Table 3 Remaining PHI analysis by tool, I2B2 corpus.

From: Protected Health Information filter (Philter): accurately and securely de-identifying free-text clinical notes

PHI category

Instances of PHI remaining (PHIlter)

Instances of PHI remaining (Physionet)

Instances of PHI remaining (Scrubber)

Age

0

1

0

Device

0

6

0

Medical record

0

524

18

Patient

2

154

92

Date

0

4590

1587

Fax

0

2

0

Phone

0

31

67

Zip

0

3

1

Username

1

92

92

Street

2

27

21

Location-other

2

9

12

Idnum

2

297

206

City

2

14

52

Doctor

5

197

186

  1. PHI counts for PHIlter, Physionet and Scrubber performance on the I2B2 corpus.