B4573 - Using statistical and machine learning approaches to predict bone health from physical activity measured with accelerometers - 26/03/2024
B number:
B4573
Principal applicant name:
Louise Millard | Department of Population Health Sciences, Bristol Medical School
Co-applicants:
Dr Gemma Clayton, Professor Jon Tobias, Zhousiying Wu
Title of project:
Using statistical and machine learning approaches to predict bone health from physical activity measured with accelerometers
Proposal summary:
Weight bearing physical activity is known to be beneficial for bone health. Predicting future bone health from accelerometer data may be useful to inform targeted interventions. We will use machine learning and statistical approaches to determine the extent to which bone health can be predicted from physical activity measured with accelerometers.
Impact of research:
This study will improve understanding of the extent that bone health can be predicted from objectively measured physical activity data, and of the types of activity features that are most predictive of subsequent bone health in adolescence.
Date proposal received:
Wednesday, 20 March, 2024
Date proposal approved:
Tuesday, 26 March, 2024
Keywords:
Statistics/methodology, Bone disorders - arthritis, osteoporosis, Statistical methods, Machine learning methods, Bones (and joints), Physical - activity, fitness, function