Machine learning models for hip fracture prediction using electronic medical records: abridged secondary publication
GHY Li1, CL Cheung2, KCB Tan3, TCY Kwok4, WCY Lau5
1 Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Hong Kong SAR, China
2 Department of Pharmacology and Pharmacy, The University of Hong Kong, Hong Kong SAR, China
3 Department of Medicine, The University of Hong Kong, Hong Kong SAR, China
4 Department of Medicine & Therapeutics and School of Public Health, The Chinese University of Hong Kong, Hong Kong SAR, China
5 School of Pharmacy, University College London, United Kingdom
- Ethnicity- and sex-specific hip fracture prediction models were developed using machine learning algorithms and electronic medical records. The performance of the prediction models was validated in independent cohorts, achieving the area under the curve values of >0.8. The prediction models may be clinically useful and generalisable to the public.
- The prediction models were developed without using bone mineral density as a potential predictor, owing to the limited availability of dual-energy X-ray absorptiometry in Hong Kong.