Table 3.
Summary of work and predictions relating to the detection of SZ using data from diffusion-weight MRI, diffusion tensor imaging and perfusion MRI scans via various artificial intelligence techniques and machine learning algorithms.
| Study | Year | Subjects | Prediction | AI/ML Technique | |
|---|---|---|---|---|---|
| Patients | Control | ||||
| Caan et al. [111] | 2006 | 34♂ | 24 | (not reported) | LDA, PCA |
| Caprihan et al. [112] | 2008 | 45 | 45 (age-matched) | 100% | DPCA |
| Ingalhalikar et al. [113] | 2010 | 27♀ | 37♀ | 90.62% | Nonlinear SVM |
| Rathi et al. [114] | 2010 | 21 (FEP) | 20 (age-matched) | SH: 78% (sensitivity) 80% (specificity) F2T: 86% (sensitivity) 85% (specificity) |
K-nearest neighbours, Parzen window classifier, SVM |
| Ardekani et al. [115] | 2011 | 50 | 50 (age- and sex-matched) | FA: 96% (sensitivity) 92% (specificity) MD: 96% (sensitivity) 100% (specificity) |
Fisher’s LDA |
| Squarcina et al. [116] | 2015 | 35 (FEP) | 35 | 83% | SVM |