Gold Standard Corpus Research Articles

Extracting social support and social isolation information from clinical psychiatry notes: comparing a rule-based natural language processing system and a large language model.

Social support (SS) and social isolation (SI) are social determinants of health (SDOH) associated with psychiatric outcomes. In electronic health records (EHRs), individual-level SS/SI is typically documented in narrative clinical notes rather than as structured coded data. Natural language processing (NLP) algorithms can automate the otherwise labor-intensive process of extraction of such information. Psychiatric encounter notes from Mount Sinai Health System (MSHS, n = 300) and Weill Cornell Medicine (WCM, n = 225) were annotated to create a gold-standard corpus. A rule-based system (RBS) involving lexicons and a large language model (LLM) using FLAN-T5-XL were developed to identify mentions of SS and SI and their subcategories (eg, social network, instrumental support, and loneliness). For extracting SS/SI, the RBS obtained higher macroaveraged F1-scores than the LLM at both MSHS (0.89 versus 0.65) and WCM (0.85 versus 0.82). For extracting the subcategories, the RBS also outperformed the LLM at both MSHS (0.90 versus 0.62) and WCM (0.82 versus 0.81). Unexpectedly, the RBS outperformed the LLMs across all metrics. An intensive review demonstrates that this finding is due to the divergent approach taken by the RBS and LLM. The RBS was designed and refined to follow the same specific rules as the gold-standard annotations. Conversely, the LLM was more inclusive with categorization and conformed to common English-language understanding. Both approaches offer advantages, although additional replication studies are warranted.

Natural Language Processing Accurately Differentiates Cancer Symptom Information in Electronic Health Record Narratives.

Identifying cancer symptoms in electronic health record (EHR) narratives is feasible with natural language processing (NLP). However, more efficient NLP systems are needed to detect various symptoms and distinguish observed symptoms from negated symptoms and medication-related side effects. We evaluated the accuracy of NLP in (1) detecting 14 symptom groups (ie, pain, fatigue, swelling, depressed mood, anxiety, nausea/vomiting, pruritus, headache, shortness of breath, constipation, numbness/tingling, decreased appetite, impaired memory, disturbed sleep) and (2) distinguishing observed symptoms in EHR narratives among patients with cancer. We extracted 902,508 notes for 11,784 unique patients diagnosed with cancer and developed a gold standard corpus of 1,112 notes labeled for presence or absence of 14 symptom groups. We trained an embeddings-augmented NLP system integrating human and machine intelligence and conventional machine learning algorithms. NLP metrics were calculated on a gold standard corpus subset for testing. The interannotator agreement for labeling the gold standard corpus was excellent at 92%. The embeddings-augmented NLP model achieved the best performance (F1 score = 0.877). The highest NLP accuracy was observed in pruritus (F1 score = 0.937) while the lowest accuracy was in swelling (F1 score = 0.787). After classifying the entire data set with embeddings-augmented NLP, we found that 41% of the notes included symptom documentation. Pain was the most documented symptom (29% of all notes) while impaired memory was the least documented (0.7% of all notes). We illustrated the feasibility of detecting 14 symptom groups in EHR narratives and showed that an embeddings-augmented NLP system outperforms conventional machine learning algorithms in detecting symptom information and differentiating observed symptoms from negated symptoms and medication-related side effects.

A tree-based corpus annotated with Cyber-Syndrome, symptoms, and acupoints

Prolonged and over-excessive interaction with cyberspace poses a threat to people’s health and leads to the occurrence of Cyber-Syndrome, which covers not only physiological but also psychological disorders. This paper aims to create a tree-shaped gold-standard corpus that annotates the Cyber-Syndrome, clinical manifestations, and acupoints that can alleviate their symptoms or signs, designating this corpus as CS-A. In the CS-A corpus, this paper defines six entities and relations subject to annotation. There are 448 texts to annotate in total manually. After three rounds of updating the annotation guidelines, the inter-annotator agreement (IAA) improved significantly, resulting in a higher IAA score of 86.05%. The purpose of constructing CS-A corpus is to increase the popularity of Cyber-Syndrome and draw attention to its subtle impact on people’s health. Meanwhile, annotated corpus promotes the development of natural language processing technology. Some model experiments can be implemented based on this corpus, such as optimizing and improving models for discontinuous entity recognition, nested entity recognition, etc. The CS-A corpus has been uploaded to figshare.

