when selecting the transients where PEREGRINE persisted ("Dwell" condition) on a given gradient permutation. Ultimately, Fit Quality Number (FQN) from linear combination modeling improved significantly for the AI-on compared to the AI-off scan. PEREGRINE enabled an adaptive and AI-integrated sequence allowing for real-time evaluation and response to OOV artifacts, identifying gradient modifications that produced less OOV contamination. Graphical AbstractO_ST_ABSGraphical Abstract SummaryC_ST_ABSPEREGRINE, Per Excitation Real-time Execution & Guided Responses with Integrated Neural-network Evaluation, allows for pulse sequence updates using on-scanner deployed neural networks. To demonstrate the utility, PEREGRINE was used to operate two neural networks (time/frequency domain), trained to detect out-of-voxel artifacts, during a MEGA-edited MRS experiment in the prefrontal cortex. When artifacts were detected, PEREGRINE updated the crusher gradients within the same TR. Real-time, adaptive, and AI-driven MR will provide a long-awaited solution for combatting artifacts and poor data quality. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=167 SRC="FIGDIR/small/722724v2_ufig1.gif" ALT="Figure 1"> View larger version (38K): org.highwire.dtl.DTLVardef@cc01e0org.highwire.dtl.DTLVardef@18f4083org.highwire.dtl.DTLVardef@1d48c8corg.highwire.dtl.DTLVardef@1574ba7_HPS_FORMAT_FIGEXP M_FIG Graphical Abstract Figure C_FIG

Explore related subjects

neuroscience Preprint

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Gudmundson, A. T., Shams, Z., Gad, A., Wang, S., Simicic, D., Murali-Manohar, S., Simegn, G. L., Özdemir, I., Davies-Jenkins, C. W., Yedavalli, V., Oeltzschner, G., Demirel, O. B., Sulam, J., schär, M., Ganji, S., Edden, R. A. E.. 2026-05-07. Real-time AI integration for MR to detect artifacts and guide pulse sequence adaptations. https://doi.org/10.64898/2026.05.04.722724

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

Neurodegeneration-inducing macromolecules exit the brain via nanovascular conduits formed by reticular fibroblasts

Accumulation of proteins such as amyloid beta (Abeta), hyperphosphorylated tau and alpha-synuclein within the brain alters neural information processing and causes neurodegeneration(1-3), but how toxic solutes are cleared from the brain remains highly controversial(4,5). Proposed exit routes include efflux across endothelial cells into the blood(6,7), and movement to the pial surface via vasomotion-induced pumping along spaces within arteriolar smooth muscle(8) or via outflow along the perivascular space of ascending venules promoted by water flux through astrocytes (the glymphatic system(9)). From the pial surface of the brain, drainage may continue to dural lymphatics, along the outer sheaths of exiting cranial nerves and across the cribriform plate(10-14). We now report the presence, in mice and humans, of 2 micron diameter conduits that remove fluorescently labelled tau and Abeta from the brain. These conduits form a spatially-organised mesh within the walls of penetrating arterioles and pial arteries, and around the surface of ascending venules and deep cerebral and pial veins. They course through the pial and arachnoid layers to span the CSF space, wrapping the brain and cranial nerves. They are formed of reticular fibroblasts, which label for VE-cadherin(15) and PDGFRalpha(16), the lymphatic markers(17) podoplanin, VEGFR3 and Prox1, and reticular fibroblast extracellular matrix components collagen I and VI(16,18-20). Parenchymal tau drains from the brain at a similar rate via arteriolar conduits and via conduits around venules, arguing against preferential removal by a glymphatic mechanism. In Alzheimer's disease model mice, Abeta is seen traversing these lymph node-like conduits. Modulation of molecular transfer via this route may accelerate or delay cognitive decline, and slowed transfer from arteriolar to pial-arachnoid conduits may initiate cerebral amyloid angiopathy.

neuroscience↗

bioRxiv2026

Analysis of the influence of gradual changes in matrix sentence similarity on neural envelope tracking

Neural tracking of speech is a well-established phenomenon in neuroscience. However, for speech signals with a fixed structure, significant correlations between speech envelopes and neurophysiological representations occur even for unheard sentences. We exploit a structured speech-in-noise matrix hearing test (Oldenburger Sentence Test, OLSA) to systematically quantify the relationship between acoustic sentence similarity and neural tracking. Simultaneous magnetoencephalography (MEG) and 76-channel electroencephalography (EEG) data, including 16 channels positioned directly around the ears (ear-EEG), were recorded from 21 young adults with normal hearing during the presentation of clean-speech audiobooks and OLSA sentences at six signal-to-noise ratios. A linear decoder trained on audiobooks reconstructed OLSA sentence envelopes. Reconstruction accuracies were compared using a linear mixed model across heard (matched) and unheard (mismatched) sentences of varying acoustic similarity. Significant reconstruction accuracies were achieved across MEG, EEG, and ear-EEG for both matched and mismatched sentences. For mismatched sentences, these accuracies gradually increased with their acoustic similarity to the heard speech data. The high similarity between sentences, which is especially prominent in matrix tests, can cause significant spurious tracking for mismatched stimuli. This effect can reach levels comparable to those of matched sentences and can be mistaken for true neural tracking. Robust neural tracking across modalities further supported the established viability of ear-EEG compared to whole-head systems.

neuroscience↗

bioRxiv2026

Seizures and tauopathy following neurotrauma are mediated by prion protein and metabotropic glutamate receptor 5

Traumatic brain injury (TBI) is one of the world's leading causes of death and disability and a major risk factor for dementias. The primary dementia associated with TBI is chronic traumatic encephalopathy (CTE), a neurodegenerative disease classified as a tauopathy, in which toxic tau molecules lead to disease pathologies and degeneration. The processes that lead to tauopathy and subsequent dementia after TBI remain unclear. Here, we built upon the finding that seizures after TBI may be a mechanism leading to tauopathy, by dissecting the functions of the metabotropic glutamate receptor 5 - cellular prion protein (mGluR5-PrPC) pathway. We delivered TBI to larval in a blast paradigm, and quantified aggregation of Tau via a genetically-encoded Tau-GFP fusion reporter. Zebrafish larvae lacking prp2 (homolog of mammalian cellular Prion Protein, PrPC) displayed a 168% increase in post-traumatic seizures activity after TBI. An mGluR5 agonist (CHPG) reduced post-traumatic seizures, whereas an mGluR5 antagonist (MPEP) increased post-traumatic seizures. Moreover, agonizing mGluR5 reduced tau aggregation and antagonizing mGluR5 increased tau burden. Larvae seizing from convulsants, rather than TBI, were treated with CHPG/MPEP and provided a similar pattern of outcomes, suggesting seizures may be a factor needed for mGluR5 activity to influence tau aggregation. The PrPC-mGluR5 pathway is proposed as one candidate pathomechanism linking TBI to subsequent seizures and tauopathy, and thus it warrants investigation as a target for prophylactic interventions.

neuroscience↗