The OpenDeID corpus for patient de-identification.
Ontology highlight
ABSTRACT:
SUBMITTER: Jonnagaddala J
PROVIDER: S-EPMC8497517 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:PurposeTo evaluate publicly available de-identification tools on a large corpus of narrative-text radiology reports.Materials and methodsIn this retrospective study, 21 categories of protected health information (PHI) in 2503 radiology reports were annotated from a large multihospital academic health system, collected between January 1, 2012 and January 8, 2019. A subset consisting of 1023 reports served as a test set; the remainder were used as domain-specific training data. The types and frequencies of PHI present within the reports were tallied. Five public de-identification tools were evaluated: MITRE Identification Scrubber Toolkit, U.S. National Library of Medicine‒Scrubber, Massachusetts Institute of Technology de-identification software, Emory Health Information DE-identification (HIDE) software, and Neuro named-entity recognition (NeuroNER). The tools were compared using metrics including recall, precision, and F1 score (the harmonic mean of recall and precision) for each category of PHI.ResultsThe annotators identified 3528 spans of PHI text within the 2503 reports. Cohen κ for interrater agreement was 0.938. Dates accounted for the majority of PHI found in the dataset of radiology reports (n = 2755 [78%]). The two best-performing tools both used machine learning methods-NeuroNER (precision, 94.5%; recall, 92.6%; microaveraged F1 score [F1], 93.6%) and Emory HIDE (precision, 96.6%; recall, 88.2%; F1, 92.2%)-but none exceeded 50% F1 on the important patient names category.ConclusionPHI appeared infrequently within the corpus of reports studied, which created difficulties for training machine learning systems. Out-of-the-box de-identification tools achieved limited performance on the corpus of radiology reports, suggesting the need for further advancements in public datasets and trained models.Supplemental material is available for this article.See also the commentary by Tenenholtz and Wood in this issue.© RSNA, 2020.
| S-EPMC8082401 | biostudies-literature
Project description:ObjectivePatient notes in electronic health records (EHRs) may contain critical information for medical investigations. However, the vast majority of medical investigators can only access de-identified notes, in order to protect the confidentiality of patients. In the United States, the Health Insurance Portability and Accountability Act (HIPAA) defines 18 types of protected health information that needs to be removed to de-identify patient notes. Manual de-identification is impractical given the size of electronic health record databases, the limited number of researchers with access to non-de-identified notes, and the frequent mistakes of human annotators. A reliable automated de-identification system would consequently be of high value.Materials and methodsWe introduce the first de-identification system based on artificial neural networks (ANNs), which requires no handcrafted features or rules, unlike existing systems. We compare the performance of the system with state-of-the-art systems on two datasets: the i2b2 2014 de-identification challenge dataset, which is the largest publicly available de-identification dataset, and the MIMIC de-identification dataset, which we assembled and is twice as large as the i2b2 2014 dataset.ResultsOur ANN model outperforms the state-of-the-art systems. It yields an F1-score of 97.85 on the i2b2 2014 dataset, with a recall of 97.38 and a precision of 98.32, and an F1-score of 99.23 on the MIMIC de-identification dataset, with a recall of 99.25 and a precision of 99.21.ConclusionOur findings support the use of ANNs for de-identification of patient notes, as they show better performance than previously published systems while requiring no manual feature engineering.
| S-EPMC7787254 | biostudies-literature
Project description:BACKGROUND:In the cuprizone model of multiple sclerosis, de- and remyelination can be studied without major interference from the adaptive immune responses. Since previous proteomic studies did not focus on the corpus callosum, where cuprizone causes the most pronounced demyelination, we performed a bottom up proteomic analysis on this brain region. METHODS:Eight week-old mice treated with 0.2% cuprizone, for 4 weeks and controls (C) were sacrificed after termination of the treatment (4wD), and 2 (2dR) or 14 (2wR) days later. Homogenates of dissected corpus callosum were analysed by quantitative proteomics. For data processing, clustering, gene ontology analysis, and regulatory network prediction, we used Perseus, PANTHER and Ingenuity Pathway Analysis softwares, respectively. RESULTS:We identified 4886 unmodified, single- or multi phosphorylated and/or gycosylated (PTM) proteins. Out of them, 191 proteins were differentially regulated in at least one experimental group. We found 57 proteins specific for demyelination, 27 for early- and 57 for late remyelinationwhile 36 proteins were affected in two, and 23 proteins in all three groups. Phosphorylation represented 92% of the post translational modifications among differentially regulated modified (PTM) proteins with decreased level, while it was only 30% of the PTM proteins with increased level. Gene ontology analysis could not classify the demyelination specific proteins into any biological process category, while allocated the remyelination specific ones to nervous system development and myelination as the most specific subcategory. We also identified a protein network in experimental remyelination, and the gene orthologues of the network were differentially expressed in remyelinating multiple sclerosis brain lesions consistent with an early remyelination pattern. CONCLUSION:Proteomic analysis seems more informative for remyelination than demyelination in the cuprizone model.