Term-BLAST-like alignment tool for concept recognition in noisy clinical texts.

Methods for concept recognition (CR) in clinical texts have largely been tested on abstracts or articles from the medical literature. However, texts from electronic health records (EHRs) frequently contain spelling errors, abbreviations, and other nonstandard ways of representing clinical concepts. Here, we present a method inspired by the BLAST algorithm for biosequence alignment that screens texts for potential matches on the basis of matching k-mer counts and scores candidates based on conformance to typical patterns of spelling errors derived from 2.9 million clinical notes. Our method, the Term-BLAST-like alignment tool (TBLAT) leverages a gold standard corpus for typographical errors to implement a sequence alignment-inspired method for efficient entity linkage. We present a comprehensive experimental comparison of TBLAT with five widely used tools. Experimental results show an increase of 10% in recall on scientific publications and 20% increase in recall on EHR records (when compared against the next best method), hence supporting a significant enhancement of the entity linking task. The method can be used stand-alone or as a complement to existing approaches. Fenominal is a Java library that implements TBLAT for named CR of Human Phenotype Ontology terms and is available at https://github.com/monarch-initiative/fenominal under the GNU General Public License v3.0.

Automated classification of lay health articles using natural language processing: a case study on pregnancy health and postpartum depression.

Evidence suggests that high-quality health education and effective communication within the framework of social support hold significant potential in preventing postpartum depression. Yet, developing trustworthy and engaging health education and communication materials requires extensive expertise and substantial resources. In light of this, we propose an innovative approach that involves leveraging natural language processing (NLP) to classify publicly accessible lay articles based on their relevance and subject matter to pregnancy and mental health. We manually reviewed online lay articles from credible and medically validated sources to create a gold standard corpus. This manual review process categorized the articles based on their pertinence to pregnancy and related subtopics. To streamline and expand the classification procedure for relevance and topics, we employed advanced NLP models such as Random Forest, Bidirectional Encoder Representations from Transformers (BERT), and Generative Pre-trained Transformer model (gpt-3.5-turbo). The gold standard corpus included 392 pregnancy-related articles. Our manual review process categorized the reading materials according to lifestyle factors associated with postpartum depression: diet, exercise, mental health, and health literacy. A BERT-based model performed best (F1 = 0.974) in an end-to-end classification of relevance and topics. In a two-step approach, given articles already classified as pregnancy-related, gpt-3.5-turbo performed best (F1 = 0.972) in classifying the above topics. Utilizing NLP, we can guide patients to high-quality lay reading materials as cost-effective, readily available health education and communication sources. This approach allows us to scale the information delivery specifically to individuals, enhancing the relevance and impact of the materials provided.

Plant Science Knowledge Graph Corpus: a gold standard entity and relation corpus for the molecular plant sciences

Abstract Natural language processing (NLP) techniques can enhance our ability to interpret plant science literature. Many state-of-the-art algorithms for NLP tasks require high-quality labelled data in the target domain, in which entities like genes and proteins, as well as the relationships between entities, are labelled according to a set of annotation guidelines. While there exist such datasets for other domains, these resources need development in the plant sciences. Here, we present the Plant ScIenCe KnowLedgE Graph (PICKLE) corpus, a collection of 250 plant science abstracts annotated with entities and relations, along with its annotation guidelines. The annotation guidelines were refined by iterative rounds of overlapping annotations, in which inter-annotator agreement was leveraged to improve the guidelines. To demonstrate PICKLE’s utility, we evaluated the performance of pretrained models from other domains and trained a new, PICKLE-based model for entity and relation extraction (RE). The PICKLE-trained models exhibit the second-highest in-domain entity performance of all models evaluated, as well as a RE performance that is on par with other models. Additionally, we found that computer science-domain models outperformed models trained on a biomedical corpus (GENIA) in entity extraction, which was unexpected given the intuition that biomedical literature is more similar to PICKLE than computer science. Upon further exploration, we established that the inclusion of new types on which the models were not trained substantially impacts performance. The PICKLE corpus is, therefore, an important contribution to training resources for entity and RE in the plant sciences.

