Table 5.

Summary of 2D CNN classification approaches on onset of symptomatic osteoarthritis.

Publication reference	Target tasks	Modality (imaging sequence)	Data set	Network architecture	Performance
Pierson et al. [56]	Predict knee pain	X-ray (plain radiography)	OAI: 4,172 subjects (2877 training, 1295 validation)	CNN	AUC: 0.69
Guan et al. [22]	Predict knee pain	X-ray (plain radiography)	OAI: 2000 subjects (1500 testing, 200 validation, 300 testing)	YOLO + DenseNet	AUC: 0.753; SN: 65.77; SP: 73.51
Chang et al. [14]	Predict knee pain	MRI (SAG-IW-TSE)	OAI: 1505 subjects (training : testing; 90% : 10%)	Siamese network	AUC: 0.808

Note. Modality (imaging sequence): magnetic resonance imaging (MRI); data set: Osteoarthritis Initiative (OAI); network architecture: convolutional neural network (CNN); performance: specificity (SP), sensitivity (SN), and area under receiver operating characteristics curve (AUC).