Please use this identifier to cite or link to this item:
|Title:||Predicting VO2max by machine learning models before volitional fatigue during incremental exercise|
|Abstract:||Maximal oxygen uptake (VO2max) is one of the most important clinical indexes that is related to all-cause mortality. However, is only obtained at maximal cardiopulmonary exercise testing (CPET). The main purpose of this study was to predict the VO2max during the early stages of the CPET by machine learning. Forty-five healthy participants (27±5 years old, 69±11 kg and 170±8 cm) performed a CPET in a cycle-ergometer until exhaustion. The time of exercise, workload, variables from the metabolic cart, and the heart rate were recorded throughout the CPET. Variables were combined into clusters according to the time of the CPET where cluster 1 contained data related to the first sec of the CPET, cluster 2 data related to sec 1 and 2, and so on. These variables were considered as inputs for the VO2max prediction for each cluster by Support Vector Machines (Fig 1). The prediction quality was evaluated by the relative error between the measured and predicted VO2max. After 316 sec (52% less of the total CPET average time), VO2max was predicted with an error of 10%. Therefore, machine learning models can be used to predict the VO2max before volitional fatigue. Future studies may explore the use of these models in patient population which would increase the practical applicability of CPET|
|Editor:||European Respiratory Society|
|Appears in Collections:||IC - Artigos e Outros Documentos|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.