Extracting Reproductive Condition and Habitat Information from Text Using a Transformer-based Information Extraction Pipeline

Understanding the biology underpinning the natural regeneration of plant species in order to make plans for effective reforestation is a complex task. This can be aided by providing access to databases that contain long-term and wide-scale geographical information on species distribution, habitat, and reproduction. Although there exists widely-used biodiversity databases that contain structured information on species and their occurrences, such as the Global Biodiversity Information Facility (GBIF) and the Atlas of Living Australia (ALA), the bulk of knowledge about biodiversity still remains embedded in textual documents. Unstructured information can be made more accessible and useful for large-scale studies if there are tools and services that automatically extract meaningful information from text and store it in structured formats, e.g., open biodiversity databases, ready to be consumed for analysis (Thessen et al. 2022). We aim to enrich biodiversity occurrence databases with information on species reproductive condition and habitat, derived from text. In previous work, we developed unsupervised approaches to extract related habitats and their locations, and related reproductive condition and temporal expressions (Gabud and Batista-Navarro 2018). We built a new unsupervised hybrid approach for relation extraction (RE), which is a combination of classical rule-based pattern-matching methods and transformer-based language models that framed our RE task as a natural language inference (NLI) task. Using our hybrid approach for RE, we were able to extract related biodiversity entities from text even without a large training dataset. In this work, we implement an information extraction (IE) pipeline comprised of a named entity recognition (NER) tool and our hybrid relation extraction (RE) tool. The NER tool is a transformer-based language model that was pretrained on scientific text and then fine-tuned using COPIOUS (Conserving Philippine Biodiversity by Understanding big data; Nguyen et al. 2019), a gold standard corpus containing named entities relevant to species occurrence. We applied the NER tool to automatically annotate geographical location, temporal expression and habitat information contained within sentences. A dictionary-based approach is then used to identify mentions of reproductive conditions in text (e.g., phrases such as "fruited heavily" and "mass flowering"). We then use our hybrid RE tool to extract reproductive condition - temporal expression and habitat - geographical location entity pairs. We test our IE pipeline on the forestry compendium available in the CABI Digital Library (Centre for Agricultural and Biosciences International), and show that our work enables the enrichment of descriptive information on reproductive and habitat conditions of species. This work is a step towards enhancing a biodiversity database with the inclusion of habitat and reproductive condition information extracted from text.

Automated tabulation of clinical trial results: A joint entity and relation extraction approach with transformer-based language representations

Evidence-based medicine, the practice in which healthcare professionals refer to the best available evidence when making decisions, forms the foundation of modern healthcare. However, it relies on labour-intensive systematic reviews, where domain specialists must aggregate and extract information from thousands of publications, primarily of randomised controlled trial (RCT) results, into evidence tables. This paper investigates automating evidence table generation by decomposing the problem across two language processing tasks: named entity recognition, which identifies key entities within text, such as drug names, and relation extraction, which maps their relationships for separating them into ordered tuples. We focus on the automatic tabulation of sentences from published RCT abstracts that report the results of the study outcomes. Two deep neural net models were developed as part of a joint extraction pipeline, using the principles of transfer learning and transformer-based language representations. To train and test these models, a new gold-standard corpus was developed, comprising over 550 result sentences from six disease areas. This approach demonstrated significant advantages, with our system performing well across multiple natural language processing tasks and disease areas, as well as in generalising to disease domains unseen during training. Furthermore, we show these results were achievable through training our models on as few as 170 example sentences. The final system is a proof of concept that the generation of evidence tables can be semi-automated, representing a step towards fully automating systematic reviews.

