Table 1.
Systematic Review of Diagnostic Test Accuracy of Different Machine Learning Algorithms for Breast Cancer Risk Calculation
| First Author, Year | Size of Dataset (n) | Dataset | Machine Learning Algorithms | Accuracy (%) | NOS |
|---|---|---|---|---|---|
| Chang et al., 2003 | 250 | Primary data (pathologically proved breast tumors) | Super Vector Machine | 85.6 | 7 |
| Polat and Gunes, 2007 | 683 | Wisconsin breast cancer dataset | Super Vector Machine | 95.89 | 7 |
| Akay, 2009 | 683 | Wisconsin breast cancer dataset | Super Vector Machine | 99.51 | 7 |
| Ayer et al., 2010 | 62,219 | Wisconsin state cancer reporting system | Artificial Neural Networks | 96.5 | 8 |
| Dramicanin et al., 2012 | 42 | Primary data (breast tissue specimens) | Super Vector Machine | 64.29 | 6 |
| Subramanian et al., 2014 | 40 | Primary data (mammographic image) | a. Super Vector Machine b. Artificial Neural Networks c. Decision Tree d. Naive Bayes |
a. 62.5 b. 75 c. 67.5 d. 75 |
6 |
| Mert et al., 2015 | 569 | Wisconsin diagnostic breast cancer dataset | a. K-Nearest Neighbor b. Artificial Neural Networks c. Radial Basis Function Neural Network d. Super Vector Machine |
a. 93.14 b. 97.53 c. 87.17 d. 95.25 |
7 |
| Milosevic et al., 2015 | 300 | The Mini Mammographic Database | a. Super Vector Machine b. K-Nearest Neighbor c. Naive Bayes |
a. 83.7 b. 54.3 c. 77.3 |
7 |
| Sun et al., 2015 | 340 | Primary data (digital mammograms) | Super Vector Machine | 72.9 | 7 |
| Asri et al., 2016 | 699 | Wisconsin breast cancer dataset | a. Support Vector Machine b. Decision Tree c. Naive Bayes d. K-Nearest Neighbors |
a. 97.13 b. 95.13 c. 95.99 d. 95.27 |
8 |
| Heidari et al., 2018 | 500 | Primary data (full-field digital mammography) | Super Vector Machine | 60.8 | 7 |
NOS, Newcastle–Ottawa Quality Assessment Scale