Bestimmung der Ausdauerleistung durch ein Rekurrentes Neuronales Netzwerk

Details

Veröffentlichungs-Nr.

EP4715835

Anmeldetag

18. September 2024

Veröffentlichung

25. März 2026

Rechtsraum

EP

Abstract

Example embodiments relate to a computer-implemented method for determining endurance performance of a subject based on sensor measurements; the computer-implemented method comprising: obtaining (201) a training dataset (214) comprising workload data (212) and heart rate data (211) measured, by respectively at least one workload sensor and at least one heart rate sensor, during respective physical activities performed by the subject within a historical window; training (202) a recurrent neural network, RNN (220), based on the training dataset (214); wherein the RNN (220) is configured to determine the heart rate (221) of the subject at an exerted workload (222) considering a historic time series of heart rate data (223) and workload data preceding the exerted workload; and determining (203) at least a portion of a workload-heart rate relationship (263), indicative for the endurance performance of the subject, by feeding one or more fixed workload time series (241, 242, 243) to the trained RNN (220).

Anmelder

Firmen

Imec VZW
Universiteit Antwerpen

Land

🇧🇪 Belgien