{"author":{"@type":"Person","name":"Alexander Bock","url":"https://abxk.ai/en/autor/alexander-bock/"},"categories":["Physical AI"],"content":{"markdown":"A brain-computer interface (BCI) records brain activity and turns it into commands: a cursor moves, a robotic arm grasps, a synthetic voice speaks words a person can no longer say. AI models do the decoding, translating patterns of electrical activity into intended movements, letters or sounds. To see how far the field really is, we analyzed all {{\u003c num \"bci\" \"studies\" \u003e}} studies on ClinicalTrials.gov, the largest public registry of clinical studies, whose entry mentions a brain-computer or brain-machine interface, and counted the research papers on PubMed.\n\n![Stacked bars: brain-computer interface studies on ClinicalTrials.gov by start year, split into implanted, non-invasive and unclear](/images/blog/en/brain-computer-interface-trials-studies.webp \"BCI studies by start year. The implant studies are a recent wave.\")\n\n## Two kinds of brain-computer interfaces\n\nNon-invasive systems read brain activity from outside the head, mostly with an EEG cap of electrodes on the scalp, sometimes with near-infrared light (fNIRS). They are safe and cheap, but the skull blurs the signal, so they work for simple choices such as \"left or right\" rather than for typing at speed. Implanted systems place electrodes on or in the brain: tiny needle arrays in the motor cortex, thin grids on its surface (ECoG), or electrodes inside a blood vessel next to the brain. They need surgery, and in return the signal is clear enough to control a cursor, a robotic arm or synthetic speech.\n\n{{\u003c data \"bci-kind\" \u003e}}\n\nThe difference shows in the size of the studies. Half of the implant studies enroll or plan to enroll {{\u003c num \"bci\" \"enroll_median_implanted\" \u003e}} participants or fewer, half of the non-invasive studies {{\u003c num \"bci\" \"enroll_median_noninvasive\" \u003e}} or fewer. In the US, a trial with an implanted device that carries a significant risk needs an investigational device exemption from the FDA, and every participant needs brain surgery, so implant studies start with a handful of people.\n\n| | Non-invasive (EEG, fNIRS) | Implanted |\n|---|---|---|\n| Where the electrodes sit | On the scalp | On or in the brain, or in a blood vessel next to it |\n| Signal quality | Blurred by the skull | Clear, down to single neurons with needle arrays |\n| Typical use in studies | Rehabilitation after stroke, simple control | Computer and robot arm control, speech for people who cannot speak |\n| Risk | Very low | Surgery, infection, long-term stability of the implant |\n| Typical study size | Dozens of participants | A few participants |\n\n## Who the studies are for\n\nStroke is the most common condition, with {{\u003c num \"bci\" \"stroke\" \u003e}} studies. Most of them are rehabilitation trials: a patient imagines moving a paralyzed hand, an EEG-based BCI detects the attempt and triggers a robotic glove or electrical muscle stimulation, which links intention and movement again. The implant studies focus on other patients: {{\u003c num \"bci\" \"implanted_spinal\" \u003e}} of the {{\u003c num \"bci\" \"implanted\" \u003e}} concern a spinal cord injury or paralysis of all four limbs, {{\u003c num \"bci\" \"implanted_als\" \u003e}} concern ALS, a disease in which people gradually lose control of all muscles, including those needed for speech.\n\n{{\u003c data \"bci-conditions\" \u003e}}\n\n## Where the research happens\n\nStudies with sites in the United States lead by far. The studies cover {{\u003c num \"bci\" \"countries\" \u003e}} countries in total, and only {{\u003c num \"bci\" \"industry_pct\" \u003e}}% are led by companies; most come from universities and hospitals.\n\n{{\u003c data \"bci-countries\" \u003e}}\n\n## The implant trials running now\n\nImplants are the fastest-growing part of the registry: {{\u003c num \"bci\" \"implanted_started_since_2024\" \u003e}} of the {{\u003c num \"bci\" \"implanted\" \u003e}} implant studies started in 2024 or later, and {{\u003c num \"bci\" \"implanted_active\" \u003e}} are recruiting or still running. {{\u003c num \"bci\" \"implanted_active_us\" \u003e}} of these ongoing studies have sites in the US and {{\u003c num \"bci\" \"implanted_active_china\" \u003e}} in China. The titles show where the field is heading: wireless systems that are fully implanted instead of a cable through the skull, more recording channels, and speech decoding, including a study of speech in Mandarin Chinese.