Table 2.

Reasons of inconsistencies among BI-RADS categories.

Reference	Test	Reason	Frequency
The reasons of the differences among BI-RADS categories for Reference data set and Test set
B0	B1–2	Lymph node and “Stability” information of the lesion was not detected by NLP Lymph node was not detected by NLP	1 2
B0	B4	Lymph node and “Stability” information of the lesion was not detected by NLP “Density” and “Shape” information of the lesion was not detected by NLP	1 1
B1–2	B3	Skin Thickening and “Stability” information of the lesion was not detected by NLP	1
B1–2	B4	Skin Lesion was not detected by NLP	1
B3	B1–2	Calcification was not found by NLP	3
B4	B1–2	Calcification was not found by NLP	3