Journal of neural engineering最新文献

{"title":"Real-time detection of spoken speech from unlabeled ECoG signals: a pilot study with an ALS participant.","authors":"Miguel Angrick, Shiyu Luo, Qinwan Rabbani, Shreya Joshi, Daniel N Candrea, Griffin W Milsap, Chad R Gordon, Kathryn Rosenblatt, Lora Clawson, Nicholas Maragakis, Francesco V Tenore, Matthew S Fifer, Nick F Ramsey, Nathan E Crone","doi":"10.1088/1741-2552/ae0965","DOIUrl":"10.1088/1741-2552/ae0965","url":null,"abstract":"<p><p><i>Objective</i>. Brain-computer interfaces hold significant promise for restoring communication in individuals with partial or complete loss of the ability to speak due to paralysis from amyotrophic lateral sclerosis (ALS), brainstem stroke, and other neurological disorders. Many of the approaches to speech decoding reported in the BCI literature have required time-aligned target representations to allow successful training-a major challenge when translating such approaches to people who have already lost their voice.<i>Approach</i>. In this pilot study, we made a first step toward scenarios in which no ground truth is available. We utilized a graph-based clustering approach to identify temporal segments of speech production from electrocorticographic (ECoG) signals alone. We then used the estimated speech segments to train a voice activity detection (VAD) model using only ECoG signals. We evaluated our approach using a leave-one-day-out cross-validation on open-loop recordings of a single dysarthric clinical trial participant living with ALS, and we compared the resulting performance to previous solutions trained with ground truth acoustic voice recordings.<i>Main results</i>. Our approach achieves a median timing error of around 530 ms with respect to the actual spoken speech. Embedded into a real-time BCI, our approach is capable of providing VAD results with a latency of only 10 ms.<i>Significance</i>. To the best of our knowledge, our results show for the first time that speech activity can be predicted purely from unlabeled ECoG signals, a crucial step toward individuals who cannot provide this information anymore due to their neurological condition, such as patients with locked-in syndrome.<i>Clinical Trial Information</i>. ClinicalTrials.gov, registration number NCT03567213.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12498269/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145093312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Decoding saccadic eye movements from brain signals using an endovascular neural interface.","authors":"Suleman Rasheed, James Bennett, Peter E Yoo, Anthony N Burkitt, David B Grayden","doi":"10.1088/1741-2552/ae0f52","DOIUrl":"https://doi.org/10.1088/1741-2552/ae0f52","url":null,"abstract":"<p><strong>Objective: </strong>An Oculomotor Brain-Computer Interface (BCI) records neural activity from brain regions involved in planning eye movements and translates this activity into control commands. While previous successful studies have relied on invasive implants in non-human primates or electrooculography (EOG) artefacts in human electroencephalogram (EEG) data, this study aimed to demonstrate the feasibility of an oculomotor BCI using a minimally invasive endovascular StentrodeTM device implanted near the supplementary motor area of a patient with Amyotrophic Lateral Sclerosis (ALS).&#xD;Approach. One participant performed self-paced visually-guided and free-viewing saccade tasks in four directions (left, right, up, down) while endovascular EEG and eye gaze recordings were collected. Visually-guided saccades were cued with visual stimuli, whereas free-viewing saccades were self-directed without explicit cues. Brain signals were pre-processed to remove cardiac artefacts, downsampled,&#xD;and classified using a Random Forest algorithm. For saccade onset classification (fixation vs. saccade), features in time and frequency domains were extracted after xDAWN denoising, while for saccade direction classification, the downsampled time series were classified directly without explicit feature extraction. &#xD;Main results. The neural responses of visually-guided saccades overlapped with cue-evoked potentials, while free-viewing saccades exhibited saccade-related potentials that began shortly before eye movement, peaked approximately 50 ms after saccade onset, and persisted for around 200 ms. In the frequency domain, these responses appeared as a low-frequency synchronisation below 15 Hz. Saccade onset classification was robust, achieving mean area under the receiver operating characteristic curve (AUC) scores of 0.88 within sessions and 0.86 across sessions. Saccade direction decoding yielded within-session AUC scores of 0.67 for four-class decoding and up to 0.75 for the best performing binary comparisons (left vs. up and left vs. down).