Binary Communication Techniques in Locked-in and Completely Locked-in Patients

DSpace Repositorium (Manakin basiert)


Dateien:

Zitierfähiger Link (URI): http://hdl.handle.net/10900/146550
http://nbn-resolving.de/urn:nbn:de:bsz:21-dspace-1465502
http://dx.doi.org/10.15496/publikation-87891
Dokumentart: Dissertation
Erscheinungsdatum: 2023-10-19
Sprache: Englisch
Fakultät: 7 Mathematisch-Naturwissenschaftliche Fakultät
Fachbereich: Informatik
Gutachter: Birbaumer, Niels (Prof. Dr.)
Tag der mündl. Prüfung: 2023-01-17
DDC-Klassifikation: 610 - Medizin, Gesundheit
Schlagworte: Myatrophische Lateralsklerose , Locked-in-Syndrom , Kommunikation , Gehirn-Computer-Schnittstelle
Lizenz: http://tobias-lib.uni-tuebingen.de/doku/lic_mit_pod.php?la=de http://tobias-lib.uni-tuebingen.de/doku/lic_mit_pod.php?la=en
Gedruckte Kopie bestellen: Print-on-Demand
Zur Langanzeige

Inhaltszusammenfassung:

Neurodegenerative Erkrankungen wie Amyotrophe Lateralsklerose (ALS) können zu teilweiser oder völliger Lähmung des Körpers führen, Patienten in diesem Zustand können dann als eingeschlossen (LIS) oder völlig eingeschlossen (CLIS) bezeichnet werden. Bevor Menschen diese Zustände erleiden, kommunizieren sie heutzutage meist mit Augenbewegungen, mit denen sie Kommunikationshilfen bedienen. Mit dem weiteren Fortschreiten der Erkrankung werden diese Geräte üblicherweise nutzlos, da man Augenbewegungen nicht mehr registrieren kann. Dann gibt es für diese Patienten nur noch die Lösung von Gehirn-Computer-Schnittstellen (BCI, brain-computer-interface), bei denen mit Hirnsignalen ohne Vermittlung von Muskelaktivität solche Kommunikationshilfen bedient werden können. Obwohl die erfassten Hirnsignale bei solchen Patienten oft pathologisch sind, hatte dies bisher keinen nennenswerten Einfluss auf die Entschlüsselung ihrer Hirnsignale und ihre Kommunikationsleistungen mit einem BCI wie diese Dissertation zeigt. Bei der Entwicklung solcher Kommunikationshilfen mit BCI ging man modular vor, indem man die Signalerfassung und -analyse getrennt von den Kommunikationsaufgaben und Kommunikationsleistungen hielt. Verschiedene physiologische Parameter wurden für unterschiedliche Patientengruppen und Aufgaben verwendet: Elektrookulographie (EOG), Elektroenzephalographie (EEG), Nah-Infrarotspektroskopie (NIRS) und intrakortikale Hirnsignale. Obwohl die Signale unterschiedlich verarbeitet wurden, ist allen Versuchen, so auch unseren, gemeinsam, dass die Patienten lernen mussten, ihre Hirnsignale willentlich zu modulieren. Nach Klassifizierung des physiologischen Signals konnte der/die Patientin damit unterschiedliche Kommunikations-Paradigmen kontrollieren (z.B. ein „Ja“ oder „Nein“ auswählen). Wir haben hier verschiedenen Paradigmen bis hin zur hirngesteuerten Auswahl von Buchstaben und Bildung von Sätzen berichtet. Wir stellen hier verschiedene Kombinationen von Signalaufnahmen und -verarbeitungen in unterschiedlichen Kommunikationssituationen vor. Die meisten davon wurden erfolgreich bei eingeschlossenen (LIS) und völlig eingeschlossenen Patienten (CLIS) erprobt. Es werden auch zwei Patienten berichtet, die über keine andere Kommunikationsmöglichkeit mehr verfügten.

Abstract:

Neurodegenerative disease, such as amyotrophic lateral sclerosis (ALS), can lead to a partial or complete paralysis of the body, and patients in these conditions are considered to be in locked-in (LIS) or complete locked-in state (CLIS). Before reaching these advanced states, patients usually communicate by moving the eyes with the help of assistive and augmentative communication devices. With the progression of the disease these devices become useless due of ocular abnormalities that prevent the correct detection of the eye movements and gaze fixation. A valuable solution for these patients is the implementation of brain-computer interfaces (BCIs) which, by decoding neural signals directly acquired from the brain activity, can be used to develop communication paradigms without the need of relying on any muscular activity. Several preliminary neurophysiological analyses have been conducted to assess possible pathological signs in the neural activity of ALS patients in LIS and CLIS. The published results included in the dissertation indicate heterogeneous conditions that, in most cases, highlight abnormal neural signals. Nevertheless, no result indicates significative cognitive problems that might affect the BCI performance. The development of communications techniques has been implemented in a modular way by keeping the signal acquisition and signal processing phases separate from the implementation of the communication paradigms. Several acquired signals have been used to target different patient populations: electrooculography, electroencephalography, functional near-infrared spectroscopy, and intracortical neural activity. In each study, the signal was processed differently, but patients were always asked to modulate their brain activity in only two different ways; the signal was then decoded using a binary classifier. The output of the classifier has been used to control different developed communication paradigms: from simple yes/no questions, to a full speller that allowed the patient to form words and sentences. The results show several combinations of different signal acquisition techniques with various communication paradigms, taking advantage of the modularity of the developed systems. Most of the communication paradigms have also been successfully deployed to LIS or CLIS patients, including two cases in which the speller system has been used by patients with no other means of communication.

Das Dokument erscheint in: