-
Notifications
You must be signed in to change notification settings - Fork 17
/
index.shtml
611 lines (485 loc) · 22.3 KB
/
index.shtml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
<html>
<head>
<title>De-identification: software and test data</title>
</head>
<body bgcolor=white>
<!--#include virtual="/links-physiotools.html"-->
<h1>De-identification: software and test data</h1>
<p>
<hr>
<em>
The <a href="#software"><tt>deid</tt> software package</a> includes
code and dictionaries for automated location and removal of protected
health information (PHI) in free text from medical records.
The <a href="#data">gold-standard corpus</a> is a collection of
deidentified nursing notes that have been reidentified with realistic
surrogate data, for use in developing and evaluating software such
as <tt>deid</tt>.</em>
<hr>
<p>
<table cellpadding=20><tr><td width="40%">
<center><h2>On this page</h2></center>
<ul>
<li><a href="#background">Background</a>
<li><a href="#software">Software</a>
<ul>
<li><a href="#prereq">Prerequisite</a>
<li><a href="#downloading">Downloading and installing <tt>deid</tt></a>
<li><a href="#testing">Testing <tt>deid</tt></a>
<li><a href="#customizing">Customizing <tt>deid</tt></a>
<li><a href="#contents">Contents of the <tt>deid</tt> package</a>
<ul>
<li><a href="#code">Code</a>
<li><a href="#dictionaries">Dictionaries</a>
<li><a href="#gsfiles">Files related to the gold-standard corpus</a>
<li><a href="#sample-outputs">Sample outputs</a>
<li><a href="#docs">Documentation</a>
</ul>
<li><a href="#notes">Notes on files in the package</a>
</ul>
<li><a href="#data">Data: the gold-standard corpus</a>
<li><a href="#refs">Further reading</a>
</ul>
</td><td bgcolor="lightblue">
For additional discussion of the methods used by the software
described here, see
<blockquote>
Neamatullah I, Douglass M, Lehman LH, Reisner A, Villarroel M, Long WJ,
Szolovits P, Moody GB, Mark RG, Clifford GD.
<a href="http://www.biomedcentral.com/1472-6947/8/32"
target="other">Automated De-Identification of Free-Text Medical
Records</a>. <em>BMC Medical Informatics and Decision Making</em>,
2008, <b>8</b>:32. doi:10.1186/1472-6947-8-32
</blockquote>
<p>
Please cite the above paper when referencing this material, and
also include the standard citation for PhysioNet:</b>
<blockquote>
Goldberger AL, Amaral LAN, Glass L, Hausdorff JM, Ivanov PCh, Mark RG,
Mietus JE, Moody GB, Peng C-K, Stanley HE. PhysioBank, PhysioToolkit, and
Physionet: Components of a New Research Resource for Complex Physiologic
Signals.
<i>Circulation</i> <b>101</b>(23):e215-e220 [Circulation Electronic Pages;
<a href="http://circ.ahajournals.org/cgi/content/full/101/23/e215"
target="other">http://circ.ahajournals.org/cgi/content/full/101/23/e215</a>];
2000 (June 13).
</blockquote>
</td></tr>
</table>
<a name=background><h2>Background</h2></a>
<p>
In the USA, the HIPAA Privacy Rule restricts exchange of medical data
containing protected health information (PHI), defined as any
information that might be used to identify the individual(s) from whom
the data were collected. Data known to contain PHI can be shared for
research purposes only under tightly controlled circumstances, typically
involving data use agreements under which the researchers involved must
obtain IRB or equivalent approvals for use of the data.
<p>
By contrast, medical data that do not contain PHI are exempt from the
restrictions of the HIPAA Privacy Rule and may be shared freely. The
data available on PhysioNet fall into this category.
<p>
Many of the research data sets currently being collected by PhysioNet are
accompanied by PHI, and the process of removing this PHI ("de-identification"
in the language of HIPAA, or "anonymization") is tedious and error-prone.
For many research projects, the cost of de-identification is a significant
barrier to data sharing.