&#xD;Significance. This proof-of-concept study demonstrates the feasibility of an endovascular oculomotor BCI in a patient with ALS, establishing a foundation for future oculomotor BCI studies in human subjects.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145226571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Co-cultured sensory neuron classification using extracellular electrophysiology and machine learning approaches for enhancing analgesic screening.","authors":"Alexander Somers, Bryan James Black","doi":"10.1088/1741-2552/ae0eef","DOIUrl":"https://doi.org/10.1088/1741-2552/ae0eef","url":null,"abstract":"<p><p>Chronic pain affects over 20% of the adult population in the United States, posing a substantial economic burden and contributing to the ongoing opioid crisis. Effective, non-addictive chronic pain treatments are urgently needed. Traditional drug discovery methods have failed to identify novel, non-addictive compounds, highlighting the need for alternative approaches such as phenotypic screening. Our lab developed a phenotypic screening assay using extracellular electrophysiological recordings from co-cultures of human induced pluripotent stem cell (hiPSC) sensory neurons and glia. This study aimed to identify responsive neuronal subtypes within these presumptively heterogeneous cultures. We induced an inflammation-like state using TNF-α and evaluated acute responses to nociceptor agonists/antagonist capsaicin and PF-05089771, which target TRVP1, and Nav1.7, respectively. By employing unsupervised learning, we labeled responsive cells based on changes in spike count and synchrony. We then used the labeled cells' baseline activity data to train and validate five classifiers, finding that a feed forward neural network yielded error values that were the most significant following pair-wise comparisons. The classifier achieved an 84% accuracy for classifying nociceptors in an unseen labeled culture. The notable accuracy suggests that machine learning techniques could be employed to enhance MEA-based neuronal phenotypic assays as cellular readouts (i.e. mean-firing rates) can be weighted based on a desired target cell (i.e. nociceptor).&#xD.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145214854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An interpretable generative multimodal neuroimaging-genomics framework for decoding Alzheimer's disease.","authors":"Giorgio Dolci, Federica Cruciani, Md Abdur Rahaman, Anees Abrol, Jiayu Chen, Zening Fu, Ilaria Boscolo Galazzo, Gloria Menegaz, Vince D Calhoun","doi":"10.1088/1741-2552/ae087d","DOIUrl":"10.1088/1741-2552/ae087d","url":null,"abstract":"<p><p><i>Objective.</i>Alzheimer's disease (AD) is the most prevalent form of dementia worldwide, encompassing a prodromal stage known as mild cognitive impairment (MCI), where patients may either progress to AD or remain stable. The objective of the work was to capture structural and functional modulations of brain structure and function relying on multimodal MRI data and single nucleotide polymorphisms, also in case of missing views, with the twofold goal of classifying AD patients versus healthy controls and detecting MCI converters.<i>Approach.</i>We propose a multimodal deep learning (DL)-based classification framework where a generative module employing cycle generative adversarial networks was introduced in the latent space for imputing missing data (a common issue of multimodal approaches). Explainable AI method was then used to extract input features' relevance allowing for post-hoc validation and enhancing the interpretability of the learned representations.<i>Main results.</i>Experimental results on two tasks, AD detection and MCI conversion, showed that our framework reached competitive performance in the state-of-the-art with an accuracy of0.926±0.02(CI [0.90, 0.95]) and0.711±0.01(CI [0.70, 0.72]) in the two tasks, respectively. The interpretability analysis revealed gray matter modulations in cortical and subcortical brain areas typically associated with AD. Moreover, impairments in sensory-motor and visual resting state networks along the disease continuum, as well as genetic mutations defining biological processes linked to endocytosis, amyloid-beta, and cholesterol, were identified.<i>Significance.</i>Our integrative and interpretable DL approach shows promising performance for AD detection and MCI prediction while shedding light on important biological insights.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12481582/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145082817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Personalized whole-brain models of seizure propagation.","authors":"Edmundo Lopez-Sola, Borja Mercadal, Èlia Lleal-Custey, Ricardo Salvador, Roser Sanchez-Todo, Fabrice Wendling, Fabrice Bartolomei, Giulio Ruffini","doi":"10.1088/1741-2552/ae08e9","DOIUrl":"10.1088/1741-2552/ae08e9","url":null,"abstract":"<p><p><i>Objective.</i>Computational modeling has recently emerged as a powerful tool to better understand seizure dynamics and guide new treatment strategies. This work aims to develop and personalize whole-brain computational models in epilepsy using multimodal clinical data to simulate and evaluate individualized therapeutic strategies.<i>Approach.</i>We present a computational framework that constructs patient-specific whole-brain models of seizure propagation by integrating SEEG, MRI, and diffusion MRI data. The pipeline uses neural mass models for each node in the network, simulating whole-brain dynamics. Model personalization involves adjusting global and local parameters representing the excitability of individual brain areas, using an evolutionary algorithm that aims to maximize the correlation between empirical and synthetic functional connectivity matrices derived from SEEG data.<i>Main results.</i>The resulting personalized models successfully reproduce individual seizure propagation patterns and can be used to simulate therapeutic interventions like surgery, stimulation, or pharmacological interventions within a unified physiological framework. Notably, model predictions reveal distinct patient-specific responses across interventions, including variable sensitivity to different pharmacological agents and identification of critical regions whose removal or modulation reduced seizure spread.<i>Significance.</i>This framework provides a mechanistic, interpretable approach to simulate and compare individualized treatment strategies. By integrating multimodal data into a unified whole-brain model, it has the potential to improve clinical decision-making in epilepsy by identifying accessible and functionally relevant targets.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145088663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"On carrier frequency in transcutaneous spinal cord electrical stimulation: a narrative review.","authors":"Natalia Shamantseva, Tatiana Moshonkina","doi":"10.1088/1741-2552/ae08e8","DOIUrl":"10.1088/1741-2552/ae08e8","url":null,"abstract":"<p><p><i>Objective.</i>Transcutaneous spinal cord stimulation (tSCS) using kilohertz frequency carrier modulation has emerged as a non-invasive neuromodulation approach to improve motor recovery and reduce pain. Early application of 5-10 kHz modulated pulses for tSCS has shown promising results in spinal cord (SC) injury and post-stroke rehabilitation, but the mechanisms underlying these effects remain poorly understood.<i>Approach.</i>This narrative review synthesizes electrophysiological, computational and clinical evidence to assess how kilohertz modulation influences spinal and corticospinal excitability and analgesia. A total of 20 preclinical and clinical studies comparing the effects of kHz-modulated and conventional stimulation pulses were reviewed.<i>Main results.</i>The results indicate that kilohertz modulated tSCS increases tolerance to stimulation, but requires a higher charge to evoke motor responses in healthy participants and individuals with post-stroke motor disorder. Compared to conventional stimulation, modulated stimulation recruits afferents less efficiently at motor threshold intensity but appears to engage broader corticospinal circuits, especially near or below threshold. Frequency-specific effects include prolonged spinal inhibition, frequency-dependent modulation of supraspinal input, and selective activation of inhibitory interneurons in the dorsal horn. Computational study supports these observations, showing that kilohertz pulses produce delayed action potential initiation due to alternating depolarization cycles. A comparative functional study has shown that modulated tSCS improves motor function in individuals with SC injury more significantly than conventional stimulation.<i>Significance.</i>This narrative review highlights gaps in our understanding of the mechanisms of modulated tSCS, suggests directions for further research and will be useful in planning studies on the mechanisms behind tSCS with and without carrier frequency. It also holds engineering relevance for the optimal design of stimulation devices.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145088592","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Helical neural implants for intracerebral drug delivery.","authors":"Batoul Khlaifat, Mahmoud Elbeh, Shreya Manjrekar, Seung-Jean Kang, Yusheng Zhang, Parima Phowarasoontorn, Sadaf Usmani, Abdel-Hameed Dabbour, Heba T Naser, Hanan Mohammed, Minsoo Kim, Khalil B Ramadi","doi":"10.1088/1741-2552/ae0523","DOIUrl":"https://doi.org/10.1088/1741-2552/ae0523","url":null,"abstract":"<p><p><i>Objective.