Brain-Computer Interface (BCI) represents the communication between the brain and a computer that may be used to interpret the real-time activity of the brain through signal processing and machine learning. Such communication offers a great gamma of possibilities for the medical focused research, for example the development of prosthetics controlled by motor imagery, calibration of cochlear implants and communication improvement of patients with amyotrophic lateral sclerosis.
Given its great performance in the presence of muscles movements, the system portability and its ease of application, fNIRS presents itself as an optimal candidate to acquire the cortical brain signals as reliable and representative inputs for a Brain-Computer Interface investigation.
Yet another whole field to be explored is based on neurofeedback, i.e. the feedback obtained from the online brain activity that may be used for the subject’s self-regulation. With the advent of fNIRS and its capability of measuring two or more subjects at the same time on realistic environments, the neurofeedback possibilities based on hemodynamic states have been widely expanded, as depicted on the schematic below. Research fields that may strongly benefit from that are social interaction, stroke rehabilitation and emotions regulation.
A key process for both BCI and Neurofeedback is the possibility of streaming the raw data along with the trigger markers from the acquisition computer (server) to the presentation computer (client). This allows for real-time processing of the raw data, which is essential for feature extraction for BCI as well as convertion to hemodynamic states to be used as input for the Neurofeedback module. A general schematic is shown below.
The real-time data stream, as depicted above, can be easily achieved with the NIRx Software Development Kit (SDK), which provides a dynamic library that allows the user to communicate with the acquisition software. *Users without programming knowledge may also easily benefit from SDK with the several examples available.
Two examples of results recently published by NIRx users focusing on feature extraction and discrimination of different tasks are shown below. These illustrate how hemodynamic states are robust for signal classification.
Left: Hemodynamic changes after unexpected delivery of pleasant and unpleasant liquids. 
Right: Features from mental arithmetic and mental counting signals are extracted and decoded for a hybrid brain-computer interface. 
 DiStasio, M.M., and Francis, J.T., “Use of frontal lobe hemodynamics as reinforcement signals to an adaptive controller”, PLoS ONE 8(7), e69541, doi:10.1371/journal.pone.0069541 (2013).
 Khan, M.J., Hong, M.J., and Hong, K.-S., "Decoding of four movement directions using hybrid NIRS-EEG brain-computer interface." Frontiers in human neuroscience 8(244), 1, doi:10.3389/fnhum.2014.00244 (2014).
 Lee, Min-Ho, Siamac Fazli, Jan Mehnert, and Seong-Whan Lee. "Subject-dependent classification for robust idle state detection using multi-modal neuroimaging and data-fusion techniques in BCI." Pattern Recognition 48, no. 8: 2725-2737 (2015).
 Tumanov, Kirill, R. Goebel, R Mockel, B Sorger, G. Weiss. "fNIRS-based BCI for Robot Control." Proceedings of the 2015 International Conference on Autonomous Agents and Multiagent Systems. International Foundation for Autonomous Agents and Multiagent Systems, (2015).
 Khan, M. Jawad, Melissa Jiyoun Hong, and Keum-Shik Hong. "Decoding of four movement directions using hybrid NIRS-EEG brain-computer interface." Frontiers in human neuroscience 8 (2014).
 DiStasio, Marcello M. ,and J. T. Francis, “Use of frontal lobe hemodynamics as reinforcement signals to an adaptive controller”, PLoS ONE 8(7), e69541 (2013).
 Fazli, Siamac, J. Mehnert, J. Steinbrink, B. Blankertz. "Using NIRS as a predictor for EEG-based BCI performance." Engineering in Medicine and Biology Society (EMBC), 2012 Annual International Conference of the IEEE. IEEE, (2012).
 Hu, Xiao-Su, K.-S. Hong, and S.S. Ge, “fNIRS-based online deception decoding,” J. Neural Engineering 9, 026012 (2012).
 Fazli, Siamac, J. Mehnert, J. Steinbrink, G. Curio, A. Villringer, K.R. Müller, and B. Blankertz, “Enhanced performance by a hybrid NIRS-EEG brain computer interface,” NeuroImage59(1), 519-529, doi: 10.1016/j.neuroimage.2011.07.084 (2012).
 Waldert, Stephan, L. Tüshaus, C.P. Kaller, and C. Mehring, “fNIRS exhibits weak tuning to hand movement direction,” PLoS ONE 7(11): e49266. doi:10.1371/journal.pone.0049266 (2012).
 Herff, Christian, F. Putze, D. Heger, C. Guan, T. Schultz, “Speaking mode recognition from functional Near Infrared Spectroscopy,” Conf Proc IEEE Eng Med Biol Soc. 2012:1715-8 (2012).
 Gottemukkula, Vikas, R. Derakhshani, “Classification-guided Feature Selection for NIRS-based BCI,” Neural Engineering (NER), 2011 5th International IEEE/EMBS Conference on. IEEE, (2011).
NIRScout is an ultra-compact fNIRS neuroimaging solution that offers the versatility and scalability to fit a broad range of research applications. For example, investigating motor sensory functions, traumatic brain injury, learning/attention disorders and brain computer interfaces.
NIRSport is the first freely configurable, wearable, multichannel NIRS imaging system to be used on any part of the head. This pocket-book sized device combines LED illumination and innovative active detection technology into a truly wearable brain imaging solution.