An analysis of entity normalization evaluation biases in specialized domains

BackgroundEntity normalization is an important information extraction task which has recently gained attention, particularly in the clinical/biomedical and life science domains. On several datasets, state-of-the-art methods perform rather well on popular benchmarks. Yet, we argue that the task is far from resolved.ResultsWe have selected two gold standard corpora and two state-of-the-art methods to highlight some evaluation biases. We present non-exhaustive initial findings on the existence of evaluation problems of the entity normalization task.ConclusionsOur analysis suggests better evaluation practices to support the methodological research in this field.

A study on methods for revising dependency treebanks: in search of gold

Reliably annotated corpora with reliable annotation are a valuable resource for Natural Language Processing, which justifies the search for methods capable of assisting linguistic revision. In this context, we present a study on methods for revising dependency treebanks, investigating the contribution of three different strategies to the corpus review: (i) linguistic rules; (ii) an adaptation of the n-grams method proposed by Boyd et al. (2008) applied to Portuguese; and (iii) Inter-Annotator Disagreement, a linguistically motivated approach that draws inspiration from the human annotation process. The results are promising, and taken together the three methods can lead to the revision of up to 58% of the errors in a specific corpus at the cost of revising only 20% of the corpus. We also present a tool that integrates treebank editing, evaluation and search capabilities with the review methods, as well as a gold-standard Portuguese corpus from the oil and gas domain.

An expectation-maximization framework for comprehensive prediction of isoform-specific functions.

Advances in RNA sequencing technologies have achieved an unprecedented accuracy in the quantification of mRNA isoforms, but our knowledge of isoform-specific functions has lagged behind. There is a need to understand the functional consequences of differential splicing, which could be supported by the generation of accurate and comprehensive isoform-specific gene ontology annotations. We present isoform interpretation, a method that uses expectation-maximization to infer isoform-specific functions based on the relationship between sequence and functional isoform similarity. We predicted isoform-specific functional annotations for 85617 isoforms of 17900 protein-coding human genes spanning a range of 17430 distinct gene ontology terms. Comparison with a gold-standard corpus of manually annotated human isoform functions showed that isoform interpretation significantly outperforms state-of-the-art competing methods. We provide experimental evidence that functionally related isoforms predicted by isoform interpretation show a higher degree of domain sharing and expression correlation than functionally related genes. We also show that isoform sequence similarity correlates better with inferred isoform function than with gene-level function. Source code, documentation, and resource files are freely available under a GNU3 license at https://github.com/TheJacksonLaboratory/isopretEM and https://zenodo.org/record/7594321.

Inter-Annotator Agreement for the Factual Status of Predicates in the TAGFACT Corpus

This paper reports on a study of the inter-annotation agreement to assess the manual annotation of the TAGFACT Gold Standard corpus. This corpus has been created as part of a larger project (TAGFACT), whose final objective is to automatize the classification of the factual status of events in a corpus of Spanish journalistic texts. In our study, six annotators labeled a corpus using the four levels of linguistic description proposed in our project to extract factual information. Each one of these levels has been assessed independently. As expected, the more fine-grained the classification is, the more problematic the annotation. This study identifies some of the most important differences and discusses the main problems encountered to obtain full agreement. We use Cohen’s Kappa to measure inter-annotation agreement as well as descriptive statistical analysis.

NLM-Chem-BC7: manually annotated full-text resources for chemical entity annotation and indexing in biomedical articles.

