PolyMorph: Increasing the Spelling Efficiency of P300 by Selection Matrix PolyMorphism and Sentence-Based Predictions

ABSTRACT

One application of the P300 brain electric signal is sentence spelling, which enables subjects who have lost control of their motor pathways to communicate by selecting characters in a matrix containing all alphabet symbols. This technology still suffers from both low communication/high error rates. A P300 speller, named PolyMorph, which jointly introduces the selection matrix polymorphism (reducing the matrix size by removing useless symbols) and sentence-based predictions (which forecast words on the basis of language statistics) is presented. This is accomplished by using a custom dynamic knowledge-base, tailored to the subject lexicon, and updated in real time with the selections of the subject. The effectiveness of the presented speller is measured in vivo and in silico. The results suggest that the use of PolyMorph increases the number of spelt characters per time unit and reduces the error rate.

KEYWORDS:

• Brain–computer interface
• EEG
• P300
• event-related potential
• predictive speller

Acknowledgments

The authors gratefully acknowledge Dr. Katie Palmer and Dr. Patrick Moore for assistance in language revision.

Additional information

Funding

This work has been partially supported by Istituto Nazionale di Alta Matematica (INdAM) and by the University of Trieste as one of the outcomes of the project FRA 2014 “Learning specifications and robustness in signal analysis (with a case study related to health care)”.

Notes on contributors

Alberto Casagrande

Alberto Casagrande is a Ph.D. computer scientist whose interests include – but are not limited to – hybrid systems, formal verification, systems biology, and algorithms. He has been assistant professor of Computer Science at University of Trieste from 2008 to 2016 and from 2017 onward.

Joanna Jarmolowska

Joanna Jarmolowska is a psychologist with a Ph.D. in Neuroscience. At present she is a researcher at the Department of Life Sciences at the University of Trieste. She is engaged in research on the development of brain computer interfaces and neurofeedback techniques (for adults and children) by using electroencephalography.

Marcello Maria Turconi

Marcello Maria Turconi has a M.Sc. in Neuroscience and Ph.D. in Neuroscience and Cognitive Science obtained with a thesis on rehabilitative B.C.I.s. He currently works as science communicator.

Pierpaolo Busan

Pierpaolo Busan is a psychologist with Ph.D. in Neuroscience. He is now a researcher at I.R.C.C.S. Fondazione Ospedale San Camillo, in Venice. He is involved in basic and translational research on brain neurophysiology (in health and disease) by using tools such as electroencephalography, transcranial magnetic stimulation, magnetoencephalography, and brain–computer interfaces.

Francesco Fabris

Francesco Fabris Ph.D., is associate professor of Computer Engineering at the University of Trieste. His main research interests are related to information theory and application of mathematics and information theory to biology and medicine, with special reference to brain computer interfaces, quality of diagnostic tests, automatic detection and forecast of patient-ventilator dyssynchrony.

Piero Paolo Battaglini

Piero Paolo Battaglini, M.D., is full professor in Physiology at the University of Trieste. For many years, he has been involved in the study of motor programming, using different approaches. More recently he is working on the development of brain computer interfaces and neurofeedback techniques.