</i>Neurological disorders often arise from specific regions of dysfunction in the brain. One approach to target these pathologic regions is through chemical delivery using intracerebral implants. Previous works have designed implants that are small and flexible, minimizing the mechanical mismatch between inorganic implants and soft organic brain tissue. Most of these implants are simple cylindrical catheters with inflow and outflow ports at either end of the cylinder. This limits the region and volume of tissue that can be dosed. We sought to develop novel catheter designs that permit targeting of larger volumes of brain tissue while maintaining minimal footprint to minimize gliosis.<i>Approach.</i>We present the design, fabrication, and testing of a novel helical-shaped microfluidic catheter we term SPIRAL (Strategic Precision Infusion for Regional Administration of Liquid). SPIRAL leverages rational fluidic design of multiple fluid outflow ports to vary infused fluid spatial distribution across brain regions. We used<i>in silico, in vitro, and in vivo</i>models to test the fluid dynamic functionality and chronic viability of SPIRAL.<i>Results.</i>Our computational fluid dynamics (CFDs) models illustrate how SPIRAL can be configured to permit simultaneous dosing through multiple outflow ports yielding a variable fluid distribution compared to a straight catheter. We show how CFD<i>in silico</i>models can be used to optimize dimensions of channel openings across SPIRAL, to achieve uniform flow through channels and validate these results<i>in vitro</i>. We show how chronically implanted SPIRAL catheters do not increase gliosis compared to standard straight catheters of similar dimensions or materials.<i>Significance.</i>Our helical intracerebral drug delivery catheter facilitates fluid localization while maintaining minimal invasiveness. SPIRAL could enable multiregional brain access and improve therapeutic efforts in the treatment of neurological diseases.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":"22 5","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145188059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"E-Sort: empowering end-to-end neural network for multi-channel spike sorting with transfer learning and fast post-processing.","authors":"Yuntao Han, Shiwei Wang","doi":"10.1088/1741-2552/ae0d33","DOIUrl":"https://doi.org/10.1088/1741-2552/ae0d33","url":null,"abstract":"<p><strong>Objective: </strong>Spike sorting, which involves detecting and attributing spikes to their putative neurons from extracellular recordings, is a common process in electrophysiology and brain-computer interface systems. Recent advances in large-scale neural recording technologies are challenging the conventional algorithms because of the intensive computational workloads required and the accuracy degradation suffered from time-variant spike patterns and significant levels of noise. Neural networks have demonstrated promising performance in processing these large-scale neural recordings. However, their applications are constrained by the labor-intensive data labeling and the lack of fully vectorised frameworks with end-to-end neural networks.</p><p><strong>Approach: </strong>We propose E-Sort, an end-to-end neural network-based spike sorter with transfer learning and parallelizable post-processing to address both obstacles.</p><p><strong>Main results: </strong>We examined our framework in both synthetic and real datasets. The results of the processing of the synthetic datasets show that our approach can reduce the number of annotated spikes required for training by 44% compared to training from scratch, achieving up to 25.7% higher accuracy. We evaluated E-Sort on various probe geometries, noise levels, and drift patterns, which demonstrates that our design can achieve an accuracy that is comparable with Kilosort4 while sorting 50 seconds of data in only 1.32 seconds. To test with real datasets, we first sorted the spikes using Kilosort4 and used the sorted spikes at the initial period to pre-train the neural network; then we compared and measured the agreement between the results from the trained model and those from Kilosort4. On average the pre-training process improved the result agreement by 30% approximately.</p><p><strong>Significance: </strong>E-Sort offers a scalable, efficient, and accurate neural network-based framework for large-scale spike sorting, significantly reducing manual labelling effort and processing time.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145194363","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A methodological framework for the efficient characterization of peripheral nerve stimulation parameters.","authors":"Rachel S Jakes, Benjamin J Alexander, Vlad I Marcu, A Bolu Ajiboye, Dustin J Tyler","doi":"10.1088/1741-2552/ae0d31","DOIUrl":"10.1088/1741-2552/ae0d31","url":null,"abstract":"<p><strong>Objective: </strong>Restoring movement and somatosensation with peripheral nerve stimulation (PNS) requires precise neural activation. Because pulse amplitude (PA) and pulse width (PW) recruit axons differently, intentionally modulating both could enable more complex PNS. However, mapping the PA-PW space is currently prohibitively time-intensive. This paper proposes and clinically validates an efficient method to characterize multiple intensities in the PA-PW space for motor and perceptual sensory applications using minimal data collection.