The automatic recognition of chemical names and their corresponding database identifiers in biomedical text is an important first step for many downstream text-mining applications. The task is even more challenging when considering the identification of these entities in the article's full text and, furthermore, the identification of candidate substances for that article's metadata [Medical Subject Heading (MeSH) article indexing]. The National Library of Medicine (NLM)-Chem track at BioCreative VII aimed to foster the development of algorithms that can predict with high quality the chemical entities in the biomedical literature and further identify the chemical substances that are candidates for article indexing. As a result of this challenge, the NLM-Chem track produced two comprehensive, manually curated corpora annotated with chemical entities and indexed with chemical substances: the chemical identification corpus and the chemical indexing corpus. The NLM-Chem BioCreative VII (NLM-Chem-BC7) Chemical Identification corpus consists of 204 full-text PubMed Central (PMC) articles, fully annotated for chemical entities by 12 NLM indexers for both span (i.e.named entity recognition) and normalization (i.e.entity linking) using MeSH. This resource was used for the training and testing of the Chemical Identification task to evaluate the accuracy of algorithms in predicting chemicals mentioned in recently published full-text articles. The NLM-Chem-BC7 Chemical Indexing corpus consists of 1333 recently published PMC articles, equipped with chemical substance indexing by manual experts at the NLM. This resource was used for the evaluation of the Chemical Indexing task, which evaluated the accuracy of algorithms in predicting the chemicals that should be indexed, i.e.appear in the listing of MeSH terms for the document. This set was further enriched after the challenge in two ways: (i) 11 NLM indexers manually verified each of the candidate terms appearing in the prediction results of the challenge participants, but not in the MeSH indexing, and the chemical indexing terms appearing in the MeSH indexing list, but not in the prediction results, and (ii) the challenge organizers algorithmically merged the chemical entity annotations in the full text for all predicted chemical entities and used a statistical approach to keep those with the highest degree of confidence. As a result, the NLM-Chem-BC7 Chemical Indexing corpus is a gold-standard corpus for chemical indexing of journal articles and a silver-standard corpus for chemical entity identification in full-text journal articles. Together, these resources are currently the most comprehensive resources for chemical entity recognition, and we demonstrate improvements in the chemical entity recognition algorithms. We detail the characteristics of these novel resources and make them available for the community. Database URL: https://ftp.ncbi.nlm.nih.gov/pub/lu/NLM-Chem-BC7-corpus/.

PFSA-ID: an annotated Indonesian corpus and baseline model of public figures statements attributions

PurposeBy far, the corpus for the quotation extraction and quotation attribution tasks in Indonesian is still limited in quantity and depth. This study aims to develop an Indonesian corpus of public figure statements attributions and a baseline model for attribution extraction, so it will contribute to fostering research in information extraction for the Indonesian language.Design/methodology/approachThe methodology is divided into corpus development and extraction model development. During corpus development, data were collected and annotated. The development of the extraction model entails feature extraction, the definition of the model architecture, parameter selection and configuration, model training and evaluation, as well as model selection.FindingsThe Indonesian corpus of public figure statements attribution achieved 90.06% agreement level between the annotator and experts and could serve as a gold standard corpus. Furthermore, the baseline model predicted most labels and achieved 82.026% F-score.Originality/valueTo the best of the authors’ knowledge, the resulting corpus is the first corpus for attribution of public figures’ statements in the Indonesian language, which makes it a significant step for research on attribution extraction in the language. The resulting corpus and the baseline model can be used as a benchmark for further research. Other researchers could follow the methods presented in this paper to develop a new corpus and baseline model for other languages.

Event-Based Clinical Finding Extraction from Radiology Reports with Pre-trained Language Model.

Radiology reports contain a diverse and rich set of clinical abnormalities documented by radiologists during their interpretation of the images. Comprehensive semantic representations of radiological findings would enable a wide range of secondary use applications to support diagnosis, triage, outcomes prediction, and clinical research. In this paper, we present a new corpus of radiology reports annotated with clinical findings. Our annotation schema captures detailed representations of pathologic findings that are observable on imaging (“lesions”) and other types of clinical problems (“medical problems”). The schema used an event-based representation to capture fine-grained details, including assertion, anatomy, characteristics, size, and count. Our gold standard corpus contained a total of 500 annotated computed tomography (CT) reports. We extracted triggers and argument entities using two state-of-the-art deep learning architectures, including BERT. We then predicted the linkages between trigger and argument entities (referred to as argument roles) using a BERT-based relation extraction model. We achieved the best extraction performance using a BERT model pre-trained on 3 million radiology reports from our institution: 90.9–93.4% F1 for finding triggers and 72.0–85.6% F1 for argument roles. To assess model generalizability, we used an external validation set randomly sampled from the MIMIC Chest X-ray (MIMIC-CXR) database. The extraction performance on this validation set was 95.6% for finding triggers and 79.1–89.7% for argument roles, demonstrating that the model generalized well to the cross-institutional data with a different imaging modality. We extracted the finding events from all the radiology reports in the MIMIC-CXR database and provided the extractions to the research community.

