AUTHOR=Lajnef Tarek , O’Reilly Christian , Combrisson Etienne , Chaibi Sahbi , Eichenlaub Jean-Baptiste , Ruby Perrine M. , Aguera Pierre-Emmanuel , Samet Mounir , Kachouri Abdennaceur , Frenette Sonia , Carrier Julie , Jerbi Karim 

TITLE=Meet Spinky: An Open-Source Spindle and K-Complex Detection Toolbox Validated on the Open-Access Montreal Archive of Sleep Studies (MASS)

JOURNAL=Frontiers in Neuroinformatics

VOLUME=Volume 11 - 2017

YEAR=2017

URL=https://www.frontiersin.org/journals/neuroinformatics/articles/10.3389/fninf.2017.00015

DOI=10.3389/fninf.2017.00015

ISSN=1662-5196

ABSTRACT=Sleep spindles and K-complexes are among the most prominent micro-events observed in electroencephalographic (EEG)  recordings during sleep. These EEG microsturctures are thought to be hallmarks of sleep-related cognitive processes. Although tedious and time-consuming, their identification and quantification is an important component of many sleep studies in healthy subjects and in patients with sleep disorders. Therefore, procedures for automatic spindle and K-complex detections could provide valuable assistance to researchers and clinicians in the field. We recently proposed a framework for joint spindle and K-complex detector (Lajnef et al. Front Human Neurosci. 9, 414, 2015) based on a tunable Q-factor wavelet transform (TQWT) (Selesnick, IEEE Trans Sign Proc, 59, 3560–3575, 2011) and morphological component analysis (MCA). The current paper provides critical validation and benchmarking of the proposed approach by applying it to open-access EEG data from the Montreal Archive of Sleep Studies (MASS) (O’Reilly et al.  J Sleep Research, 23, 628-635, 2014), using a wide range of performance metrics. Importantly, these scores were compared to those previously reported for other methods tested on the same database. For spindle detection, our method provided higher performance than most of the alternative methods, as evidenced by all the statistics that take into account both sensitivity and precision (i.e., MCC, F1, Cohen k). Finally, our method is made available to the community via an open-source tool we called Spinky (for spindle and K-complex detection).Thanks to GUI implementation, and access to Matlab and Pyhton resources, Spinky is expected to contribute to an open-science approach which will enhance replicability and reliable comparisons of classifier performances for the detection of sleep EEG microstructure in healthy and diseased subjects.