</p><p><strong>Approach: </strong>We used cuff electrodes implanted in one participant with a spinal cord injury to generate iso-EMG activation contours and two participants with upper limb loss to generate somatosensory perceptual iso-intensity contours in the PA-PW space. Strength-duration (SD) curves were mapped to the contours using varying sample point subsets and assessed for fit quality. Finite element modeling of a human nerve and activation simulations evaluated differences in recruited axon populations across the PA-PW space.</p><p><strong>Main results: </strong>SD curves accurately fit all levels of motor activation and perceptual intensity (median R^2 = 0.996 and 0.984, respectively). Reliable estimates of SD curves at any intensity require only two sufficiently-spaced points (motor R2 = 0.991, sensory R2 = 0.977). Using this data, we present and validate a novel method for efficiently characterizing the PA-PW space using SD curves, including a metric that quantifies mapping accuracy based on two sampled points. In silico, intensity-matched high-PW and high-PA stimulation recruited overlapping, but not equivalent, axon sets, with high-PA stimuli preferentially recruiting large-diameter fibers and axons farther from the contact.</p><p><strong>Significance: </strong>This method enables rapid, accurate mapping of the stimulation parameter space for clinical motor and sensory PNS. The efficiency of the proposed characterization approach enhances the clinical feasibility of multiparameter modulation, establishing a framework for further exploration of two-parameter modulation for increased selectivity and resolution, reduced fatigue, and unique percept generation. (ClinicalTrials.gov ID NCT03898804).</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145194343","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From zero- to few-shot: deep temporal learning of wrist EMG enables scalable cross-user gesture recognition.","authors":"Fady S Botros, Heather E Williams, Angkoon Phinyomark, Erik J Scheme","doi":"10.1088/1741-2552/ae08eb","DOIUrl":"10.1088/1741-2552/ae08eb","url":null,"abstract":"<p><p><i>Objective.</i>Wrist electromyography (EMG) is emerging as an enticing wearable input modality for human-machine interaction. Traditionally recorded from the forearm for use in transradial prostheses, wrist-based EMG sensors are now being integrated into devices such as watches and wristbands for hand gesture recognition (HGR). Consumer familiarity with wrist-worn devices makes wrist EMG a compelling option, but the need for individualized user calibration remains a challenge.<i>Approach.</i>This study therefore evaluated various cross-user models to reduce the calibration burden and compared wrist- and forearm-based models. Eight different machine learning architectures were evaluated across 33 users, using varying amounts of data from the end user.<i>Main results.</i>A temporal convolutional network-bidirectional long short-term memory architecture, applied for the first time to EMG classification, was found to significantly (<i>p</i> < 0.05) outperform other tested machine learning architectures. An inter-day feature set combined with<i>Z</i>-score normalization achieved the best performance when classifying five gestures (plus a rest class) using either wrist or forearm EMG. Consistent with other recent results, wrist EMG consistently outperformed forearm EMG in all analyses, including within- and across-user comparisons (<i>p</i> < 0.05). In cross-user models, wrist EMG demonstrated a zero-shot performance of 78.2%, compared to 71.6% for forearm EMG (<i>p</i> < 0.05). Introducing one calibration repetition from the end user increased one-shot performance of wrist EMG to 91.6%, compared to 86.9% for forearm EMG (<i>p</i> < 0.05). Adding further training repetitions boosted wrist EMG performance to 98.3%, compared to 97.4% for forearm EMG.<i>Significance.</i>These findings provide new evidence supporting the viability of wrist EMG for cross-user HGR models that generalize to new users with minimal calibration, suggesting promising potential for its broader adoption in wearable devices.</p>","PeriodicalId":94096,"journal":{"name":"Journal of neural engineering","volume":" ","pages":""},"PeriodicalIF":3.8,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145088512","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}