<p>
The <a href="/physiobank/database/mimic2cdb/">MIMIC II</a> project has
invested several years' effort to develop and evaluate software that
is capable of removing PHI from text and, where appropriate, replacing
it by realistic surrogate PHI. (For example, names are replaced by
fictitious names, medical record numbers by fictitious medical record
numbers, dates by fictitious dates, geographic locations by other
geographic locations, etc.) The MIMIC II project has contributed to
PhysioNet both its deidentification software, which may be useful to
other researchers needing to remove PHI from their own data, and a
large corpus of text drawn from ICU nursing notes that were gathered
simultaneously with the signals, trends, laboratory reports, discharge
summaries and other data in the MIMIC
II <a href="/physiobank/database/mimic2cdb/">clinical</a>
and <a href="/physiobank/database/mimic2db/">waveform</a> databases.
<hr>
<a name=software><h2>Software</h2></a>
<p>
As is the case with all software available on PhysioNet, the
de-identification software (<tt>deid</tt>) is provided in source form so that
its workings can be studied, customized, and improved. This software
is free software. You can redistribute it and/or modify it under the
terms of the <a href="COPYING">GNU General Public License</a> as
published by the Free Software Foundation, either version 2 of the
License, or (at your option) any later version.
<p>
The authors of the <tt>deid</tt> software package are Margaret M Douglass,
Li-wei H Lehman, William J Long, and Ishna Neamatullah. An experimental
prototype by Jason M Levine provided useful guidance in the design of
<tt>deid</tt>.
<a name=prereq><h3>Prerequisite</h3></a>
<p>
The <tt>deid</tt> package requires Perl (version 5 or later), which is
freely available for all popular platforms
<a href="http://perl.org/" target=other>here</a> if you don't have it
already.
<p>
The current version of the <tt>deid</tt> software package was
developed and tested using perl 5.8.8 and perl 5.10 on GNU/Linux
(Fedora Core 10). It has also been tested on Fedora 6, 7 and 8,
Ubuntu 8.04, MacOS X 10.4.5, and on MS-Windows XP.
<a name=downloading><h3>Downloading and installing <tt>deid</tt></h3></a>
<p>
The <tt>deid</tt> software package can be downloaded as a single
file, <a
href="/physiotools/sources/deid/deid-1.1.tar.gz"><tt>deid-1.1.tar.gz</tt></a>.
(Instructions on unpacking <tt>tar.gz</tt> files
are <a href="/faq.shtml#tar-gz">here</a>.) The contents
of <tt>deid-1.1.tar.gz</tt> unpack into a directory named
<tt>deid-1.1</tt>, which contains the files
listed <a href=#contents>below</a>.
<p>
The <tt>deid</tt> software package is command-line
(<a href=/faq.shtml#text-mode>text mode</a>) software that must be run
in a terminal emulator (console) window.
<p>
Previous versions of the <tt>deid</tt> package are available for
reference purposes <a href=/physiotools/sources/deid/>here</a>.
<a name=testing><h3>Testing <tt>deid</tt></h3></a>
<p>
If you wish to test <tt>deid</tt>, see <a href="#data">Data: the gold-standard
corpus of deidentified medical text</a> below for information on obtaining
<tt>id.text</tt>. Copy <tt>id.text</tt> into the <tt>deid-1.1</tt>
directory (created when <tt>deid-1.1.tar.gz</tt> was unpacked in the
previous step), and enter that directory. Type the command
<pre>
perl deid.pl id deid.config
</pre>
<p>
The software will print a greeting message:
<pre>
*******************************************************************************************************************
De-Identification Algorithm: Identifies Protected Health Information (PHI) in Discharge Summaries and Nursing Notes
*******************************************************************************************************************
Starting de-identification (version 1.1) ...
Running deid in output mode. Output files will be:
id.phi: the PHI locations found by the code.
id.res: the scrubbed text.
id.info: debug info about the PHI locations.
</pre>
<p>
On a typical current PC, the process will run to completion in
approximately 10 to 20 minutes. The three output files will be
written into the current directory; they should be identical to
the files of the same names that are included in the <tt>GSoutput</tt>
directory.
<p>
The program <tt>runStat.pl</tt> can be used to derive performance
statistics by comparing the <tt>id.phi</tt> file generated
by <tt>deid.pl</tt> with the reference <tt>id.deid</tt> file
provided with the <tt>deid</tt> software package. To do this, type
the command
<pre>
perl runStat.pl id.deid id.phi
</pre>
which should produce the output
<pre>
==========================
Num of true positives = 1720
Num of false positives = 546
Num of false negatives = 59
Sensitivity/Recall = 0.967
PPV/Specificity = 0.748
==========================
</pre>
To use the gold standard corpus to test the performance of another
deidentifier, arrange for that deidentifier to produce a file in the
format of <tt>id.phi</tt>, then use <tt>runStat.pl</tt> as above.
<a name=customizing><h3>Customizing <tt>deid</tt></h3></a>
<p>
Examine the configuration file, <tt>deid.config</tt>, to see how
to modify the behavior of <tt>deid.pl</tt>. For additional information,
see the <a href="doc/DeidUserManual.pdf">user manual</a>.
<p>
In order to customize this software to de-identify free text in other
medical records, you may replace our filter modules with your
data-specific filters. Additionally, at a minimum, you will have to
replace the six files that contain a priori information
(see <a href=#apriori>below</a>).
<p>
Depending on the text you wish to de-identify, you may wish to re-classify names
as ambiguous or not. For example, "Mae" is an unambiguous name in most
contexts, but in nursing and discharge notes, it may be an acronym (meaning
"moving all extremities") and it is therefore ambiguous in those contexts.
<hr>
<a name=contents><h3>Contents of the <tt>deid</tt> package</h3></a>
See <a href=#downloading>above</a> for information about downloading
the entire <tt>deid</tt> package in a single archive file. The
individual files in the package can also be viewed or downloaded by
following the links below.
<dl>
<dt><a name=code><b>Code</b></a></dt>
<dl>
<dt><tt><a href=deid.pl>deid.pl</a></tt></dt>
<dd>perl program for de-identifying medical notes</dd>
<dt><tt><a href=runStat.pl>runStat.pl</a></tt></dt>
<dd>perl program for calculating performance statistics</dd>
<dt><tt><a href=stat.pm>stat.pm</a></tt></dt>
<dd>perl subroutines shared by deid.pl and runStat.pl</dd>
<dt><tt><a href=deid.config>deid.config</a></tt></dt>
<dd>sample configuration file for use with deid.pl</dd>
<br>
<dt><tt><a href=COPYING>COPYING</a></tt></dt>
<dd>GNU General Public License, version 2</dd>
<br>
</dl>
<dt><a name=dictionaries><b>Dictionaries</b></a> (see the <a href=doc/DeidUserManual.pdf>User
Manual</a> for details)</dt>
<dl>
<dt><tt>dict/</tt></dt>
<dd>Subdirectory containing lists of words and phrases not likely to be PHI</dd>
<dl>
<dt><tt><a href=dict/common_words.txt>common_words.txt</a></tt></dt>
<dd>49,668 words that are common in medical records</dd>
<dt><tt><a href=dict/commonest_words.txt>commonest_words.txt</a></tt></dt>
<dd>5,126 words that are very common in medical records</dd>
<dt><tt><a href=dict/medical_phrases.txt>medical_phrases.txt</a></tt></dt>
<dd>28 medical phrases</dd>
<dt><tt><a href=dict/notes_common.txt>notes_common.txt</a></tt></dt>
<dd>66 very common words found in nursing notes</dd>
<dt><tt><a href=dict/sno_edited.txt>sno_edited.txt</a></tt></dt>
<dd>175,313 medical terms from UMLS/SNOMED</dd>
</dl>
<br>
<dt><tt>lists/</tt></dt>
<dd>Subdirectory containing lists of words and phrases that are or might be PHI</dd>
<dl>
<br>
<dt><em>Names</em></dt>
<dt><tt><a href=lists/last_names_unambig.txt>last_names_unambig.txt</a></tt></dt>
<dd>81,497 unambiguous family names</dd>
<dt><tt><a href=lists/last_names_ambig.txt>last_names_ambig.txt</a></tt></dt>
<dd>7,298 ambiguous family names</dd>
<dt><tt><a href=lists/last_names_popular.txt>last_names_popular.txt</a></tt></dt>
<dd>93 popular family names</dd>
<dt><tt><a href=lists/prefixes_unambig.txt>prefixes_unambig.txt</a></tt></dt>
<dd>17 family name prefixes (von, de la, etc.)</dd>
<dt><tt><a href=lists/last_name_prefixes.txt>last_name_prefixes.txt</a></tt></dt>
<dd>138 prefixes that may appear before a family name</dd>
<dt><tt><a href=lists/female_names_unambig.txt>female_names_unambig.txt</a></tt></dt>
<dd>3843 unambiguous female given names</dd>
<dt><tt><a href=lists/female_names_ambig.txt>female_names_ambig.txt</a></tt></dt>
<dd>616 ambiguous female given names</dd>
<dt><tt><a href=lists/female_names_popular.txt>female_names_popular.txt</a></tt></dt>
<dd>125 popular female given names</dd>
<dt><tt><a href=lists/male_names_unambig.txt>male_names_unambig.txt</a></tt></dt>
<dd>1144 unambiguous male given names</dd>
<dt><tt><a href=lists/male_names_ambig.txt>male_names_ambig.txt</a></tt></dt>
<dd>419 ambiguous male given names</dd>
<dt><tt><a href=lists/male_names_popular.txt>male_names_popular.txt</a></tt></dt>
<dd>130 popular male given names</dd>
<br>
<dt><em>Locations</em></dt>
<dt><tt><a href=lists/countries_unambig.txt>countries_unambig.txt</a></tt></dt>
<dd>179 country names</dd>
<dt><tt><a href=lists/us_states.txt>us_states.txt</a></tt></dt>
<dd>59 US states and territories</dd>
<dt><tt><a href=lists/us_states_abbre.txt>us_states_abbre.txt</a></tt></dt>
<dd>59 standard US state and territorial abbreviations</dd>
<dt><tt><a href=lists/more_us_state_abbreviations.txt>more_us_state_abbreviations.txt</a></tt></dt>
<dd>53 non-standard US state name abbreviations</dd>
<dt><tt><a href=lists/locations_unambig.txt>locations_unambig.txt</a></tt></dt>
<dd>3341 unambiguous location names</dd>
<dt><tt><a href=lists/locations_ambig.txt>locations_ambig.txt</a></tt></dt>
<dd>135 words that may be (parts of) location names</dd>
<br>
<dt><em>Other possible PHI</em></dt>
<dt><tt><a href=lists/us_area_code.txt>us_area_code.txt</a></tt></dt>
<dd>382 US telephone area codes</dd>
<dt><tt><a href=lists/company_names_unambig.txt>company_names_unambig.txt</a></tt></dt>
<dd>484 unambiguous company names</dd>
<dt><tt><a href=lists/company_names_ambig.txt>company_names_ambig.txt</a></tt></dt>
<dd>18 ambiguous company names</dd>
<dt><tt><a href=lists/ethnicities_unambig.txt>ethnicities_unambig.txt</a></tt></dt>
<dd>195 ethnicities</dd>
<br>
<dt><a name=apriori><em>Known PHI (a priori information)</em></a></dt>
<dd>[The six files in this group contain the surrogate PHI from
the gold standard corpus, not the original PHI from the files used to
create it! You will need to create your own versions of these six
files, using these as models, if you wish to use <tt>deid</tt> on your
own data.]</dd>
<dt><tt><a href=lists/pid_patientname.txt>pid_patientname.txt</a></tt></dt>
<dd>163 full names and ids of the patients in the gold standard corpus</dd>
<dt><tt><a href=lists/doctor_first_names.txt>doctor_first_names.txt</a></tt></dt>
<dd>56 given names of doctors</dd>
