Feasibility of computer-assisted diagnosis for breast ultrasound: the results of the diagnostic performance of S-detect from a single center in China

Figures & data

Figure 1 The schematic representation of the study flow.

Abbreviation: US, ultrasound.

Table 1 The pathological types of the 266 breast lesions

Pathological type	Number	Ratio (%)
Malignant lesions
Intraductal carcinoma	10	9.52
Invasive ductal carcinoma, not otherwise specified	80	76.2
Invasive lobular carcinoma	6	5.71
Apocrine carcinoma	2	1.90
Invasive micropapillary carcinoma	1	0.95
Mucinous carcinoma	2	1.90
Neuroendocrine intraductal carcinoma	1	0.95
Adenocarcinoma with spindle cell metaplasia	2	1.90
Diffuse large B-cell lymphoma	1	0.95
Total	105
Benign lesions
Adenosis	37	23.0
Adenomas	97	60.2
Intraductal papillomas	20	12.4
Lobular tumor	1	0.62
Chronic inflammation	5	3.11
Adiponecrosis	1	0.62
Total	161

Figure 2 The ROC for the experienced radiologists, S-detect, and the in-training resident.

Abbreviation: ROCs, receiver operating characteristic curves.

Figure 3 The breast lesion of a 51-year-old woman, without any symptoms, which proved to be ductal carcinoma in situ.

Note: (A) The longitudinal section of the lesion; (B) the cross-section of the lesion; (C) the abundant peripheral blood vessels of the tumor; and (D) the lesion was diagnosed by S-detect as a possibly benign tumor.

Table 2 The diagnostic performance of the experienced radiologists, S-detect, and the in-training resident

Rater	SE (%)	SP (%)	PLR	NLR	PPV (%)	NPV (%)	Accuracy (%)	AUC
	95% CI	95% CI	95% CI	95% CI	95% CI	95% CI	95% CI	95% CI
S-detect	87.07	72.27	3.14	0.18	79.5	81.9	80.45	0.807
	88.55–92.04	63.32–80.08	2.33–4.22	0.12–0.28	72.44–85.45	73.19–88.74	75.25–84.79	0.751–0.863
Radiologist	98.1	65.43	2.84	0.03	64.78	98.15	78.20	0.817
	93.29–99.77	57.57–72.72	2.29–3.51	0.01–0.12	56.82–72.18	93.47–99.77	72.84–82.75	0.766–0.867
Resident	82.86	68.94	2.67	0.25	63.5	86.05	74.44	0.758
	74.27–89.51	61.18–75.99	2.09–3.41	0.16–0.38	54.85–71.56	78.85–91.52	68.86–79.32	0.699–0.819

Abbreviations: AUC, area under the receiver operator characteristics curve; NLR, negative likelihood ratio; NPV, negative predictive value; PLR, positive likelihood ratio; PPV, positive predictive value; SE, sensitivity; SP, specificity.

Table 3 The subcategorization of breast lesions by an experienced radiologist

BI-RADS categories classified by an experienced radiologist	Pathological result	S-detect		Total	P-value
		Possibly benign	Possibly malignant
BI-RADS 3	Benign	94	11	105	0.0010
	Malignant	0	2	2
BI-RADS 4a	Benign	25	15	40	0.0192
	Malignant	4	2	6
BI-RADS 4b	Benign	7	5	12	>0.05
	Malignant	5	8	13
BI-RADS 4c	Benign	2	2	4	>0.05
	Malignant	7	33	40
BI-RADS 5	Benign	0	0	0
	Malignant	3	44	47	>0.05

Abbreviation: BI-RADS, Breast Imaging Report and Data System.

Figure 4 The breast lesion of a 46-year-old woman, without any symptoms, which proved to be chronic inflammation by histology.

Note: (A) The longitudinal section of the lesion; (B) the cross-section of the lesion; (C) the abundant peripheral blood vessels of the tumor; and (D) the lesion was diagnosed by S-detect as a possibly benign tumor.

Figure 5 The breast lesion of a 35-year-old woman who found a palpable breast mass with a diameter of 2 cm 6 months ago, whereas it measured as 4 cm in this US examination. The pathological result was mucinous carcinoma.

Note: (A) The longitudinal section of the lesion; and (B) the cross-section of the lesion; (C, D) the lesion was misdiagnosed by S-detect as a possibly benign tumor.

Figure 6 The breast lesion of a 41-year-old woman with a history of invasive phyllodes tumor.

Note: (A) The longitudinal section of the lesion; (B) the cross-section of the lesion; and (C, D) the lesion was misdiagnosed by S-detect as a possibly benign tumor.