Table 2

Summary of the different deep learning methods by applications in gastric problems

Based method	Task	Application	Reference
Convolutional neural network	Classification	Benign or malign of stomach biopsy specimens	79
		Benign or malignant images	58
		Gastric cancer or non-cancer	32
		Gastric cancer or non-cancer	64
		Gastritis or non-gastritis	34
		Helicobacter pylori-positive or helicobacter pylori-negative	38
			39
		Helicobacter pylori–related gastritis, reactive gastropathy and histologically normal gastric mucosa	40
		Neoplasm or non-neoplasm	80
		Normal gastric images or early gastric cancer images	60
		Normal or abnormal gastric slow wave	61
		With and without histology-proven atrophic gastritis	35
Convolutional neural network and residual neural network		Benign ulcer and gastric cancer	42
Convolutional neural network and residual neural network		Early gastric cancer, advanced gastric cancer, high grade dysplasia, low grade dysplasia or non-neoplasm	41
Convolutional neural network with deep generalised multi-instance learning		Differentiation degree (poorly and well/moderately) and lauren type (intestinal, diffuse and mixed)	81
Convolutional neural network and deep reinforcement learning		Gastric sites	62
Residual neural network		Gastric cancer type (intestinal type or diffuse type)	52
Residual neural network		Microsatellite instable or microsatellite stability	53
Recurrent neural network		Live or dead probability	51
Convolutional neural network	Classification/detection	Benign or malignant gastric ulcer/gastric ulcer	43
	Classification/detection	HER2+ tumour, HER2 tumour or non-tumour/necrosis detection	5
	Detection	Gastric cancer	65
		Gastritis or non-gastritis	36
		Gastritis or non-gastritis	37
		Lymphocyte or non-lymphocyte	66
		Normal mucosa, non-cancerous pathology, cancer	63
		Signet ring cell cancer	29
		Signet-ring cell carcinoma component intramucosal or advanced	82
		Gastric ulcer	43
		Necrosis detection	5
Generative adversarial network	Generation	Gastritis image generation	56
Generative adversarial network	Generation	Gastritis image generation	57
Residual neural network	Segmentation	Gastric cancer	54
Fully convolutional network		Gastric tumour	48
Fully convolutional network		Recognise small cancerous tissues	49