In the field of sport/health, the prescription is generic and individualisation, which is still very rare, is generally linked only to physical performance. These limitations on the implementation of the sessions will result in limiting the effects of the training programme and increasing the risk of injury. It is therefore necessary to develop knowledge and tools to assist physicians and physical activity professionals in their decision making. The aim of the study is to improve the precision of the calculation of the training load in order to better individualise the management of the participants.
Physical activity has been recommended for many years by various health authorities. Indeed, numerous studies have demonstrated the effects of physical activity on the quality of life, the gain of autonomy and the reduction of functional limitations in the elderly. This problem is becoming more and more present with the increase in life expectancy, which increases the prevalence of individuals reaching the status of elderly person. These forecasts suggest that more and more people will have to live with the consequences of age. From this point of view, the management of these people through adapted physical activity is necessary to relieve the health system by limiting the risks of frailty and thus hospitalisation among the elderly. In order to promote adapted physical activity, the role of specialists in this field is to set up physical activity sessions that solicit the neurocognitive, neuromuscular and cardiorespiratory aspects through balance, muscle strengthening and aerobic exercises. These sessions will be defined according to different criteria such as pathology, level of accomplishment of the exercises and the perceived difficulty of the session. For participants, the most common method of quantification is the Borg Rating of Perceived Exertion (RPE) scale. This measure consists of measuring the perception of effort on a scale ranging from 6 for "no perceived effort" to 20 for "maximum effort". The use of this tool has the advantage of having a strong correlation with heart rate at moderate intensities. These different parameters will allow the adapted physical activity engineer to determine a training load according to these different parameters. This training load will then allow the next session(s) to be adjusted to best suit the functional capacities of the participants and thus individualise the treatment of these people. However, this training load is subjective. For professionals in adapted physical activity, the quantification of the training load of the different sessions is based on personal experience. This method is difficult to quantify and generalise to other professionals. Similarly, the feeling of the difficulty of the session is specific to each participant. The quantification of the RPE will be strongly impacted by the last exercises of the session and the group effect. To address this limitation, various tools have been developed. One of the most common solutions is the use of a heart rate monitor. The advantage of this sensor is the quantification of heart rate during the session without the subjective inter-individual bias. However, this method of quantification has a poor correlation with the training load of muscle strengthening. This limitation has a strong impact on the estimation of adapted activity sessions due to the large proportion of exercises focused on the neuromuscular aspect. To address this issue, the analysis of human movement would be a complementary tool to estimate energy expenditure according to the different types of exercises proposed during the sessions. This analysis has been greatly facilitated with the arrival of deep learning for the classification of different movements according to the kinematics of the movement or actimetry. In this context two measurement systems stand out. On the one hand, the Actigraph is one of the most widely used tools in this field to quantify the training load according to the magnitude vector calculated by the accelerometer. This training load will be identified as low, moderate or high depending on the activities recorded by the sensor. On the other hand, the use of depth cameras to segment the human body is a rapidly evolving area. The data collected with this measurement system coupled with deep learning can classify a wide range of movements on several individuals at the same time. The use of these measurement systems would allow a better understanding of the estimation of the workload. Indeed, the association of the classification of exercises with the time of execution and the number of repetitions coupled with the heart rate would make it possible to have an accurate energy expenditure in ecological conditions. These data would allow for more precise individualisation of the management of elderly people for adapted physical activity sessions. This type of personalised care for frailty would be part of the future challenges of ageing well.
Study Type
OBSERVATIONAL
Enrollment
87
A group of participants with physical activity (protocolized)
GUERIN
Nice, France
Training Load
The training load measured after each physical activity session measured with the different methods. The training load will be estimate with: the number of sets x the number of repetitions x the intensity of the exercise. The measure of the intensity of the exercises will be compute from vector magnitude from actimetry and motion capture. For the control method, the intensity of the exercise will be estimate by adapted physical activity engineers.
Time frame: Measured at day 0
Training Load
The training load measured after each physical activity session measured with the different methods. The training load will be estimate with: the number of sets x the number of repetitions x the intensity of the exercise. The measure of the intensity of the exercises will be compute from vector magnitude from actimetry and motion capture. For the control method, the intensity of the exercise will be estimate by adapted physical activity engineers.
Time frame: Measured at day 3
Training Load
The training load measured after each physical activity session measured with the different methods. The training load will be estimate with: the number of sets x the number of repetitions x the intensity of the exercise. The measure of the intensity of the exercises will be compute from vector magnitude from actimetry and motion capture. For the control method, the intensity of the exercise will be estimate by adapted physical activity engineers.
Time frame: Measured at day 7
Training Load
The training load measured after each physical activity session measured with the different methods. The training load will be estimate with: the number of sets x the number of repetitions x the intensity of the exercise. The measure of the intensity of the exercises will be compute from vector magnitude from actimetry and motion capture. For the control method, the intensity of the exercise will be estimate by adapted physical activity engineers.
Time frame: Measured at day 10
Training Load
The training load measured after each physical activity session measured with the different methods. The training load will be estimate with: the number of sets x the number of repetitions x the intensity of the exercise. The measure of the intensity of the exercises will be compute from vector magnitude from actimetry and motion capture. For the control method, the intensity of the exercise will be estimate by adapted physical activity engineers.
Time frame: Measured at day 14.
Training Load
The training load measured after each physical activity session measured with the different methods. The training load will be estimate with: the number of sets x the number of repetitions x the intensity of the exercise. The measure of the intensity of the exercises will be compute from vector magnitude from actimetry and motion capture. For the control method, the intensity of the exercise will be estimate by adapted physical activity engineers.
Time frame: Measured at day 17
Training Load
The training load measured after each physical activity session measured with the different methods. The training load will be estimate with: the number of sets x the number of repetitions x the intensity of the exercise. The measure of the intensity of the exercises will be compute from vector magnitude from actimetry and motion capture. For the control method, the intensity of the exercise will be estimate by adapted physical activity engineers.
Time frame: Measured at day 21
Training Load
The training load measured after each physical activity session measured with the different methods. The training load will be estimate with: the number of sets x the number of repetitions x the intensity of the exercise. The measure of the intensity of the exercises will be compute from vector magnitude from actimetry and motion capture. For the control method, the intensity of the exercise will be estimate by adapted physical activity engineers.
Time frame: Measured at day 24.
Exercises Intensity
Assessment of exercises intensity will be quantify by the Rate Perceived Exertion (RPE). The RPE scale ranges from 0 to 10. The number indicates how easy or difficult the participant finds an activity. For example, 2-3 is a light activity and 10 is a maximal effort activity.
Time frame: Measured at day 0, day 3, day 7, day 10, day 14, day 17, day 21, day 24.
Exercises classification
To evaluate the prediction of the exercises, we will analyse the accuracy of the measurement with a confusion matrix on the prediction of the movements that decompose the different exercises. This matrix will allow us to determine the accuracy of the classifier according to the rate of false positives and the rate of false negatives. accuracy = (TP + TN) / (TP + TN + FP + FN) Where TP = number of True Positives, TN = number of True negatives, FP = number of False Positives, and FN = number of False Negatives.
Time frame: Measured at day 0, day 3, day 7, day 10, day 14, day 17, day 21, day 24.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.