Machine listening in a neonatal intensive care unit - Archive ouverte HAL
Conference Papers Year : 2024

Machine listening in a neonatal intensive care unit

Abstract

Oxygenators, alarm devices, and footsteps are some of the most common sound sources in a hospital. Detecting them has scientific value for environmental psychology but comes with challenges of its own: namely, privacy preservation and limited labeled data. In this paper, we address these two challenges via a combination of edge computing and cloud computing. For privacy preservation, we have designed an acoustic sensor which computes third-octave spectrograms on the fly instead of recording audio waveforms. For sample-efficient machine learning, we have repurposed a pretrained audio neural network (PANN) via spectral transcoding and label space adaptation. A small-scale study in a neonatological intensive care unit (NICU) confirms that the time series of detected events align with another modality of measurement: i.e., electronic badges for parents and healthcare professionals. Hence, this paper demonstrates the feasibility of polyphonic machine listening in a hospital ward while guaranteeing privacy by design.
Fichier principal
Vignette du fichier
tailleur2024dcase.pdf (1.3 Mo) Télécharger le fichier
Origin Files produced by the author(s)
Licence

Dates and versions

hal-04698561 , version 1 (16-09-2024)
hal-04698561 , version 2 (04-10-2024)

Licence

Identifiers

Cite

Modan Tailleur, Vincent Lostanlen, Jean-Philippe Rivière, Pierre Aumond. Machine listening in a neonatal intensive care unit. Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE), Oct 2024, Tokyo, Japan. ⟨hal-04698561v2⟩
0 View
0 Download

Altmetric

Share

More