\n\n{{\u003c data \"bci-implanted\" \"n=20\" \u003e}}\n\n## Research output\n\nPubMed lists {{\u003c num \"bci\" \"pubmed_last_full_year\" \u003e}} papers from {{\u003c num \"bci\" \"last_full_year\" \u003e}} with a brain-computer or brain-machine interface in the title or abstract, {{\u003c num \"bci\" \"pubmed_growth_since_2015\" \u003e}} times as many as in 2015. Research grows much faster than clinical use: in the same year, only {{\u003c num \"bci\" \"started_last_full_year\" \u003e}} new studies with participants started.\n\n![Bar chart: number of PubMed papers per year with brain-computer interface or brain-machine interface in the title or abstract, from 2000 to the last full year](/images/blog/en/brain-computer-interface-trials-papers.webp \"Research papers per year with a brain-computer or brain-machine interface in the title or abstract.\")\n\n## What the numbers cannot tell\n\nA registry entry shows what a team plans, not what it achieved; results appear later, if at all. Not every study is registered at ClinicalTrials.gov: studies that only appear in other registries, such as the Chinese Clinical Trial Registry, are missing here. The split between implanted and non-invasive studies is our own, based on keywords in each entry, and {{\u003c num \"bci\" \"unclear\" \u003e}} entries do not say clearly enough which kind of interface they use.\n\n## How we counted\n\nWe retrieved all studies from the ClinicalTrials.gov API whose entry contains \"brain-computer interface\" or \"brain-machine interface\", with or without a hyphen, and grouped them by start date, status, condition, country of the study sites and type of interface. Implanted means the entry mentions an implant, intracortical or surface electrodes on the brain, depth electrodes or a device in a blood vessel; non-invasive means it mentions EEG, fNIRS, MEG or a non-invasive method. Papers were counted on PubMed by publication year with the same terms in the title or abstract. The source and date are below each table.\n\nHow such an interface turns brain signals into commands is explained in our learning path in [Physical AI](/en/physical-ai/).","text":"A brain-computer interface (BCI) records brain activity and turns it into commands: a cursor moves, a robotic arm grasps, a synthetic voice speaks words a person can no longer say. AI models do the decoding, translating patterns of electrical activity into intended movements, letters or sounds. To see how far the field really is, we analyzed all 280 studies on ClinicalTrials.gov, the largest public registry of clinical studies, whose entry mentions a brain-computer or brain-machine interface, and counted the research papers on PubMed.\nBCI studies by start year. The implant studies are a recent wave. Two kinds of brain-computer interfaces Non-invasive systems read brain activity from outside the head, mostly with an EEG cap of electrodes on the scalp, sometimes with near-infrared light (fNIRS). They are safe and cheap, but the skull blurs the signal, so they work for simple choices such as \u0026ldquo;left or right\u0026rdquo; rather than for typing at speed. Implanted systems place electrodes on or in the brain: tiny needle arrays in the motor cortex, thin grids on its surface (ECoG), or electrodes inside a blood vessel next to the brain. They need surgery, and in return the signal is clear enough to control a cursor, a robotic arm or synthetic speech.\nStudies by type of interface TypeStudiesShare Non-invasive (EEG, fNIRS and similar)12946.1% Not clear from the entry9232.9% Implanted electrodes5921.1% Source as for the first table on this page.\nThe difference shows in the size of the studies. Half of the implant studies enroll or plan to enroll 6 participants or fewer, half of the non-invasive studies 30 or fewer. In the US, a trial with an implanted device that carries a significant risk needs an investigational device exemption from the FDA, and every participant needs brain surgery, so implant studies start with a handful of people.