Automatic recognition of tumour-infiltrating lymphocytes in pathological biopsy images of the gastric mucosa

Table 1 Performance indicators of the gastric-tumour-infiltrating lymphocytes-convolutional neural network model in the training set and test set (%)

Dataset	Accuracy (95%CI)	Specificity (95%CI)	Sensitivity (95%CI)	Cohen’s Kappa (95%CI)	F1 score (95%CI)	Compared with manual assessment (P value)
Training	99.5 (99.2-99.8)	99.8 (99.6-100.0)	99.3 (98.9-99.7)	0.99 (0.98-1.00)	99.5 (99.2-99.8)	< 0.001
Testing	99.2 (98.8-99.6)	100.0 (99.9-100.0)	98.7 (98.1-99.3)	0.98 (0.97-0.99)	99.3 (98.9-99.7)

P < 0.001 vs manual assessment. The training set contains 58240 valid image patches, and the test set contains 31104 valid image patches. The manual assessment accuracy was 86.3%, with a kappa coefficient of 0.72. CI: Confidence interval.

Table 2 Confusion matrix of the gastric-tumour-infiltrating lymphocytes-convolutional neural network model on the test set, n (%)

Actual label/predicted label	TIL	Non-TIL	Total	Recall (%)	Miss rate (%)	TNR (%)
TIL	7632 (24.5)	98 (0.3)	7730 (24.8)	98.7	1.3	-
Non-TIL	0 (0.0)	23374 (75.2)	23374 (75.2)	100.0	0.0	100.0
Total	7632 (24.5)	23472 (75.5)	31104 (100)	-	-	-
Precision (%)	100.0	99.6	-	-	-	-
False discovery rate (%)	0.0	0.4	-	-	-	-

This table is based on statistics from 31104 valid image patches in the test set. TNR: True negative rate (specificity); TIL: Tumour-infiltrating lymphocyte.

Table 3 Tumour-infiltrating lymphocyte recognition performance of the gastric-tumour-infiltrating lymphocytes-convolutional neural network for different gastric mucosal lesions (%)

Lesion type	Accuracy (95%CI)	Specificity (95%CI)	Sensitivity (95%CI)
Chronic atrophic gastritis	99.1 (98.5-99.7)	99.5 (99.1-99.9)	98.8 (98.0-99.6)
Intestinal metaplasia	98.6 (97.8-99.4)	99.2 (98.6-99.8)	97.9 (96.8-99.0)
High-grade intraepithelial neoplasia	99.3 (98.7-99.9)	99.7 (99.3-100.0)	99.0 (98.2-99.8)
Early gastric cancer	99.5 (99.0-100.0)	100.0 (99.8-100.0)	99.1 (98.4-99.8)

P < 0.05. This table is based on the statistics of 96 samples in the test set, with 24 cases for each lesion type. The accuracy of the intestinal metaplasia group was the lowest, mainly because of the slight misjudgment caused by the similar morphology between the goblet cells and scattered tumour-infiltrating lymphocytes. CI: Confidence interval.

Table 4 Ablation experiment performance indicators of the gastric-tumour-infiltrating lymphocytes-convolutional neural network model (independent test set)

Ablation experiment type	Experimental group	Accuracy (95%CI)	Specificity (95%CI)	Sensitivity (95%CI)	Cohen’s Kappa	P value (vs optimal group)
Feature fusion weight	Dynamic learning weight (0.4/0.3/0.3)	99.2 (98.8-99.6)	100.0 (99.9-100.0)	98.7 (98.1-99.3)	0.98	< 0.001
Feature fusion weight	Equal weight (0.33/0.33/0.33)	96.5 (95.8-97.2)	97.8 (97.1-98.5)	95.1 (94.2-96.0)	0.89
Input channel type	2-channel H/E (Colour Deconvolution)	99.2 (98.8~99.6)	100.0 (99.9~100.0)	98.7 (98.1-99.3)	0.98	< 0.001
Input channel type	3-channel RGB	94.7 (93.9-95.5)	92.3 (91.2-93.4)	95.1 (94.0-96.2)	0.87

P < 0.001 vs optimal group. The optimal group was the dynamic learning weight reassignment/2-channel haematoxylin/eosin group, which was the final model adopted. H/E: Haematoxylin/eosin; RGB: Red green blue (red, green, blue channels); CI: Confidence interval.

Table 5 Multivariate Cox proportional hazards regression analysis of prognostic factors in patients with early gastric cancer

Prognostic Index	Variable	HR	95%CI	P value
DFS	G-AI-TIL ≥ 28.5% (vs < 28.5%)	0.58	0.37-0.91	0.018
	Age (per 1-year increase)	1.04	1.02-1.06	< 0.001
	Tumour invasion depth (submucosa vs mucosa)	1.87	1.12-3.12	0.016
	Ulcers (present vs absent)	1.32	0.81-2.15	0.257
OS	G-AI-TIL ≥ 28.5% (vs < 28.5%)	0.55	0.35-0.86	0.009
	Age (per 1-year increase)	1.05	1.03-1.07	< 0.001
	Tumour invasion depth (submucosa vs mucosa)	1.93	1.15-3.24	0.013
	Ulcers (present vs absent)	1.41	0.85-2.34	0.180

Hazard ratio (HR) < 1 indicates an independent protective factor, and HR > 1 indicates an independent risk factor; Variable assignment: Gastric artificial intelligence-based tumor-infiltrating lymphocytes (< 28.5% = 0, ≥ 28.5% = 1), depth of tumour invasion (mucosal layer = 0, submucosal layer = 1), ulcer (absent = 0, present = 1), age is a continuous variable (per 1-year increase). DFS: Disease-free survival; OS: Overall survival; HR: Hazard ratio; CI: Confidence interval; G-AI-TIL: Gastric artificial intelligence-based tumor-infiltrating lymphocytes.