On Events Detection During Anaesthesia via Sparse Piece-Wise Polynomial Normality Characterization
Year: 2026
Type: preprint
Abstract: In this paper, a general framework is proposed for normality characterization of health indicators-related timeseries during anaesthesia. This characterization offers an efficient incidental events detection in datasets collected during surgery with unknown workflows. The resulting framework can therefore be used in facilitating off-line data annotation/cleaning as well as an on-line alarm system in operation rooms. The method leverages a recently proposed implicit identification structure involving piece-wise polynomial relationships and an associated sparse identification algorithm. The proposed framework is tested using a dataset involving 19 patients with events-related annotations by anaesthesiologists.
Source: HAL
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7131014781
