Table 4.

– Comparison of performance of different systems reported on the same i2b2–2010 corpus

Publications	Models	Precision (%)	Recall (%)	F-measure (%)
Rink et al. [33]	SVM	67.44	57.85	59.31
Sahu te al. [19]	Multi-CNN-Max	55.73	50.08	49.42
Sahu and Anand [34]	LSTM-ATT	65.23	56.77	60.04
Wang et al. [35]	RCNN	50.07	45.34	46.47
Raj et al. [20]	CRNN	67.91	61.98	64.38
*Luo et al. [21]	Seg-CNN	–	–	74.20
	Our model	75.69	73.03	74.34

*Luo et al. used the original dataset from the challenge (871 documents in total)