Scholarlink

<p>Since the advent of computing, humans have sought computer input technologies that are expressive, intuitive and universal. While diverse modalities have been developed, including keyboards, mice and touchscreens, they require interaction with a device that can be limiting, especially in on-the-go scenarios. Gesture-based systems use cameras or inertial sensors to avoid an intermediary device, but tend to perform well only for unobscured movements. By contrast, brain–computer or neuromotor interfaces that directly interface with the body’s electrical signalling have been imagined to solve the interface problem<sup><CitationRef CitationID="CR1">1</CitationRef></sup>, but high-bandwidth communication has been demonstrated only using invasive interfaces with bespoke decoders designed for single individuals<sup><CitationRef AdditionalCitationIDS="CR3" CitationID="CR2">2</CitationRef>–<CitationRef CitationID="CR4">4</CitationRef></sup>. Here, we describe the development of a generic non-invasive neuromotor interface that enables computer input decoded from surface electromyography (sEMG). We developed a highly sensitive, easily donned sEMG wristband and a scalable infrastructure for collecting training data from thousands of consenting participants. Together, these data enabled us to develop generic sEMG decoding models that generalize across people. Test users demonstrate a closed-loop median performance of gesture decoding of 0.66 target acquisitions per second in a continuous navigation task, 0.88 gesture detections per second in a discrete-gesture task and handwriting at 20.9 words per minute. We demonstrate that the decoding performance of handwriting models can be further improved by 16% by personalizing sEMG decoding models. To our knowledge, this is the first high-bandwidth neuromotor interface with performant out-of-the-box generalization across people.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A generic non-invasive neuromotor interface for human-computer interaction

