# MI-030: Motor Imagery for Severely Motor-Impaired Patients: Evidence for Brain-Computer Interfacing as Superior Control Solution

> Generated from a local paper card. Do not edit this page directly; edit the source card and rerun `vp run docs:generate`.

> Internal wiki boundary: paper-card evidence and user-provided PDF downloads are available for private reading. Extracted full-paper text is not published.

> Source: `library/paper_cards/MI-030.md`

## Paper Access

* Internal PDF: <a href={"/papers/MI-030.pdf"} download style={{ display: "inline-flex", alignItems: "center", justifyContent: "center", minHeight: "2.25rem", padding: "0.45rem 0.8rem", borderRadius: "6px", backgroundColor: "#047857", color: "#ffffff", fontWeight: 700, lineHeight: 1, textDecoration: "none", boxShadow: "0 1px 2px rgba(15, 23, 42, 0.22)" }}>Download Paper</a>
* DOI / official page: [10.1371/journal.pone.0104854](https://doi.org/10.1371/journal.pone.0104854)
* Open-access page: [Open access page](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0104854)
* Deployment boundary: these PDF links are intended only for a private/protected internal wiki.

## MI-030: Motor Imagery for Severely Motor-Impaired Patients: Evidence for Brain-Computer Interfacing as Superior Control Solution

## Metadata

* ID: MI-030
* Title: Motor Imagery for Severely Motor-Impaired Patients: Evidence for Brain-Computer Interfacing as Superior Control Solution
* Year: 2014
* DOI / URL: 10.1371/journal.pone.0104854
* Local PDF: see Paper Access section above
* Text artifact: local-only path withheld from docs site
* Review status: `extracted`

## Study Type

* Track: MI / clinical BCI
* Task: establish MI or attempted-execution BCI control for severely motor-impaired patients
* Participants or dataset: four patients
* Device/electrode setup: screening with 63-channel EEG; feedback sessions with a reduced setup noted in local text
* Protocol/task: patient-tailored BCI screening, MI training, and feedback applications

## Methods

* Signal processing or analysis: machine-learning features, LDA, feature adaptation, binary control with no-decision trials
* Training/calibration: six experimental sessions, user-centered tailoring
* Online/offline: online patient feedback/control plus offline estimates

## Key Results

* Three of four patients gained significant BCI control within the short study period.
* One highly affected patient showed BCI performance better than available assistive technology on accuracy, reaction time, and ITR.
* Local text reports up to about 90% binary accuracy for one patient context.

## Limitations

* Very small clinical sample.
* Patient-tailored setup and intense supervision limit generalization.
* Does not evaluate robotic grasping.

## Relevance To Current Review

* Supports MI as clinically meaningful but training/user-tailoring intensive.
* Helps separate rehabilitation/assistive claims from the engineering prototype claims of SAH-BRI-Grasp.

## Evidence Status

| Claim | Status | Evidence Note |
| --- | --- | --- |
| MI/attempted execution BCI can work for some severely motor-impaired patients. | verified | Abstract reports three of four patients gained significant control. |
| User-centered tailoring matters for clinical MI BCI. | verified | Abstract credits user-centered design and flexible technical setup. |
| These results prove SAH-BRI-Grasp rehabilitation efficacy. | needs confirmation | The project lacks clinical trials. |

## Open Questions

* Should clinical MI evidence be kept in product/rehab direction rather than core system validation?
* What no-decision strategy could transfer to project stop/cancel commands?