| S-EPMC7145428 | biostudies-literature
Project description:BackgroundUK government guidelines and initiatives emphasise equity in delivery of care, shared decision-making, and patient-centred care. This includes sharing information with patients as partners in health decisions and empowering them to manage their health effectively. In the UK, general practitioners (GPs) routinely receive hospital discharge letters; while patients receiving copies of such letters is seen as "good practice" and recommended, it is not standardised. The effects and consequences of whether or not this happens remains unclear. The aim of this study (one of three forming the Discharge Communication Study) was to explore patient perspectives on receiving discharge letters and their views on how this could be improved in order to optimise patient experience and outcomes.MethodsSemi-structured interviews were conducted with a diverse sample of 50 patients recruited from 17 GP surgeries within the West Midlands, UK. All participants were adults with a recent episode of general hospital inpatient or outpatient care. Data were audio recorded, transcribed and analysed using mixed methods corpus linguistics techniques.ResultsParticipants reported inconsistent access to discharge letters. Most wanted to receive a copy of their discharge letter although some expressed reservations. Perceived benefits included: increased understanding of their condition and treatment, reduced anxiety, and increased satisfaction. Consequences where participants had not received letters included: letter inaccuracies being overlooked, missed follow up actions, failure to fully remember diagnosis, treatment, or self-management or recommendations, and confusion and anxiety at what occurred and what will happen next. Participants felt the usefulness of receiving copies of letters could be increased by: including a patient information section, avoidance of acronyms, and jargon or technical terms explained with lay language.ConclusionsMost patients value receiving copies of hospital discharge letters, and should be consistently offered them. Patients' preferences for letter receipt could be logged in their health records. To enable positive outcomes letters should have a clear and accessible format that reflects the priorities and information needs of patients. Patients appear not to be receiving or being offered copies of letters consistently despite UK policies and guidelines supporting this practice; this suggests a need for greater standardisation of practice.
| S-EPMC7294646 | biostudies-literature
Project description:We report an 8-year-old boy with a complex cerebral malformation, intellectual disability, and complex partial seizures. Whole-exome sequencing revealed a yet unreported de novo variant in the PIK3R2 gene that was recently associated with megalencephaly-polymicrogyria-polydactyly-hydrocephalus (MPPH) syndrome and bilateral perisylvian polymicrogyria (BPP). Our patient showed cerebral abnormalities (megalencephaly, perisylvian polymicrogyria, and mega corpus callosum) that were consistent with these conditions. Imaging also showed right temporal anomalies suggestive of cortical dysplasia. Until now, only three variants (c.1117G>A (p.(G373R)), c.1126A>G (p.(K376E)) and c.1202T>C (p.(L401P))) affecting the SH2 domain of the PIK3R2 protein have been reported in MPPH and BPP syndromes. In contrast to the variants reported so far, the patient described herein exhibits the c.1669G>C (p.(D557H)) variant that affects a highly conserved residue at the interface with the PI3K catalytic subunit α. The phenotypic spectrum associated with variants in this gene and its pathway are likely to continue to expand as more cases are identified.
