Automatic Cardiopulmonary Endurance Assessment: A Machine Learning Approach Based on GA-XGBOOST

OBJECTIVE

Among various assessment paradigms, the cardiopulmonary exercise test (CPET) provides rich evidence as part of the cardiopulmonary endurance (CPE) assessment. However, methods and strategies for interpreting CPET results are not in agreement. The purpose of this study is to validate the possibility of using machine learning to evaluate CPET data for automatically classifying the CPE level of workers in high-latitude areas.

METHODS

A total of 120 eligible workers were selected for this cardiopulmonary exercise experiment, and the physiological data and completion of the experiment were recorded in the simulated high-latitude workplace, within which 84 sets of data were used for XGBOOST model training and36 were used for the model validation. The model performance was compared with Support Vector Machine and Random Forest. Furthermore, hyperparameter optimization was applied to the XGBOOST model by using a genetic algorithm.

RESULTS

The model was verified by the method of tenfold cross validation; the correct rate was 0.861, with a Micro-F1 Score of 0.864. Compared with RF and SVM, all data achieved a better performance.

CONCLUSION

With a relatively small number of training samples, the GA-XGBOOST model fits well with the training set data, which can effectively evaluate the CPE level of subjects, and is expected to provide automatic CPE evaluation for selecting, training, and protecting the working population in plateau areas.

Improvement in exercise capacity and delayed anaerobic metabolism induced by far-infrared-emitting garments in active healthy subjects: A pilot study

Background Far-infrared-emitting garments have several biological properties including the capability to increase blood perfusion in irradiated tissues. Design The aim of the study was to evaluate whether far-infrared radiation increases exercise capacity and delays anaerobic metabolism in healthy subjects. Methods With a double-blind, crossover protocol, a maximal cardiopulmonary exercise test was performed in 20 volunteers, wearing far-infrared or common sport clothes, identical in texture and colour. Results Comparing far-infrared with placebo garments, higher oxygen uptake at peak of exercise and longer endurance time were observed (peak oxygen uptake 38.0 ± 8.9 vs. 36.2 ± 8.5 ml/kg/min, endurance time 592 ± 85 vs. 570 ± 71 seconds; P < 0.01); the anaerobic threshold was significantly delayed (anaerobic threshold time 461 ± 93 vs. 417 ± 103 seconds) and anaerobic threshold oxygen uptake and anaerobic threshold oxygen pulse were significantly higher (25.3 ± 6.4 vs. 20.9 ± 5.4 ml/kg/min and 13.3 ± 3.8 vs. 12.4 ± 3.3 ml/beat, respectively). In 10 subjects the blood lactate concentration was measured every 2 minutes during exercise and at peak; lower values were observed with far-infrared fabrics compared to placebo from the eighth minute of exercise, reaching a significant difference at 10 minutes (3.6 ± 0.83 vs. 4.4 ± 0.96 mmol/l; P = 0.02). Conclusions In healthy subjects, exercising with a far-infrared outfit is associated with an improvement in exercise performance and a delay in anaerobic metabolism. In consideration of the acknowledged non-thermic properties of functionalised clothes, these effects could be mediated by an increase in oxygen peripheral delivery secondary to muscular vasodilation. These data suggest the need for testing far-infrared-emitting garments in patients with exercise limitation or in chronic cardiovascular and respiratory patients engaged in rehabilitation programmes.

Biology of VO2 max: looking under the physiology lamp

In this review, we argue that several key features of maximal oxygen uptake (VO₂ max) should underpin discussions about the biological and reductionist determinants of its interindividual variability: (i) training-induced increases in VO₂ max are largely facilitated by expansion of red blood cell volume and an associated improvement in stroke volume, which also adapts independent of changes in red blood cell volume. These general concepts are also informed by cross-sectional studies in athletes that have very high values for VO₂ max. Therefore, (ii) variations in VO₂ max improvements with exercise training are also likely related to variations in these physiological determinants. (iii) All previously untrained individuals will respond to endurance exercise training in terms of improvements in VO₂ max provided the stimulus exceeds a certain volume and/or intensity. Thus, genetic analysis and/or reductionist studies performed to understand or predict such variations might focus specifically on DNA variants or other molecular phenomena of relevance to these physiological pathways.