Automatic de-identification of electronic medical records using token-level and character-level conditional random fields

Abstract

De-identification, identifying and removing all protected health information (PHI) present in clinical data including electronic medical records (EMRs), is a critical step in making clinical data publicly available. The 2014 i2b2 (Center of Informatics for Integrating Biology and Bedside) clinical natural language processing (NLP) challenge sets up a track for de-identification (track 1). In this study, we propose a hybrid system based on both machine learning and rule approaches for the de-identification track. In our system, PHI instances are first identified by two (token-level and character-level) conditional random fields (CRFs) and a rule-based classifier, and then are merged by some rules. Experiments conducted on the i2b2 corpus show that our system submitted for the challenge achieves the highest micro F-scores of 94.64%, 91.24% and 91.63% under the “token”, “strict” and “relaxed” criteria respectively, which is among top-ranked systems of the 2014 i2b2 challenge. After integrating some refined localization dictionaries, our system is further improved with F-scores of 94.83%, 91.57% and 91.95% under the “token”, “strict” and “relaxed” criteria respectively.

De-identification of clinical notes via recurrent neural network and conditional random field

“…The representative works are three natural language processing (NLP) challenges, two organized by the Center of Informatics for Integrating Biology and Bedside (i2b2) in 2006 [2] and 2014 [3, 4, 5], and one organized by the Centers of Excellence in Genomic Science (CEGS) Neuropsychiatric Genome-scale and RDOC Individualized Domains (N-GRID) in 2016 [6]. The organizers of the three challenges provide manually annotated corpora for participants to develop various kinds of systems for de-identification [7, 8, 9, 10, 11, 12, 13, 14, 15]. …”

“…In our system, an ensemble classifier is deployed to combine the outputs of three individual machine learning-based subsystems, and a rule-based subsystem is used to identify some formulaic PHI instances. The three machine learning-based subsystems are a CRF-based system with a large number of hand-crafted features [12], a bidirectional LSTM-based system without any hand-crafted features [16, 17], and a variant of bidirectional LSTM-based system with a small quantity of hand-crafted features [18, 19]. Moreover, we also evaluate our system on the 2014 i2b2 challenge corpus and compare it with other state-of-the-art systems.…”

“…Lots of teams from all around the world participated in this three challenges. In the two i2b2 NLP challenges, the proposed de-identification systems may fall in three categories: rule-based [26], machine learning-based [10, 14, 27], and hybrid [11, 12, 13, 15]. The rule-based systems can exactly recognize formulaic PHI instances (i.e., phone numbers, emails, licenses, etc.…”

The UAB Informatics Institute and 2016 CEGS N-GRID de-identification shared task challenge

“…These rules increase the system coverage by permitting the uses of more relaxed patterns and ambiguous terms. Researchers can combine the pattern and dictionary methods in the machine-learning model by using matching results as features (11, 19). This method is simple and likely optimal within the scope of the challenge.…”

“…Successful systems used the machine-learning algorithm Conditional Random Field (CRF) for labeling a sequence of tokens(11, 18, 19). The popular rule-based method was pattern-matching using a formal language such as regular expressions.…”