Model

DORMI (Sleepacta sleep/wake; EU MDR Class I; Banfi lightCNNA Sci Rep 2021)

Cite
Venue listedCatalog paper date: 2021Added to atlas: Nov 9, 2026Source

Activity counts measure attention, not model quality. Missing counts are unknown. Publication status follows the linked source; a listed venue alone does not confirm peer review.

Sleepacta’s EU MDR 2017/745 Class I sleep/wake analysis software (BD/RDM 2640727, CND Z12030692, registered 28/08/2024). Classifies sleep vs wake on 30 s epochs; does not estimate sleep stages and does not replace PSG (company).

Parameters
0.001361M
Signals
IMU
Form factor
Wrist
Year
2021
Architecture

lightCNNA: 3× Conv1D (8 filters, kernel 8, dilations 6/3/1, no bias) + dense(16, ReLU) + sigmoid; 1,361 trainable params; ~2,727 FLOPs; optional TensorFlow Lite / C-array embed

License

proprietary (Sleepacta commercial MD); paper methods disclosed; weights/code not released

At a glance
Type
Task model
Modalities
IMU
Form factor
Wrist
Parameters
0.001361M
Architecture
lightCNNA: 3× Conv1D (8 filters, kernel 8, dilations 6/3/1, no bias) + dense(16, ReLU) + sigmoid; 1,361 trainable params; ~2,727 FLOPs; optional TensorFlow Lite / C-array embed
Venue
Scientific Reports 2021;11:345 (method); Sleepacta commercial DORMI CE 2024
Year
2021
arXiv
Not reported
Code
Not released
Weights
Not released
Licence
proprietary (Sleepacta commercial MD); paper methods disclosed; weights/code not released
Caveats
Not reported
Datasets
Not reported
Added
Nov 9, 2026
Updated
Nov 9, 2026

Abstract

This study presents a thorough analysis of sleep/wake detection algorithms for efficient on-device sleep tracking using wearable accelerometric devices. It develops a novel end-to-end algorithm using convolutional neural network applied to raw accelerometric signals recorded by an open-source wrist-worn actigraph. The aim of the study is to develop an automatic classifier that: (1) is highly generalizable to heterogenous subjects, (2) would not require manual features’ extraction, (3) is computationally lightweight, embeddable on a sleep tracking device, and (4) is suitable for a wide assortment of actigraphs. Hereby, authors analyze sleep parameters, such as total sleep time, waking after sleep onset and sleep efficiency, by comparing the outcomes of the proposed algorithm to the gold standard polysomnographic concurrent recordings. The relatively substantial agreement (Cohen’s kappa coefficient, median, equal to 0.78 ± 0.07) and the low-computational cost (2727 floating-point operations) make this solution suitable for an on-board sleep-detection approach.

Verbatim fromthe publisher's DOI record· Publisher licence: https://creativecommons.org/licenses/by/4.0

DORMI (Sleepacta sleep/wake; EU MDR Class I; Banfi lightCNNA Sci Rep 2021): ChatPPG research summary card

Key results

How it works

How DORMI (Sleepacta sleep/wake; EU MDR Class I; Banfi lightCNNA Sci Rep 2021) works: Input signals → Model → Downstream tasksInput signals: IMU (Accelerometer, 3-axis). Model: DORMI (Sleepacta sleep/wake; EU MDR Class I; Banfi lightCNNA Sci Rep 2021), 0.001361M parameters (Convolutional network). Downstream tasks: Sleep staging.Input signalsIMU: Accelerometer, 3-axisxyzIMUAccelerometer, 3-axisModelDORMI (Sleepacta sleep/wake…0.001361M parametersConvolutional networkDownstream tasksSleep stagingSleep staging
Architecture
lightCNNA: 3× Conv1D (8 filters, kernel 8, dilations 6/3/1, no bias) + dense(16, ReLU) + sigmoid; 1,361 trainable params; ~2,727 FLOPs; optional TensorFlow Lite / C-array embed
Inputs
IMU
Approach
sleep · sleep wake · actigraphy · CNN · embedded · EU MDR

Leaderboard rows· 2

All leaderboards
TaskDatasetMetricValueRank
Sleep/wake detection (actigraphy-style)own-pisa-ax3-psg-banfi-2021Accuracy ↑0.8932#1 / 1
Sleep/wake detection (actigraphy-style)own-pisa-ax3-psg-banfi-2021Cohen's κ ↑0.78#1 / 1

Datasets used

No public datasets linked yet.

Papers· 1

Cite this model

Cite the publication recorded for this model. Check it against the publisher before submitting.

Banfi, T., Valigi, N., di Galante, M., d’Ascanio, P., Ciuti, G., & Faraguna, U. (2021). Efficient embedded sleep wake classification for open-source actigraphy. Scientific Reports 2021, 11, 345. https://doi.org/10.1038/s41598-020-79294-y
Banfi, Tommaso, Nicolò Valigi, Marco di Galante, Paola d’Ascanio, Gastone Ciuti, and Ugo Faraguna. 2021. "Efficient embedded sleep wake classification for open-source actigraphy." Scientific Reports 2021 11: 345. https://doi.org/10.1038/s41598-020-79294-y.
Banfi, T., Valigi, N., di Galante, M., d’Ascanio, P., Ciuti, G. and Faraguna, U. (2021) 'Efficient embedded sleep wake classification for open-source actigraphy', Scientific Reports 2021, 11, p. 345. doi: 10.1038/s41598-020-79294-y.
Banfi, Tommaso, et al. "Efficient embedded sleep wake classification for open-source actigraphy." Scientific Reports 2021, vol. 11, 2021, p. 345. https://doi.org/10.1038/s41598-020-79294-y.
T. Banfi, N. Valigi, M. di Galante, P. d’Ascanio, G. Ciuti, and U. Faraguna, "Efficient embedded sleep wake classification for open-source actigraphy," Scientific Reports 2021, vol. 11, p. 345, 2021, doi: 10.1038/s41598-020-79294-y.
@article{banfi2021efficient,
  title         = {{Efficient embedded sleep wake classification for open-source actigraphy}},
  author        = {Banfi, Tommaso and Valigi, Nicolò and di Galante, Marco and d’Ascanio, Paola and Ciuti, Gastone and Faraguna, Ugo},
  year          = {2021},
  journal       = {Scientific Reports 2021},
  volume        = {11},
  pages         = {345},
  doi           = {10.1038/s41598-020-79294-y}
}

History

  1. Added to the ChatPPG research hub