Table 2.

Comparison of different machine learning models on the performance in estimation OV infectious level grading. The best performance and the next best performance are indicated in bold and italic respectively. represents a classification model and represents a regression model. The proposed models that outperform the most performant baseline, ResNet50 () are marked with .

Task	Models	Performance (± s.d)
		Accuracy	Recall	Precision	F1-score
OV-RDT grading (0, +1, +2, +3, +4)	SVM	0.48 ± 0.07	0.48 ± 0.07	0.50 ± 0.06	0.48 ± 0.07
	RF	0.44 ± 0.06	0.44 ± 0.06	0.46 ± 0.07	0.44 ± 0.06
	MobileNetV2	0.48 ± 0.05	0.48 ± 0.05	0.49 ± 0.06	0.46 ± 0.05
	MobileNetV2	0.55 ± 0.06	0.55 ± 0.06	0.59 ± 0.06	0.55 ± 0.06
	ResNet50	0.58 ± 0.07	0.58 ± 0.07	0.57 ± 0.09	0.57 ± 0.07
	ResNet50	0.58 ± 0.07	0.58 ± 0.07	0.61 ± 0.08	0.58 ± 0.07
	EffNet-B5 (Ours)	0.63 ± 0.06	0.63 ± 0.06	0.64 ± 0.06	0.62 ± 0.06
	EffNet-B5 (Ours)	0.66	0.66	0.68	0.66
OV-RDT Status (Negative (0), Positive (1–4))	SVM	0.84 ± 0.04	0.84 ± 0.04	0.87 ± 0.04	0.85 ± 0.04
	RF	0.78 ± 0.04	0.78 ± 0.04	0.81 ± 0.04	0.79 ± 0.04
	MobileNetV2	0.87 ± 0.04	0.87 ± 0.04	0.91 ± 0.03	0.88 ± 0.04
	MobileNetV2	0.91 ± 0.04	0.91 ± 0.04	0.92 ± 0.03	0.91 ± 0.03
	Resnet50	0.90 ± 0.04	0.90 ± 0.04	0.93 ± 0.02	0.90 ± 0.03
	ResNet50	0.92 ± 0.03	0.92 ± 0.03	0.93 ± 0.03	0.92 ± 0.03
	EffNet-B5 (Ours)	0.95	0.95	0.95	0.95
	EffNet-B5 (Ours)	0.95	0.95	0.96	0.95