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

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

doi:10.3389/fninf.2017.00015

Cited by 19 publications

(41 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recording raw physiological signals is critical for sleep research because sleep stages provide only limited insight into sleep quality. Furthermore, these specific patterns can now be automatically detected using deep neural networks (15,33); though, analyses of the latter patterns are not reported here. Second, the data show that our method for detecting breathing frequency and RRV using an accelerometer has excellent agreement with the gold standard.…”

Section: Discussionmentioning

confidence: 99%

The Dreem Headband as an Alternative to Polysomnography for EEG Signal Acquisition and Sleep Staging

Thorey²,

et al. 2019

Preprint

View full text Add to dashboard Cite

Despite the central role of sleep in our lives and the high prevalence of sleep disorders, sleep is still poorly understood. The development of ambulatory technologies capable of monitoring brain activity during sleep longitudinally is critical to advancing sleep science and facilitating the diagnosis of sleep disorders. We introduced the Dreem headband (DH) as an affordable, comfortable, and user-friendly alternative to polysomnography (PSG). The purpose of this study was to assess the signal acquisition of the DH and the performance of its embedded automatic sleep staging algorithms compared to the gold-standard clinical PSG scored by 5 sleep experts. Thirty-one subjects completed an over-night sleep study at a sleep center while wearing both a PSG and the DH simultaneously. We assessed 1) the EEG signal quality between the DH and the PSG, 2) the heart rate, breathing frequency, and respiration rate variability (RRV) agreement between the DH and the PSG, and 3) the performance of the DH's automatic sleep staging according to AASM guidelines vs. PSG sleep experts manual scoring. Results demonstrate a strong correlation between the EEG signals acquired by the DH and those from the PSG, and the signals acquired by the DH enable monitoring of alpha (r= 0.71 ± 0.13), beta (r= 0.71 ± 0.18), delta (r = 0.76 ± 0.14), and theta (r = 0.61 ± 0.12) frequencies during sleep. The mean absolute error for heart rate, breathing frequency and RRV was 1.2 ± 0.5 bpm, 0.3 ± 0.2 cpm and 3.2 ± 0.6 %, respectively. Automatic Sleep Staging reached an overall accuracy of 83.5 ± 6.4% (F1 score : 83.8 ± 6.3) for the DH to be compared with an average of 86.4 ± 8.0% (F1 score: 86.3 ± 7.4) for the five sleep experts. These results demonstrate the capacity of the DH to both precisely monitor sleep-related physiological signals and process them accurately into sleep stages. This device paves the way for high-quality, large-scale, longitudinal sleep studies. Sleep | EEG | Machine learning | Sleep stages | DeviceCorrespondence: research@dreem.com

show abstract

Section: Discussionmentioning

confidence: 99%

The Dreem Headband as an Alternative to Polysomnography for EEG Signal Acquisition and Sleep Staging

Thorey²,

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…KCs were semi-automatically detected using an open-access validated method (Lajnef et al, 2015, 2017) which is based on a combination of the tunable Q-factor wavelet transform and a morphological component analysis. This approach requires an initial calibration step where a small subset of the data is visually scored for KCs and then used to derive an optimal threshold.…”

Section: Methodsmentioning

confidence: 99%

Increased Evoked Potentials to Arousing Auditory Stimuli during Sleep: Implication for the Understanding of Dream Recall

Vallat

Lajnef

Eichenlaub

et al. 2017

Front. Hum. Neurosci.

Self Cite

View full text Add to dashboard Cite

High dream recallers (HR) show a larger brain reactivity to auditory stimuli during wakefulness and sleep as compared to low dream recallers (LR) and also more intra-sleep wakefulness (ISW), but no other modification of the sleep macrostructure. To further understand the possible causal link between brain responses, ISW and dream recall, we investigated the sleep microstructure of HR and LR, and tested whether the amplitude of auditory evoked potentials (AEPs) was predictive of arousing reactions during sleep. Participants (18 HR, 18 LR) were presented with sounds during a whole night of sleep in the lab and polysomnographic data were recorded. Sleep microstructure (arousals, rapid eye movements (REMs), muscle twitches (MTs), spindles, KCs) was assessed using visual, semi-automatic and automatic validated methods. AEPs to arousing (awakenings or arousals) and non-arousing stimuli were subsequently computed. No between-group difference in the microstructure of sleep was found. In N2 sleep, auditory arousing stimuli elicited a larger parieto-occipital positivity and an increased late frontal negativity as compared to non-arousing stimuli. As compared to LR, HR showed more arousing stimuli and more long awakenings, regardless of the sleep stage but did not show more numerous or longer arousals. These results suggest that the amplitude of the brain response to stimuli during sleep determine subsequent awakening and that awakening duration (and not arousal) is the critical parameter for dream recall. Notably, our results led us to propose that the minimum necessary duration of an awakening during sleep for a successful encoding of dreams into long-term memory is approximately 2 min.

show abstract

“…We furthermore report some statistics about the K-complexes annotations over SS2 in Figure 7. The proposed approach outperforms the approach from Lajnef et al 2017 [15] in terms of Precision / Recall at IoU = 0.3, and in terms of F1 score for any IoU. Furthermore, the proposed approach seems to predict start and end times quite precisely.…”

Section: Spindles Detectionmentioning

confidence: 62%

“…Benchmarks relied on codes provided by the original authors 2 . For K-complexes detection, Lajnef et al 2017 [15] was considered as the baseline comparator.…”

Section: Datasetsmentioning

confidence: 99%

DOSED: A deep learning approach to detect multiple sleep micro-events in EEG signal

Chambon

Thorey²,

Arnal³

et al. 2019

Journal of Neuroscience Methods

View full text Add to dashboard Cite

Background : Electroencephalography (EEG) monitors brain activity during sleep and is used to identify sleep disorders. In sleep medicine, clinicians interpret raw EEG signals in so-called sleep stages, which are assigned by experts to every 30 s window of signal. For diagnosis, they also rely on shorter prototypical micro-architecture events which exhibit variable durations and shapes, such as spindles, K-complexes or arousals. Annotating such events is traditionally performed by a trained sleep expert, making the process time consuming, tedious and subject to interscorer variability. To automate this procedure, various methods have been developed, yet these are event-specific and rely on the extraction of hand-crafted features.New method : We propose a novel deep learning architecure called Dreem One Shot Event Detector (DOSED).DOSED jointly predicts locations, durations and types of events in EEG time series. The proposed approach, applied here on sleep related micro-architecture events, is inspired by object detectors developed for computer vision such as YOLO and SSD. It relies on a convolutional neural network that builds a feature representation from raw EEG signals, as well as two modules performing localization and classification respectively.Results and comparison with other methods: The proposed approach is tested on 4 datasets and 3 types of events (spindles, K-complexes, arousals) and compared to the current state-of-the-art detection algorithms. Conclusions:Results demonstrate the versatility of this new approach and improved performance compared to the current state-of-the-art detection methods.

show abstract

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

Cited by 19 publications

References 57 publications

The Dreem Headband as an Alternative to Polysomnography for EEG Signal Acquisition and Sleep Staging

The Dreem Headband as an Alternative to Polysomnography for EEG Signal Acquisition and Sleep Staging

Increased Evoked Potentials to Arousing Auditory Stimuli during Sleep: Implication for the Understanding of Dream Recall

DOSED: A deep learning approach to detect multiple sleep micro-events in EEG signal

Contact Info

Product

Resources

About