Using Machine Learning to Diagnose Osteoporosis Risk in Adults
Author Information
Author(s): Tabib Saghar, Alizadeh Seyed Danial, Andishgar Aref, Pezeshki Babak, Keshavarzian Omid, Tabrizi Reza
Primary Institution: Fasa University of Medical Sciences
Hypothesis
Can machine learning techniques accurately identify osteoporosis risk factors in the Fasa Adults Cohort Study?
Conclusion
The XGB model had the best performance in assessing the risk of osteoporosis in the Iranian population.
Supporting Evidence
- The XGB model achieved an AUC of 0.78, indicating good performance.
- Age, calcium intake, and red blood cell count were identified as key risk factors.
- The study included a large sample size of over 10,000 participants.
- Machine learning methods showed potential for early detection of osteoporosis.
- Traditional diagnostic tools are often impractical in developing countries.
Takeaway
Researchers used computers to help find out who might have osteoporosis, a disease that makes bones weak. They found that a special computer method worked best for this.
Methodology
The study analyzed data from the Fasa Adults Cohort Study using eight machine learning methods to evaluate osteoporosis risk.
Potential Biases
Self-reported data may lead to underestimation of osteoporosis prevalence.
Limitations
The study relied on self-reported data, which may introduce errors and affect the accuracy of results.
Participant Demographics
Participants included 10,108 individuals aged 35 to 70, with a male-to-female ratio of 0.8:1.
Statistical Information
P-Value
p<0.05
Confidence Interval
95% CI: 0.74–0.82
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website