| S-EPMC4989218 | biostudies-literature
Project description:BackgroundCornelia de Lange syndrome is characterized by dysmorphic facial features, hirsutism, severe growth and developmental delay. Germline mutations in the NIPBL gene with an autosomal dominant pattern and in the SMC1A gene with an X-linked pattern have been identified in Cornelia de Lange syndrome.Case presentationA two-month-old Iranian boy who showed multiple congenital anomalies was referred to the genetic center of a welfare organization in southwest Iran. He was the second child of a non-consanguineous marriage, born after full term with normal delivery. His birth weight was 3110 g, his length was 46 cm and his head circumference was 30 cm. Both parents were clinically asymptomatic, with no positive history of any deformity in their respective families.ConclusionsSequencing of the NIPBL gene from our patient revealed a single-base deletion of thymidine in exon 10 (c.516delT). This mutation presumably results in premature termination at codon 526. We did not observe this mutation in the parents of our patient with Cornelia de Lange syndrome. The results presented here enlarge the spectrum of NIPBL gene mutations associated with Cornelia de Lange syndrome by identifying a novel de novo mutation in an Iranian patient with Cornelia de Lange syndrome and further support the hypothesis that NIPBL mutations are disease-causing mutations leading to Cornelia de Lange syndrome.
| S-EPMC3138439 | biostudies-literature
Project description:BACKGROUND:More than half of patients with genetic leukoencephalopathies remain without a specific diagnosis; this is particularly true in individuals with a likely primary neuronal etiology, such as those in which abnormal white matter occurs in combination with severe epilepsy. PATIENT:A child with a severe early infantile epileptic encephalopathy and abnormal myelination underwent whole exome sequencing. RESULTS:Whole exome sequencing identified a heterozygous de novo mutation in KCNT1, a sodium-gated potassium channel gene. CONCLUSIONS:Severely delayed myelination was anecdotally reported in previous patients with KCNT1 mutations. This case reinforces that KCNT1 sequencing should be included in an investigation of patients with severely delayed myelination and epilepsy.
| S-EPMC4303471 | biostudies-literature
Project description:Barth syndrome (BTHS) is a rare genetic disorder characterized by various types of cardiomyopathy, neutropenia, failure to thrive, skeletal myopathy, and 3-methylglutaconic aciduria. BTHS is caused by loss-of-function mutations in the tafazzin (TAZ) gene located on chromosome Xq28, leading to cardiolipin deficiency. We report a 13-month-old boy with BTHS who had a novel de novo mutation in the TAZ gene. To the best of our knowledge, this is the first reported case of a BTHS patient with a de novo mutation in Korea. This report will contribute towards expanding the knowledge on the mutation spectrum of the TAZ gene in BTHS.
| S-EPMC4925393 | biostudies-literature
Project description:Objective: 1q44 microdeletion syndrome is difficult to diagnose due to the wide phenotypic spectrum and strong genetic heterogeneity. We explore the correlation between the chromosome microdeletions and phenotype in a child with 1q44 microdeletion syndrome, we collected the clinical features of the patient and combined them with adjacent copy number variation (CNV) regions previously reported. Methods: We collected the full medical history of the patient and summarized her clinical symptoms. Whole-exome sequencing (WES) and CapCNV analysis were performed with DNA extracted from both the patient's and her parents' peripheral blood samples. Fluorescent quantitative PCR (q-PCR) was performed for the use of verification to the CNV regions. Results: A 28.7 KB microdeletion was detected in the 1q44 region by whole-exome sequencing and low-depth whole-genome sequencing. The deleted region included the genes COX20 and HNRNPU. As verification, karyotype analysis showed no abnormality, and the results of qPCR were consistent with that of whole-exome sequencing and CapCNV analysis. Conclusion: The patient was diagnosed with 1q44 microdeletion syndrome with clinical and genetic analysis. Analyzing both whole-exome sequencing and CapCNV analysis can not only improve the diagnostic rate of clinically suspected syndromes that present with intellectual disability (ID) and multiple malformations but also support further study of the correlation between CNVs and clinical phenotypes. This study lays the foundation for the further study of the pathogenesis of complex diseases.
| S-EPMC8173053 | biostudies-literature
Project description:Bartonella henselae is associated with a wide spectrum of clinical manifestations, including cat scratch disease, endocarditis and meningoencephalitis, in immunocompetent and immunocompromised patients. We report the first molecularly confirmed case of B. henselae infection in an AIDS patient in state of Rio de Janeiro, Brazil. Although DNA sequence of B. henselae has been detected by polymerase chain reaction in a lymph node biopsy, acute and convalescent sera were nonreactive.
| S-EPMC4131826 | biostudies-literature