RESUMO
The free-form portions of clinical notes are a significant source of information for research, but before they can be used, they must be de-identified to protect patients' privacy. De-identification efforts have focused on known identifier types (names, ages, dates, addresses, ID's, etc.). However, a note can contain residual "Demographic Traits" (DTs), unique enough to re-identify the patient when combined with other such facts. Here we examine whether any residual risks remain after removing these identifiers. After manually annotating over 140,000 words worth of medical notes, we found no remaining directly identifying information, and a low prevalence of demographic traits, such as marital status or housing type. We developed an annotation guide to the discovered Demographic Traits (DTs) and used it to label MIMIC-III and i2b2-2006 clinical notes as test sets. We then designed a "bootstrapped" active learning iterative process for identifying DTs: we tentatively labeled as positive all sentences in the DT-rich note sections, used these to train a binary classifier, manually corrected acute errors, and retrained the classifier. This train-and-correct process may be iterated. Our active learning process significantly improved the classifier's accuracy. Moreover, our BERT-based model outperformed non-neural models when trained on both tentatively labeled data and manually relabeled examples. To facilitate future research and benchmarking, we also produced and made publicly available our human annotated DT-tagged datasets. We conclude that directly identifying information is virtually non-existent in the multiple medical note types we investigated. Demographic traits are present in medical notes, but can be detected with high accuracy using a cost-effective human-in-the-loop active learning process, and redacted if desired.2.
Assuntos
Aprendizado Profundo , Confidencialidade , Demografia , Humanos , Fenótipo , Aprendizagem Baseada em ProblemasRESUMO
BACKGROUND: Automated machine-learning systems are able to de-identify electronic medical records, including free-text clinical notes. Use of such systems would greatly boost the amount of data available to researchers, yet their deployment has been limited due to uncertainty about their performance when applied to new datasets. OBJECTIVE: We present practical options for clinical note de-identification, assessing performance of machine learning systems ranging from off-the-shelf to fully customized. METHODS: We implement a state-of-the-art machine learning de-identification system, training and testing on pairs of datasets that match the deployment scenarios. We use clinical notes from two i2b2 competition corpora, the Physionet Gold Standard corpus, and parts of the MIMIC-III dataset. RESULTS: Fully customized systems remove 97-99% of personally identifying information. Performance of off-the-shelf systems varies by dataset, with performance mostly above 90%. Providing a small labeled dataset or large unlabeled dataset allows for fine-tuning that improves performance over off-the-shelf systems. CONCLUSION: Health organizations should be aware of the levels of customization available when selecting a de-identification deployment solution, in order to choose the one that best matches their resources and target performance level.
Assuntos
Anonimização de Dados/normas , Registros Eletrônicos de Saúde , Aprendizado de Máquina/normas , Conjuntos de Dados como Assunto , HumanosRESUMO
In moth-pollinated petunias, production of floral volatiles initiates when the flower opens and occurs rhythmically during the day, for optimal flower-pollinator interaction. To characterize the developmental transcriptomic response to time of day, we generated RNA-Seq databases for corollas of floral buds and mature flowers in the morning and in the evening. Around 70% of transcripts accumulating in petals demonstrated significant changes in expression levels in response to the flowers' transition from a 4.5-cm bud to a flower 1 day postanthesis (1DPA). Overall, 44% of the petal transcripts were differentially expressed in the morning vs. evening. Morning/evening changes were affected by flower developmental stage, with a 2.5-fold larger transcriptomic response to daytime in 1DPA flowers compared to buds. Analyzed genes known to encode enzymes in volatile organic compound biosynthesis were upregulated in 1DPA flowers vs. buds-in parallel with the activation of scent production. Based on analysis of global changes in the petal transcriptome, PhWD2 was identified as a putative scent-related factor. PhWD2 is a protein that is uniquely present in plants and has a three-domain structure: RING-kinase-WD40. Suppression of PhWD2 (termed UPPER - Unique Plant PhEnylpropanoid Regulator) resulted in a significant increase in the levels of volatiles emitted from and accumulated in internal pools, suggesting that it is a negative regulator of petunia floral scent production.