Knowledge Vault 3/83 - G.TEC BCI & Neurotechnology Spring School 2024 - Day 9
BCIs for the assessment of locked-in and complete locked-in patients
Rossella Spataro, University of Palermo (IT)
<Resume Image >

Concept Graph & Resume using Claude 3 Opus | Chat GPT4 | Llama 3:

graph LR classDef spataro fill:#f9d4d4, font-weight:bold, font-size:14px; classDef lockedin fill:#d4f9d4, font-weight:bold, font-size:14px; classDef bci fill:#d4d4f9, font-weight:bold, font-size:14px; classDef communication fill:#f9f9d4, font-weight:bold, font-size:14px; classDef future fill:#f9d4f9, font-weight:bold, font-size:14px; A[Rossella Spataro] --> B[Spataro: BCI for locked-in patients 1] A --> C[Locked-in syndrome diagnosis:
quadriplegia, anartia 2] C --> D[Classic, incomplete,
complete locked-in classifications 3] C --> E[Causes: stroke, ALS,
tumors, encephalitis 4] C --> F[Limited quality of life 5] A --> G[BCI benefits: consciousness
assessment, communication 6] A --> H[Locked-in communication
history: eye movements 7] H --> I[Eye trackers for
classic locked-in 8] A --> J[BCI autonomy idea: Star Trek 9] A --> K[Early BCI attempts: slow cortical potentials 10] A --> L[Visual P300 spellers: faster,
no training 11] L --> M[Evolved spellers: matrix sizes,
predictive typing, motor imagery 12] L --> N[SSVEP spellers: focus on
flashing arrows 13] A --> O[ALS progression requires non-visual BCIs 14] O --> P[MindBeagle: vibrotactile P300, motor
imagery for completely locked-in 15] P --> Q[MindBeagle validation: healthy subjects,
locked-in patients at home 16] P --> R[Binary choice: left/right vibrotactile
or motor imagery 17] P --> S[16 ALS patients: 79% two-stimulus,
73% three-stimulus accuracy 18] S --> T[12/16 vibrotactile, 4/16 motor
imagery communication success 19] S --> U[ALS outperformed stroke with
vibrotactile due to somatosensory preservation 20] A --> V[Endogenous BCIs: mental imagery,
math for locked-in communication 21] A --> W[Invasive BCIs: electrocorticography, intracortical
microelectrodes for locked-in communication 22,23] A --> X[BCIs for locked-in robot
control and assistance 24] X --> Y[Visual P300 for locked-in
humanoid robot grasping 25] X --> Z[Wearable wireless BCIs for
non-visual robot control 26] X --> AA[Brain-controlled wheelchairs need
safety improvements 27] X --> AB[BCIs for drone control
in VR/AR 28] A --> AC[BCI goals: restoring communication,
autonomy, participation, quality of life 29] A --> AD[Locked-in depression decreases over
time, caregiver depression persists 30] class A,B spataro; class C,D,E,F,AD lockedin; class G,H,I,J,K,L,M,N,O,P,Q,R,S,T,U,V,W,X,Y,Z,AA,AB,AC bci; class H,I,L,M,N,O,P,Q,R,S,T,U,V,W communication; class AA,AB,AC future;

Resume:

1.- Dr. Rossella Spataro, a neurologist from the University of Palermo, presented on BCI assessment of locked-in and completely locked-in patients.

2.- Locked-in syndrome is diagnosed by observing quadriplegia and anartia, with preserved consciousness and cognitive function.

3.- Locked-in syndrome is classified into classic, incomplete, and complete forms based on preserved motor control.

4.- Stroke in the pons is the most frequent cause of locked-in syndrome, followed by ALS, brain tumors, and encephalitis.

5.- Quality of life for locked-in patients is very limited, with restricted mobility, communication, and leisure activities.

6.- BCI can help assess consciousness, cognitive function, and provide communication and autonomy for locked-in patients.

7.- Historically, locked-in patients have communicated through residual eye movements, as exemplified by the case of Jean-Dominique Bauby.

8.- Eye trackers are commonly used by classic locked-in patients to communicate, but many progress to a completely locked-in state.

9.- The idea of using BCI for autonomy in locked-in patients dates back to a 1966 Star Trek episode.

10.- Early BCI communication attempts with locked-in ALS patients in the late 1990s used slow cortical potentials but were very slow.

11.- Visual P300 spellers allowed faster BCI communication without training in locked-in patients, with accuracy around 86% on first use.

12.- BCI spellers have evolved to use different matrix sizes, integrations with predictive typing, and non-visual modalities like motor imagery.

13.- Alternative BCI spellers use steady-state visual evoked potentials to select letters by focusing on peripheral flashing arrows.

14.- As ALS progresses, visual BCI systems become unusable, necessitating the development of visual-independent auditory and tactile BCIs.

15.- The MindBeagle system uses vibrotactile P300 potentials and motor imagery for communication in completely locked-in patients.

16.- Validation of MindBeagle was conducted with healthy subjects and locked-in patients in home settings, adapting to practical challenges.

17.- Binary choice communication is established by attending to left/right vibrotactile stimuli or imagining left/right hand movements for yes/no responses.

18.- In a study of 16 locked-in ALS patients, accuracy was 79% for two-stimulus and 73% for three-stimulus vibrotactile paradigms.

19.- 12/16 patients could communicate with vibrotactile and 4/16 with motor imagery BCIs, including some completely locked-in patients.

20.- ALS patients performed better than stroke patients with vibrotactile BCIs due to preserved somatosensory pathways.

21.- Endogenous BCIs using mental imagery tasks like limb movement and mental math show mixed results for communication in locked-in patients.

22.- Invasive BCIs using electrocorticography have allowed a locked-in ALS patient to communicate at 2 characters/minute with high satisfaction.

23.- Intracortical microelectrodes recording from motor cortex enabled a locked-in patient to achieve a 9 word vocabulary with low error rate.

24.- Beyond communication, BCIs are being developed to allow locked-in patients to control robots for personal assistance in daily life.

25.- Visual P300 BCIs have allowed locked-in ALS patients to successfully control a humanoid robot to grasp and deliver objects.

26.- Wearable wireless BCIs are being tested for hands-free robot control using non-visual modalities suitable for locked-in patients.

27.- Brain-controlled wheelchairs are another assistive application being developed, but require further advancement to ensure safety for practical use.

28.- BCIs are also being applied to control drones in virtual and augmented reality environments using visual P300 signals.

29.- The lecture highlighted many BCI solutions aiming to restore communication, autonomy, social participation and quality of life for locked-in patients.

30.- Depression is common in locked-in patients at diagnosis but decreases over time, while caregiver depression remains a major challenge.

Knowledge Vault built byDavid Vivancos 2024