DEDUCE: A pattern matching method for automatic de-identification of Dutch medical text

Menger, Vincent Jorn; Scheepers, Floortje; Wijk, Lisette Maria van; Spruit, Marco

doi:10.1016/j.tele.2017.08.002

Cited by 57 publications

(54 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We used deidentified data sets by deidentifying clinical notes using the Deindentification Method for Dutch Medical Text (DEDUCE) method. 19 Demographic variables were limited to sex, year of birth, and Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition) diagnosis. The study was reviewed and approved by the University Medical Center Utrecht ethical committee.…”

Section: Methodsmentioning

confidence: 99%

Machine Learning Approach to Inpatient Violence Risk Assessment Using Routinely Collected Clinical Notes in Electronic Health Records

et al. 2019

Self Cite

View full text Add to dashboard Cite

Key Points Question To what extent can inpatient violence risk assessment be performed by applying machine learning techniques to clinical notes in patients’ electronic health records? Findings In this prognostic study, machine learning was used to analyze clinical notes recorded in electronic health records of 2 independent psychiatric health care institutions in the Netherlands to predict inpatient violence. Internal predictive validity was measured using areas under the curve, which were 0.797 for site 1 and 0.764 for site 2; however, applying pretrained models to data from other sites resulted in significantly lower areas under the curve. Meaning The findings suggest that inpatient violence risk assessment can be performed automatically using already available clinical notes without sacrificing predictive validity compared with existing violence risk assessment methods.

show abstract

Section: Methodsmentioning

confidence: 99%

Machine Learning Approach to Inpatient Violence Risk Assessment Using Routinely Collected Clinical Notes in Electronic Health Records

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…The complete corpus of doctor and nurse notes (i.e., all notes written before, during or after admission) in the same time period was also made available, totaling 1,015,931 doctor and nurse notes combined. All notes are de-identified using the De-identification Method for Dutch Medical Text (DEDUCE) [54] before any other processing took place. The subset of notes that was available at the start of admission served as input for the prediction problem, while the entire corpus of notes were used to learn representations of text.…”

Section: Text Datasetmentioning

confidence: 99%

Comparing Deep Learning and Classical Machine Learning Approaches for Predicting Inpatient Violence Incidents from Clinical Text

2018

Self Cite

View full text Add to dashboard Cite

Machine learning techniques are increasingly being applied to clinical text that is already captured in the Electronic Health Record for the sake of delivering quality care. Applications for example include predicting patient outcomes, assessing risks, or performing diagnosis. In the past, good results have been obtained using classical techniques, such as bag-of-words features, in combination with statistical models. Recently however Deep Learning techniques, such as Word Embeddings and Recurrent Neural Networks, have shown to possibly have even greater potential. In this work, we apply several Deep Learning and classical machine learning techniques to the task of predicting violence incidents during psychiatric admission using clinical text that is already registered at the start of admission. For this purpose, we use a novel and previously unexplored dataset from the Psychiatry Department of the University Medical Center Utrecht in The Netherlands. Results show that predicting violence incidents with state-of-the-art performance is possible, and that using Deep Learning techniques provides a relatively small but consistent improvement in performance. We finally discuss the potential implication of our findings for the psychiatric practice.

show abstract

“…To evaluate performance between English and Dutch datasets, the nursing notes corpus dataset [1] (2,434 records, about 1,800 PHI instances) and the 2014 i2b2 dataset were also used. DEDUCE, a rule-based approach developed for Dutch medical records, was adopted by the researchers [65]. For the CRF approach, a subset of features from a token-based approach by Liu et al [37] was utilized.…”

Section: Trienes Et Al 2020 [47] (Comparing Rule-based Feature-basementioning

confidence: 99%

Survey on RNN and CRF models for de-identification of medical free text

Leevy

Khoshgoftaar

Villanustre³

2020

J Big Data

View full text Add to dashboard Cite

As the use and volume of medical records continues to rapidly grow in various areas, including research, there is a growing need to safeguard patient privacy for ethical and legal reasons [1]. In the USA, the confidentiality of patient information is legislated by the Health Insurance Portability and Accountability Act (HIPAA) [2]. The act lists 18 categories of protected health information (PHI), such as telephone numbers, geographic data, social security numbers, email addresses, and full face photos [3], that require special attention (see Table 1). PHI is health information capable of being linked, through the operations of a HIPAA-covered entity or business associate of the entity, to an individual patient. In the HIPAA world, the de-identification of PHI involves the reduction of risk to an acceptable level not subject to predefined privacy restrictions [4]. This process is carried out through the Expert Determination Method or the Safe Harbor method [5]. The Expert Determination method requires the opinion of a qualified statistician to

show abstract

DEDUCE: A pattern matching method for automatic de-identification of Dutch medical text

Cited by 57 publications

References 22 publications

Machine Learning Approach to Inpatient Violence Risk Assessment Using Routinely Collected Clinical Notes in Electronic Health Records

Machine Learning Approach to Inpatient Violence Risk Assessment Using Routinely Collected Clinical Notes in Electronic Health Records

Comparing Deep Learning and Classical Machine Learning Approaches for Predicting Inpatient Violence Incidents from Clinical Text

Survey on RNN and CRF models for de-identification of medical free text

Contact Info

Product

Resources

About