BiodivNERE: Gold standard corpora for named entity recognition and relation extraction in the biodiversity domain

Biodiversity is the assortment of life on earth covering evolutionary, ecological, biological, and social forms. To preserve life in all its variety and richness, it is imperative to monitor the current state of biodiversity and its change over time and to understand the forces driving it. This need has resulted in numerous works being published in this field. With this, a large amount of textual data (publications) and metadata (e.g. dataset description) has been generated. To support the management and analysis of these data, two techniques from computer science are of interest, namely Named Entity Recognition (NER) and Relation Extraction (RE). While the former enables better content discovery and understanding, the latter fosters the analysis by detecting connections between entities and, thus, allows us to draw conclusions and answer relevant domain-specific questions. To automatically predict entities and their relations, machine/deep learning techniques could be used. The training and evaluation of those techniques require labelled corpora. In this paper, we present two gold-standard corpora for Named Entity Recognition (NER) and Relation Extraction (RE) generated from biodiversity datasets metadata and abstracts that can be used as evaluation benchmarks for the development of new computer-supported tools that require machine learning or deep learning techniques. These corpora are manually labelled and verified by biodiversity experts. In addition, we explain the detailed steps of constructing these datasets. Moreover, we demonstrate the underlying ontology for the classes and relations used to annotate such corpora.

A Gated Recurrent Unit based architecture for recognizing ontology concepts from biological literature

BackgroundAnnotating scientific literature with ontology concepts is a critical task in biology and several other domains for knowledge discovery. Ontology based annotations can power large-scale comparative analyses in a wide range of applications ranging from evolutionary phenotypes to rare human diseases to the study of protein functions. Computational methods that can tag scientific text with ontology terms have included lexical/syntactic methods, traditional machine learning, and most recently, deep learning.ResultsHere, we present state of the art deep learning architectures based on Gated Recurrent Units for annotating text with ontology concepts. We use the Colorado Richly Annotated Full Text Corpus (CRAFT) as a gold standard for training and testing. We explore a number of additional information sources including NCBI’s BioThesauraus and Unified Medical Language System (UMLS) to augment information from CRAFT for increasing prediction accuracy. Our best model results in a 0.84 F1 and semantic similarity.ConclusionThe results shown here underscore the impact for using deep learning architectures for automatically recognizing ontology concepts from literature. The augmentation of the models with biological information beyond that present in the gold standard corpus shows a distinct improvement in prediction accuracy.

Development and validation of an automated basal cell carcinoma histopathology information extraction system using natural language processing.

Routinely collected healthcare data are a powerful research resource, but often lack detailed disease-specific information that is collected in clinical free text such as histopathology reports. We aim to use natural Language Processing (NLP) techniques to extract detailed clinical and pathological information from histopathology reports to enrich routinely collected data. We used the general architecture for text engineering (GATE) framework to build an NLP information extraction system using rule-based techniques. During validation, we deployed our rule-based NLP pipeline on 200 previously unseen, de-identified and pseudonymised basal cell carcinoma (BCC) histopathological reports from Swansea Bay University Health Board, Wales, UK. The results of our algorithm were compared with gold standard human annotation by two independent and blinded expert clinicians involved in skin cancer care. We identified 11,224 items of information with a mean precision, recall, and F1 score of 86.0% (95% CI: 75.1-96.9), 84.2% (95% CI: 72.8-96.1), and 84.5% (95% CI: 73.0-95.1), respectively. The difference between clinician annotator F1 scores was 7.9% in comparison with 15.5% between the NLP pipeline and the gold standard corpus. Cohen's Kappa score on annotated tokens was 0.85. Using an NLP rule-based approach for named entity recognition in BCC, we have been able to develop and validate a pipeline with a potential application in improving the quality of cancer registry data, supporting service planning, and enhancing the quality of routinely collected data for research.

