Machine learning algorithms for predicting the risk of fracture in patients with diabetes in China

AI algorithms for accurate prediction of osteoporotic fractures in patients with diabetes: an up-to-date review

“…In the following year, two exciting studies emerged. Chu et al [ 57 ] developed Probabilistic Classification Vector Machines (PCVM) algorithm to construct risk prediction models for fractures apart from the 6 algorithms LR, SVM, RF, DT, GBDT, XGBoost, which were conducted in the research above. What is delightful is that PCVM achieved the best f1 scores (0.97), surpassing LR (0.75), SVM (0.83), RF (0.84), DT (0.85), GBDT (0.87), XGBoost (0.88).…”

“…What is delightful is that PCVM achieved the best f1 scores (0.97), surpassing LR (0.75), SVM (0.83), RF (0.84), DT (0.85), GBDT (0.87), XGBoost (0.88). Research of Chen et al [ 40 ] determined 18 influencing factors of fracture risks of patients with diabetes while Chu et al [ 57 ] determined 17 influencing factors, these influencing factors were easy to obtain and do not require precise inspection. To predict the risk of hip fractures in a more accurate way, Yosibash et al [ 58 ] developed a ML algorithm with autonomous finite element analyses (AFE) based on CT scans for hip fracture risk assessment in type 2 diabetic mellitus (T2DM).…”