\nBrain-Computer Interfaces in Numbers: Trials, Implants and Research Non-invasive (EEG, fNIRS) Implanted Where the electrodes sit On the scalp On or in the brain, or in a blood vessel next to it Signal quality Blurred by the skull Clear, down to single neurons with needle arrays Typical use in studies Rehabilitation after stroke, simple control Computer and robot arm control, speech for people who cannot speak Risk Very low Surgery, infection, long-term stability of the implant Typical study size Dozens of participants A few participants Who the studies are for Stroke is the most common condition, with 112 studies. Most of them are rehabilitation trials: a patient imagines moving a paralyzed hand, an EEG-based BCI detects the attempt and triggers a robotic glove or electrical muscle stimulation, which links intention and movement again. The implant studies focus on other patients: 40 of the 59 concern a spinal cord injury or paralysis of all four limbs, 26 concern ALS, a disease in which people gradually lose control of all muscles, including those needed for speech.\nStudies by condition (a study can belong to several groups) ConditionStudies Stroke112 Spinal cord injury, tetraplegia69 ALS, locked-in syndrome52 Parkinson's, epilepsy, other neurological38 Other35 Other paralysis, loss of speech27 Mental health, disorders of consciousness26 Healthy volunteers21 Source as for the first table on this page.\nWhere the research happens Studies with sites in the United States lead by far. The studies cover 35 countries in total, and only 13.2% are led by companies; most come from universities and hospitals.\nCountries with the most studies (study sites) CountryStudies United States96 China27 Singapore22 France21 Switzerland10 Canada10 United Kingdom7 Italy6 Taiwan6 Germany5 Source as for the first table on this page.\nThe implant trials running now Implants are the fastest-growing part of the registry: 35 of the 59 implant studies started in 2024 or later, and 46 are recruiting or still running. 18 of these ongoing studies have sites in the US and 11 in China. The titles show where the field is heading: wireless systems that are fully implanted instead of a cable through the skull, more recording channels, and speech decoding, including a study of speech in Mandarin Chinese.\nOngoing studies with implanted electrodes, newest first StudyTitle in the registryCountryStartParticipants NCT07521930Interfacing With NeuroTechnology to Expand Neural Throughput (INTENT)United States2027-015 NCT06898138Acute Stimulation and Modulation of Stereotyped High-Frequency OscillationsUnited States2026-11-018 NCT05470478iBCI Optimization for Veterans With ParalysisUnited States2026-10-022 NCT07745686VALSE: Integration of Neuromodulation and Exoskeleton Systems for Mobility in Individuals With Chronic Spinal Cord InjurySwitzerland2026-09-1512 NCT07610850Brain-controlled Spinal Cord Stimulation in Participants With Chronic Stroke for Lower and Upper Limb RehabilitationSwitzerland2026-09-013 NCT07647315Wireless Implantable Mandarin Chinese Speech Neuroprosthesis EvaluationChina2026-09-014 NCT07791459BrainGate2: Feasibility Study of an Intracortical Neural Interface System for Persons With Aphasia–2026-085 NCT07640711Innovative Brain-Computer Interface for People With Spinal Cord InjuryNetherlands2026-081 NCT07784088Implantable Wireless Brain-Computer Interface Systems in Patients With TetraplegiaChina2026-07-1534 NCT07720882A Clinical Study of a Fully Implanted, Wireless and Functional BCI for Upper-Limb Functional Replacement in Tetraplegia Due to Spinal Cord Injury: Efficacy and SafetyChina2026-07-0732 NCT07465796A First-in-Human Study of the CIPHER System During Brain Surgery for Newly Diagnosed GliomaAustralia2026-05-2210 NCT07533903Functional Outcomes and Control Using Synchron BCI - AustraliaAustralia2026-05-2210 NCT07457645Safety and Efficacy of High-Channel Implanted BCI for Motor Function Improvement in Tetraplegia PatientsChina2026-05-1510 NCT07553416Prospective, Randomized, Parallel-Controlled Study of Brain Computer Interface Integrated Robotic Mirror Therapy for Post-Stroke Upper Limb Motor Function DisorderChina2026-05-0140 NCT07603752Implantable Wireless Brain-Machine Interface System for Spinal Cord Injury: Efficacy and Safety StudyChina2026-04-2834 NCT07477613Implantable Brain-Computer Interface for Upper-Limb Recovery After StrokeFrance2026-04-0810 NCT07446114Functional Outcomes and Control Using Synchron BCI - CanadaCanada2026-03-1810 NCT07460037Chinese-Specific Speech Imagery Coding Using High-Density ECoGChina2026-03-1550 NCT05936619MindEx: A Novel, Multifocal, Cognitive Brain-Machine Interface SystemUnited States2026-02-202 NCT07225582Development of Optimal Sensory Feedback Strategies to Maximize Function After TetraplegiaUnited States2026-01-133 NCT06939660BCI-Assisted SCS-EXS for Gait OptimizationChina2025-04-233 NCT06642519Brain-Machine Interface for Freezing of GaitCanada2025-04-1010 NCT06944834A Study on the Safety and Functionality of the Implantable Wireless Brain-Computer Interfaces for