<dt><tt><a href=lists/doctor_last_names.txt>doctor_last_names.txt</a></tt></dt>
<dd>254 family names of doctors</dd>
<dt><tt><a href=lists/stripped_hospitals.txt>stripped_hospitals.txt</a></tt></dt>
<dd>143 names of nearby hospitals</dd>
<dt><tt><a href=lists/local_places_unambig.txt>local_places_unambig.txt</a></tt></dt>
<dd>48 unambiguous names of nearby towns and cities</dd>
<dt><tt><a href=lists/local_places_ambig.txt>local_places_ambig.txt</a></tt></dt>
<dd>4 ambiguous names of nearby towns and cities</dd>
</dl>
</dl>
<br>
<dt><a name=gsfiles><b>Files related to the gold-standard corpus (excluding a priori
information, see above)</b></a><dt>
<dl>
<dt><tt><a href=id.deid>id.deid</a></tt></dt>
<dd>PHI locations in the gold standard corpus</dd>
<dt><tt><a href=id-phi.phrase>id-phi.phrase</a></tt></dt>
<dd>PHI locations, types, and PHI terms in the gold standard corpus</dd>
<dt><tt><a href=shift.txt>shift.txt</a></tt></dt>
<dd>Date shifts for patients in the gold standard corpus</dd>
<br>
<dt><em>[The gold standard corpus itself must be obtained separately
(see below); it is not part of this package.]</em></dt>
</dl>
<br>
<dt><a name=sample-outputs><b>Sample outputs generated by <tt>deid.pl</tt></b></a></dt>
<dl>
<dt><tt>GSoutput/</tt></dt>
<dd>[The three files provided in this directory were generated
from the gold standard corpus, using the test procedure described
below running on Fedora 10.]</dd>
<dl>
<dt><a href=GSoutput/id.phi><tt>id.phi</tt></a></dt>
<dd>locations of detected PHI in the input text</dd>
<dt><a href=GSoutput/id.res><tt>id.res</tt></a></dt>
<dd>the deidentified output text, with PHI replaced by tags</dd>
<dt><a href=GSoutput/id.info><tt>id.info</tt></a></dt>
<dd>debugging information</dd>
</dl>
</dl>
<br>
<dt><a name=docs><b>Documentation</b></a></dt>
<dl>
<dt><tt>doc/</tt></dt>
<dl>
<dt><tt><a href=doc/DeidUserManual.pdf>DeidUserManual.pdf</a></tt></dt>
<dd>De-Identification of Free-Text Medical Records: User Manual</dd>
<dt><tt><a href=doc/Changes.log>Changes.log</a></tt></dt>
<dd>Changes made since the initial release (<tt>deid-1.0</tt>).</dd>
<br>
</dl>
<dt><em>[Also see the references at the top and bottom of this page.]</em></dt>
</dl>
</dl>
<a name=notes><h3>Notes on the files in the package</h3></a>
<p>
Except for functions related to calculating performance statistics (in
<tt>stats.pm</tt>), the deidentification code is contained in a perl
script (<tt>deid.pl</tt>). Its configuration file, <tt>deid.config</tt>,
can be used to set parameters of the deidentification process at run time.
Associated word and phrase lists are in <tt>lists/</tt> (containing known
and possible PHI) and <tt>dict/</tt> (containing probable non-PHI).
<p>
If the date shift filter is on, <tt>deid</tt> replaces all dates in
its input with surrogate dates. The <tt>shift.txt</tt> file contains
a randomly assigned date shift (between 1000 and 3000 days) for each
patient in the gold standard corpus; surrogate dates are generated by
adding the specified number of days to the dates in the input file.
The method used to generate the date shifts in <tt>shift.txt</tt>
differs from that used in PhysioBank databases.
<p>
The file <tt>id-phi.phrase</tt> is provided as a convenient index
to the PHI in the gold standard corpus; it is not used by the deid
software. Each line of this file contains 6 fields: PID,
Record_Number, PHI_Start_Location, PHI_End_Location, PHI_Type, and
PHI_Text.
