Table 6.

Evaluation on authentic discharge summaries. Highest F-measures are in bold. The F-measure differences from Stat De-id in PHI and in non-PHI are significant at α = 0.05.

Method	Class	Precision	Recall	F-measure
Stat De-id	PHI	98.46%	95.24%	96.82%
IFinder	PHI	26.17%	61.98%	36.80%*
H+D	PHI	82.67%	87.30%	84.92%*
CRFD	PHI	91.16%	84.75%	87.83%*
Stat De-id	Non-PHI	99.84%	99.95%	99.90%
IFinder	Non-PHI	98.68%	94.19%	96.38%*
H+D	Non-PHI	99.58%	99.39%	99.48%*
CRFD	Non-PHI	99.62%	99.86%	99.74%*