Motor RehabilitationChina2025-045 NCT06829212Research on Wireless Brain Implant System for General Control of External DevicesChina2025-034 NCT06511934Feasibility of the BrainGate2 Neural Interface System in Persons With Tetraplegia (BG-Tablet-01)United States2024-07-226 NCT06432907StereoEEG Motor Neuronal Potentials DecodingRussia2024-06-0630 NCT06243952Brain Controlled Spinal Cord Stimulation In Participants With Spinal Cord Injury For Lower Limb RehabilitationSwitzerland2024-05-033 NCT06207591Investigation on the Cortical Communication SystemNetherlands2023-12-152 NCT05920174Clinical Trial to Evaluate the Safety and Efficacy of an Implantable Neural Acquisitor \u0026amp; Stimulator System in Patients With Motor DisabilityChina2023-10-179 NCT06094205Feasibility of the BrainGate2 Neural Interface System in Persons With Tetraplegia (BG-Speech-02)United States2023-10-162 NCT05665998Brain Controlled Spinal Cord Stimulation in Participants With Cervical Spinal Cord Injury for Upper Limb RehabilitationSwitzerland2023-05-223 NCT05035823COMMAND Early Feasibility Study: Implantable BCI to Control a Digital Device for People With ParalysisUnited States2022-04-276 NCT03567213Investigation on the Cortical Communication (CortiCom) SystemUnited States2021-12-143 NCT05439655Acute Recording of Stereotyped High-Frequency OscillationsUnited States2021-09-0117 NCT03698149ECoG BMI for Motor and Speech ControlUnited States2018-11-093 NCT03161067Investigation on the Bidirectional Cortical Neuroprosthetic SystemUnited States2017-08-015 NCT02550522Brain Computer Interface: Neuroprosthetic Control of a Motorized ExoskeletonFrance2015-095 NCT01894802Cortical Recording and Stimulating Array Brain-Machine InterfaceUnited States2013-12-0130 NCT01958086Visuomotor Prosthetic for ParalysisUnited States2013-10-012 NCT00912041BrainGate2: Feasibility Study of an Intracortical Neural Interface System for Persons With TetraplegiaUnited States2009-0527 Source as for the first table on this page.\nResearch output PubMed lists 1,390 papers from 2,025 with a brain-computer or brain-machine interface in the title or abstract, 2.4 times as many as in 2015. Research grows much faster than clinical use: in the same year, only 36 new studies with participants started.\nResearch papers per year with a brain-computer or brain-machine interface in the title or abstract. What the numbers cannot tell A registry entry shows what a team plans, not what it achieved; results appear later, if at all. Not every study is registered at ClinicalTrials.gov: studies that only appear in other registries, such as the Chinese Clinical Trial Registry, are missing here. The split between implanted and non-invasive studies is our own, based on keywords in each entry, and 92 entries do not say clearly enough which kind of interface they use.\nHow we counted We retrieved all studies from the ClinicalTrials.gov API whose entry contains \u0026ldquo;brain-computer interface\u0026rdquo; or \u0026ldquo;brain-machine interface\u0026rdquo;, with or without a hyphen, and grouped them by start date, status, condition, country of the study sites and type of interface. Implanted means the entry mentions an implant, intracortical or surface electrodes on the brain, depth electrodes or a device in a blood vessel; non-invasive means it mentions EEG, fNIRS, MEG or a non-invasive method. Papers were counted on PubMed by publication year with the same terms in the title or abstract. The source and date are below each table.\nHow such an interface turns brain signals into commands is explained in our learning path in Physical AI.\n","words":1380},"date":"2026-10-18","description":"Who tests brain-computer interfaces, on which patients and with what kind of implant? We analyzed every BCI study on ClinicalTrials.gov and the research output on PubMed.","image":"https://abxk.ai/images/cover/brain-computer-interface-trials.webp","json":"https://abxk.ai/en/physical-ai/brain-computer-interface-trials/index.json","kind":"page","language":"en","modified":"2026-10-18","section":"physical-ai","seo_title":"Brain-Computer Interface Clinical Trials: The Data","tags":["brain-computer-interface","bci","neurotechnology","clinical-trials","physical-ai"],"title":"Brain-Computer Interfaces in Numbers: Trials, Implants and Research","translations":[{"json":"https://abxk.ai/de/physical-ai/brain-computer-interface-trials/index.json","language":"de","url":"https://abxk.ai/de/physical-ai/brain-computer-interface-trials/"}],"url":"https://abxk.ai/en/physical-ai/brain-computer-interface-trials/"}