Three-dimensional, clinically rated posture data from people aged 10 to 69 years

Area of Science:

• Biomedical Engineering
• Public Health
• Data Science

Background:

• Weak posture is a prevalent issue across all age groups, contributing to back pain and imposing a significant socio-economic burden.
• Early detection of postural deficiencies through posture assessment is crucial for proactive interventions and public health promotion.
• Existing methods for posture assessment may lack objectivity, highlighting the need for data-driven approaches.

Purpose of the Study:

• To provide a comprehensive dataset of spinal posture parameters for a diverse age range (10-69 years).
• To enable the development of objective, data-driven methods for posture assessment.
• To facilitate the creation of machine learning applications for diagnostic support in identifying spinal conditions.

Main Methods:

• Stereophotogrammetry was used to measure posture in 1,149 subjects.
• Raw and normalized sagittal posture parameters (flèche cervicale, flèche lombaire, kyphosis index) were calculated.
• Biomedical experts classified spinal conditions (hyperkyphosis, hyperlordosis), with algorithmic quality checks and expert reassessment for label refinement.

Main Results:

• A dataset including anthropometrics, raw and normalized posture parameters (FC, FL, KI, FC%, FL%, KI%), and expert-validated classifications of hyperkyphosis and hyperlordosis is presented.
• The data encompasses measurements from 1,149 individuals aged 10 to 69 years.
• Refined labels, resulting from algorithmic quality checks and expert reassessment, are provided alongside original expert ratings.

Conclusions:

• The presented posture data is valuable for developing objective posture assessment techniques.
• The dataset has significant potential for the creation of machine learning models to support the diagnosis of postural abnormalities.
• This work contributes to advancing public health by enabling more accurate and objective methods for evaluating spinal health.