Skip to main content

Table 7 Main studies in the use of artificial intelligence as an aid to the diagnosis of osteoporosis

From: Artificial intelligence on the identification of risk groups for osteoporosis, a general review

Author/article Conventional method
Y AI % VAR PAC Country Gender
Kung et al. [50] 2002 OSTA 91.0 22 722 China F
Rizzi et al. [15] 2004 MoG N/A 3 845 Italy F
Wenjia et al. [16] 2005 Hybrid 85.7 5 2.158 Iran F
Chiu et al. [18] 2006 ANN 79.2 7 1.403 Taiwan M/F
Leslie et al. [23] 2009 Algorithm 93.3 5 4.015 Canada F
Mantzaris et al. [24] 2010 LVQ 96.6 4 3.426 Greece M/F
Cos Juez et al. [25] 2010 MLP 97.9 10 200 Spain F
Jennane et al. [44] 2012 SVM 87.0 20 69 Argentina F
Harrar et al. [26] 2012 MLP 97.0 5 120 France F
Yoo et al. [27] 2013 SVM 76.7 11 1674 South Korea F
Anburajan et al. [28] 2013 SVM 90.0 5 50 India F
Kavitha et al. [29] 2013 SVM 91.8 3 100 Japan F
Tafraouti et al. [30] 2014 SVM 93.0 16 77 France M/F
Iliou et al. [31] 2015 MLP 83.0 35 589 Greece M/F
Liu et al. [32] 2015 MLP 93.0 10 725 Taiwan M/F
Xinghu et al. [33] 2016 ANN 95.0 17 119 China M/F
  1. SVM support vector machines, RF random forests, ANN artificial neural networks, MoG mixture of Gaussian, OSTA Osteoporosis Self-Assessment Tool for Asian, PNN probabilistic neural network, LVQ learning vector quantization, MLP Multilayer Perceptron, HAC histogram-based automatic clustering, M masculine, F feminine, Y year, AI artificial intelligence, % precision, VAR amount of variables, PAC number of patients