<p>
The file <tt>id.deid</tt> also contains PHI locations in the gold standard
corpus, and it is used as a reference for calculating performance statistics
by the functions in <tt>stats.pm</tt>. It contains two types of lines. The
first type is of the form
<pre>
Patient <em>PID</em> Note <em>Record_Number</em>
</pre>
and the second type is of the form
<pre>
<em>PHI_Start_Location PHI_Start_Location PHI_End_Location</em>
</pre>
<hr>
<a name=data><h2>Data: the gold-standard corpus of deidentified medical text</h2></a>
<p>
Evaluating software for de-identification turns out to be quite difficult.
As in many of the projects contributing to PhysioNet, a reference database
is highly useful. Ideally the developer of de-identification software needs
an appropriate corpus of text in which all of the PHI has been labelled, so
that the software's PHI detection performance can be assessed quickly and
quantitatively. To compare different approaches to PHI removal objectively,
we need a standard corpus. And here's the dilemma: we can't share such a
corpus if it has any PHI in it!
<p>
The <tt>deid</tt> software package was developed and tested using a
gold standard corpus of 2,434 nursing notes that have been thoroughly
deidentified by a multi-pass process that included meticulous reviews
by three or more experts working independently, as well as by a
variety of automated methods. All detected instances of PHI in these
nursing notes have been replaced by realistic surrogate data in the
gold standard corpus. Although the <tt>deid</tt> software, as noted
above, may be redistributed under the terms of the GPL, the gold
standard corpus, because of the very small possibility that it may
contain one or more instances of undetected PHI, is currently
available only under terms that do not permit it to be redistributed.
This corpus is available only under the following conditions:
<ol>
<li> You agree not to attempt to identify any of the individual
subjects of the file.
<li> If you find information within the file that you believe might
permit identification of any of the individual subjects of the file,
you agree to report this promptly by email
to <a href="mailto:[email protected]">[email protected]</a>,
citing the specific information in question so that it can be
investigated and removed if necessary.
<li> You agree to exercise all reasonable and prudent care to avoid
disclosure of the identity of any of the individual subjects of the
file in any publication or other communication of your analysis of the
file.
<li> If you agree to all of these terms and conditions, access to this
file will be granted to you as an individual. You may not share it
with anyone else. Your colleagues can obtain access to this file as
individuals via the same procedure you are following.
</ol>
<p>
To download the corpus, login to PhysioNetWorks, then follow the "Deidentified
Medical Text" link on your PhysioNetWorks home page. (You can create a free
PhysioNetWorks account in a few minutes <a href="/pnw/login">here</a> if you
don't have one already.)
<hr>
<a name=refs><h2>Further reading</h2></a>
<ol>
<li>
Neamatullah I. <a href="doc/ishna-meng-thesis.pdf">Automated De-Identification
of Free-Text Medical Records</a>. MIT Dept of EECS, MEng thesis, 2006.
<li>
Douglass M. <a href="doc/douglass-meng-thesis.pdf">Computer-Assisted
De-identification of Free-text Nursing Notes</a>.
MIT Dept of EECS, MEng thesis, 2005.
<li>
Douglass M, Clifford GD, Reisner A, Long WJ, Moody GB, Mark RG.
<a href="doc/douglass-cinc-2005.pdf">De-Identification Algorithm for
Free-Text Nursing Notes</a>. <em>Computers In Cardiology</em> 32:331-334,
2005.
<li>
Douglass M, Clifford GD, Reisner A, Moody GB, Mark RG.
<a href="doc/douglass-cinc-2004.pdf">Computer-Assisted Deidentification of
Free Text in the MIMIC II Database</a>. <em>Computers In Cardiology</em>
31:341-344, 2004.
<li>
Levine JM. <a href="doc/levine-meng-thesis.pdf">De-identification of ICU patient
records</a>. MIT Dept of EECS, MEng thesis, 2003.
</ol>
<h3>Acknowledgements</h3>
<p>
This work was performed with support from the National Institute of Biomedical
Imaging and Bioengineering (NIBIB), under grant R01 EB001659.
<h3>Contact information</h3>
<blockquote>
Li-wei H Lehman (lilehman AT alum DOT mit DOT edu)<br>
Gari D Clifford (gari AT mit DOT edu)
</blockquote>
<hr>
<!--#include virtual="/footer.shtml"-->
</body>
</html>