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月23日

查看全文
  • Patrick Kaifosh,
  • Thomas R. Reardon,
  • Brian D. Allen,
  • Chris Anderson,
  • Sacha Arnoud,
  • Rahul Arora,
  • Mridu Atray,
  • Lana Awad,
  • Francisco Ayerbe,
  • Christopher Baker,
  • Nicholas Baker,
  • Alexandre Barachant,
  • Philip Bard,
  • Wilman Pimentel Beltran,
  • Adam Berenzweig,
  • Rohin Bhasin,
  • Joe Bienkowski,
  • Sean Bittner,
  • Luke Boegner,
  • Anu Bolarinwa,
  • Don Bosley,
  • Matthew Bracaglia,
  • Mario Bräcklein,
  • Maclyn Brandwein,
  • Joe Bravate,
  • Matt Butler,
  • Adam J. Calhoun,
  • Chia-Jung Chang,
  • Daniel Chenet,
  • Joshua Chester,
  • Rudi Chiarito,
  • Rohan Chitnis,
  • John Choi,
  • Won Chun,
  • Jeremiah Chung,
  • James Connors,
  • Jota Costa,
  • Mark Cramer,
  • Raven Cunningham,
  • William F. Cusack,
  • Nathan Danielson,
  • Thomas J. Davidson,
  • Bruno De Araujo,
  • Bob DiMaiolo,
  • Scott Draves,
  • Alan Du,
  • Zaina Edelson,
  • Phina Enemuo,
  • Mina Fahmi,
  • Nariman Farsad,
  • Ali Farshchian,
  • Randy Feliz,
  • Jake Fine,
  • Emanuele Formento,
  • Dustin Freeman,
  • Jianing Fu,
  • Jean-Christophe Gagnon-Audet,
  • Rupesh Gajurel,
  • Jonathan Gamutan,
  • Sida Gao,
  • Jonateal Garcia,
  • Nathalie Therese Helene Gayraud,
  • Minha Ghani,
  • Sayan Ghosh,
  • Vickram Gidwani,
  • Danny Giebisch,
  • Greg Gimler,
  • Alexandre Gramfort,
  • Lauren Grosberg,
  • Bryn Gunther,
  • Ning Guo,
  • Chetan Gupta,
  • Sinem Guven Kaya,
  • Austin Ha,
  • Katarina Hadjer,
  • Carlos Xavier Hernández,
  • Stav Hertz,
  • Carl Hewitt,
  • Daniel N. Hill,
  • Kirak Hong,
  • Lillian Hong,
  • Helen Hou,
  • Stepan Hruda,
  • Alex Hsieh,
  • Vivian Hsiung,
  • Rongqing Huang,
  • Yue Hui,
  • Hazel Hulet,
  • Shaker Islam,
  • Vinay Jayaram,
  • Connie Jiang,
  • Xiaodong Jiang,
  • Brooke Juarez,
  • James Jaeyoon Jun,
  • Na Young Jun,
  • Nirag Kadakia,
  • Nishant Kakar,
  • Ajay Kamdar,
  • Ta-Chu Kao,
  • Steven Kober,
  • TW Koh,
  • Christina Shabu Koshy,
  • Andrzej Lawn,
  • Claire Lee,
  • Jennifer Lee,
  • JinHyung Lee,
  • Juheui Amy Lee,
  • Tiffanie Li,
  • Jonathan Liao,
  • Yingru Liu,
  • Yuxuan Liu,
  • Saar Lively,
  • Kati London,
  • Roddy Louie,
  • Francisco Luongo,
  • Attila Maczak,
  • Niru Maheswaranathan,
  • Michael Mandel,
  • Jesse Marshall,
  • Najja Marshall,
  • Mirek Martincik,
  • Nicolas Yvan Masse,
  • Stephen McAnearney,
  • Ashley McHugh,
  • Jorge Aurelio Menendez,
  • Josh Merel,
  • David Miller,
  • Ilya Milyavskiy,
  • Ricardo Pio Monti,
  • Sean Moore,
  • Yonathan Morin,
  • Brock Morrell,
  • Dano Morrison,
  • Anthony Moschella,
  • Suman Mulumudi,
  • Conner Muth,
  • Krunal Naik,
  • Norris Nakagaki,
  • Ajay Nathan,
  • Romario Nelson,
  • Jimson Ngeo,
  • Keven Nguyen,
  • Luke O’Connor,
  • Shay Ohayon,
  • Garrick Orchard,
  • Chris Osborn,
  • Timothy M. Otchy,
  • Emmanuella Owolabi,
  • Adam M. Packer,
  • Tejaswy Pailla,
  • Julia Paredes,
  • Sean Parker,
  • Diogo Peixoto,
  • Matias Perez,
  • Zavion Perez,
  • Adrien Piérard,
  • Stephen M. Plaza,
  • Natalie Plotkin,
  • Eftychios Pnevmatikakis,
  • Brandon Pool,
  • Shanil Puri,
  • Sunaina Rajani,
  • Jose Ramirez Fuentes,
  • Julian Ramos Rojas,
  • Tanvi Ranjan,
  • Devin Reardon,
  • Jonathan Reid,
  • Jason Reisman,
  • Lain Warawao Nemo Mora y Rivera,
  • Sebi Rolotti,
  • Andrew Rosenkranz,
  • Ian Roth,
  • Likhon Roy,
  • Ran Rubin,
  • Alexander Rudnicki,
  • Sam Russell,
  • Abby Russo,
  • James Sacra,
  • Amir Sadoughi,
  • Roxanna Salim,
  • Aichatou Savane,
  • Collin Schlager,
  • David Schwab,
  • Jeffrey Seely,
  • Mike Seltzer,
  • Nurettin Dorukhan Sergin,
  • Ami Shah,
  • Anish Shah,
  • Philip Shamash,
  • Vandita Sharma,
  • Stephie Shen,
  • Kevin Shi,
  • Olivia Shiah,
  • Yasmin Siahpoosh,
  • Noor Siddiqi,
  • Jeremy Simpson,
  • Gagandip Singh,
  • Viswanath Sivakumar,
  • Jeff Smith,
  • Seyyid Emre Sofuoglu,
  • Ivy Jiyoung Song,
  • Morgan Springer,
  • Adrian Spurr,
  • Fabio Stefanini,
  • Connor Stout,
  • Emanuel Strauss,
  • Swetha Suresh,
  • Ananya Suri,
  • David Sussillo,
  • Ziyi Tang,
  • Vikram Tank,
  • Jesslyn Tannady,
  • Aliqyan Tapia,
  • Tugce Tasci,
  • Tiberiu Tesileanu,
  • Aman Tiwari,
  • Anoushka Tiwari,
  • Calvin Tong,
  • Blizelle Tormis,
  • Julia Trabulsi,
  • Migmar Tsering,
  • Kyle Urquhart,
  • Peter Walkington,
  • Megan Wang,
  • Renxiong Wang,
  • Zhuo Wang,
  • Christy Warden,
  • Richard Warren,
  • Claire L. Warriner,
  • Ron J. Weiss,
  • Daniel Z. Wetmore,
  • Ezri White,
  • Christopher Wiebe,
  • Steve Williams,
  • Yuguan Xing,
  • Chris Ye,
  • Akshay Yembarwar,
  • Shuibenyang Yuan,
  • Michael Zawadzki,
  • Mingrui Zhang,
  • Jiesi Zhao,
  • Kevin Zheng,
  • Joseph Zhong,
  • Lei Zhou,
  • Danny Zlobinsky

摘要

Since the advent of computing, humans have sought computer input technologies that are expressive, intuitive and universal. While diverse modalities have been developed, including keyboards, mice and touchscreens, they require interaction with a device that can be limiting, especially in on-the-go scenarios. Gesture-based systems use cameras or inertial sensors to avoid an intermediary device, but tend to perform well only for unobscured movements. By contrast, brain–computer or neuromotor interfaces that directly interface with the body’s electrical signalling have been imagined to solve the interface problem1, but high-bandwidth communication has been demonstrated only using invasive interfaces with bespoke decoders designed for single individuals2–4. Here, we describe the development of a generic non-invasive neuromotor interface that enables computer input decoded from surface electromyography (sEMG). We developed a highly sensitive, easily donned sEMG wristband and a scalable infrastructure for collecting training data from thousands of consenting participants. Together, these data enabled us to develop generic sEMG decoding models that generalize across people. Test users demonstrate a closed-loop median performance of gesture decoding of 0.66 target acquisitions per second in a continuous navigation task, 0.88 gesture detections per second in a discrete-gesture task and handwriting at 20.9 words per minute. We demonstrate that the decoding performance of handwriting models can be further improved by 16% by personalizing sEMG decoding models. To our knowledge, this is the first high-bandwidth neuromotor interface with performant out-of-the-box generalization across people.