Classifying the lifestyle status for Alzheimer’s disease from clinical notes using deep learning with weak supervision

BackgroundSince no effective therapies exist for Alzheimer’s disease (AD), prevention has become more critical through lifestyle status changes and interventions. Analyzing electronic health records (EHRs) of patients with AD can help us better understand lifestyle’s effect on AD. However, lifestyle information is typically stored in clinical narratives. Thus, the objective of the study was to compare different natural language processing (NLP) models on classifying the lifestyle statuses (e.g., physical activity and excessive diet) from clinical texts in English.MethodsBased on the collected concept unique identifiers (CUIs) associated with the lifestyle status, we extracted all related EHRs for patients with AD from the Clinical Data Repository (CDR) of the University of Minnesota (UMN). We automatically generated labels for the training data by using a rule-based NLP algorithm. We conducted weak supervision for pre-trained Bidirectional Encoder Representations from Transformers (BERT) models and three traditional machine learning models as baseline models on the weakly labeled training corpus. These models include the BERT base model, PubMedBERT (abstracts + full text), PubMedBERT (only abstracts), Unified Medical Language System (UMLS) BERT, Bio BERT, Bio-clinical BERT, logistic regression, support vector machine, and random forest. The rule-based model used for weak supervision was tested on the GSC for comparison. We performed two case studies: physical activity and excessive diet, in order to validate the effectiveness of BERT models in classifying lifestyle status for all models were evaluated and compared on the developed Gold Standard Corpus (GSC) on the two case studies.ResultsThe UMLS BERT model achieved the best performance for classifying status of physical activity, with its precision, recall, and F-1 scores of 0.93, 0.93, and 0.92, respectively. Regarding classifying excessive diet, the Bio-clinical BERT model showed the best performance with precision, recall, and F-1 scores of 0.93, 0.93, and 0.93, respectively.ConclusionThe proposed approach leveraging weak supervision could significantly increase the sample size, which is required for training the deep learning models. By comparing with the traditional machine learning models, the study also demonstrates the high performance of BERT models for classifying lifestyle status for Alzheimer’s disease in clinical notes.

PhenoDEF: a corpus for annotating sentences with information of phenotype definitions in biomedical literature

BackgroundAdverse events induced by drug-drug interactions are a major concern in the United States. Current research is moving toward using electronic health record (EHR) data, including for adverse drug events discovery. One of the first steps in EHR-based studies is to define a phenotype for establishing a cohort of patients. However, phenotype definitions are not readily available for all phenotypes. One of the first steps of developing automated text mining tools is building a corpus. Therefore, this study aimed to develop annotation guidelines and a gold standard corpus to facilitate building future automated approaches for mining phenotype definitions contained in the literature. Furthermore, our aim is to improve the understanding of how these published phenotype definitions are presented in the literature and how we annotate them for future text mining tasks.ResultsTwo annotators manually annotated the corpus on a sentence-level for the presence of evidence for phenotype definitions. Three major categories (inclusion, intermediate, and exclusion) with a total of ten dimensions were proposed characterizing major contextual patterns and cues for presenting phenotype definitions in published literature. The developed annotation guidelines were used to annotate the corpus that contained 3971 sentences: 1923 out of 3971 (48.4%) for the inclusion category, 1851 out of 3971 (46.6%) for the intermediate category, and 2273 out of 3971 (57.2%) for exclusion category. The highest number of annotated sentences was 1449 out of 3971 (36.5%) for the “Biomedical & Procedure” dimension. The lowest number of annotated sentences was 49 out of 3971 (1.2%) for “The use of NLP”. The overall percent inter-annotator agreement was 97.8%. Percent and Kappa statistics also showed high inter-annotator agreement across all dimensions.ConclusionsThe corpus and annotation guidelines can serve as a foundational informatics approach for annotating and mining phenotype definitions in literature, and can be used later for text mining applications.