A content-based literature recommendation system for datasets to improve data reusability - A case study on Gene Expression Omnibus (GEO) datasets.

OBJECTIVE: The centrality of data to biomedical research is difficult to understate, and the same is true for the importance of the biomedical literature in disseminating empirical findings to scientific questions made on such data. But the connections between the literature and related datasets are often weak, hampering the ability of scientists to easily move between existing datasets and existing findings to derive new scientific hypotheses. This work aims to recommend relevant literature articles for datasets with the ultimate goal of increasing the productivity of researchers. Our approach to literature recommendation for datasets is a part of the dataset reusability platform developed at the University Texas Health Science Center at Houston for datasets related to gene expression. This platform incorporates datasets from Gene Expression Omnibus (GEO). An average of 34 datasets were added to GEO daily in the last five years (i.e. 2014 to 2018), demonstrating the need for automatic methods to connect these datasets with relevant literature. The relevant literature for a given dataset may describe that dataset, provide a scientific finding based on that dataset, or even describe prior and related work to the dataset's topic that is of interest to users of the dataset. MATERIALS AND METHODS: We adopt an information retrieval paradigm for literature recommendation. In our experiments, distributional semantic features are created from the title and abstract of MEDLINE articles. Then, related articles are identified for datasets in GEO. We evaluate multiple distributional methods such as TF-IDF, BM25, Latent Semantic Analysis, Latent Dirichlet Allocation, word2vec, and doc2vec. Top similar papers are recommended for each dataset using cosine similarity between the dataset's vector representation and every paper's vector representation. We also propose several novel re-ranking and normalization methods over embeddings to improve the recommendations. RESULTS: The top-performing literature recommendation technique achieved a strict precision at 10 of 0.8333 and a partial precision at 10 of 0.9000 using BM25 based on a manual evaluation of 36 datasets. Evaluation on a larger, automatically-collected benchmark shows small but consistent gains by emphasizing the similarity of dataset and article titles. CONCLUSION: This work is the first step toward developing a literature recommendation tool by recommending relevant literature for datasets. This will hopefully lead to better data reuse experience.

Detection of Personal and Family History of Suicidal Thoughts and Behaviors using Deep Learning and Natural Language Processing: A Multi-Site Study.

Objective: Personal and family history of suicidal thoughts and behaviors (PSH and FSH, respectively) are significant risk factors associated with future suicide events. These are often captured in narrative clinical notes in electronic health records (EHRs). Collaboratively, Weill Cornell Medicine (WCM), Northwestern Medicine (NM), and the University of Florida (UF) developed and validated deep learning (DL)-based natural language processing (NLP) tools to detect PSH and FSH from such notes. The tool's performance was further benchmarked against a method relying exclusively on ICD-9/10 diagnosis codes. Materials and Methods: We developed DL-based NLP tools utilizing pre-trained transformer models Bio_ClinicalBERT and GatorTron, and compared them with expert-informed, rule-based methods. The tools were initially developed and validated using manually annotated clinical notes at WCM. Their portability and performance were further evaluated using clinical notes at NM and UF. Results: The DL tools outperformed the rule-based NLP tool in identifying PSH and FHS. For detecting PSH, the rule-based system obtained an F1-score of 0.75 ± 0.07, while the Bio_ClinicalBERT and GatorTron DL tools scored 0.83 ± 0.09 and 0.84 ± 0.07, respectively. For detecting FSH, the rule-based NLP tool's F1-score was 0.69 ± 0.11, compared to 0.89 ± 0.10 for Bio_ClinicalBERT and 0.92 ± 0.07 for GatorTron. For the gold standard corpora across the three sites, only 2.2% (WCM), 9.3% (NM), and 7.8% (UF) of patients reported to have an ICD-9/10 diagnosis code for suicidal thoughts and behaviors prior to the clinical notes report date. The best performing GatorTron DL tool identified 93.0% (WCM), 80.4% (NM), and 89.0% (UF) of patients with documented PSH, and 85.0%(WCM), 89.5%(NM), and 100%(UF) of patients with documented FSH in their notes. Discussion: While PSH and FSH are significant risk factors for future suicide events, little effort has been made previously to identify individuals with these history. To address this, we developed a transformer based DL method and compared with conventional rule-based NLP approach. The varying effectiveness of the rule-based tools across sites suggests a need for improvement in its dictionary-based approach. In contrast, the performances of the DL tools were higher and comparable across sites. Furthermore, DL tools were fine-tuned using only small number of annotated notes at each site, underscores its greater adaptability to local documentation practices and lexical variations. Conclusion: Variations in local documentation practices across health care systems pose challenges to rule-based NLP tools. In contrast, the developed DL tools can effectively extract PSH and FSH information from unstructured clinical notes. These tools will provide clinicians with crucial information for assessing and treating patients at elevated risk for suicide who are rarely been diagnosed.

A novel NIH research grant recommender using BERT.

Research grants are important for researchers to sustain a good position in academia. There are many grant opportunities available from different funding agencies. However, finding relevant grant announcements is challenging and time-consuming for researchers. To resolve the problem, we proposed a grant announcements recommendation system for the National Institute of Health (NIH) grants using researchers' publications. We formulated the recommendation as a classification problem and proposed a recommender using state-of-the-art deep learning techniques: i.e. Bidirectional Encoder Representations from Transformers (BERT), to capture intrinsic, non-linear relationship between researchers' publications and grants announcements. Internal and external evaluations were conducted to assess the system's usefulness. During internal evaluations, the grant citations were used to establish grant-publication ground truth, and results were evaluated against Recall@k, Precision@k, Mean reciprocal rank (MRR) and Area under the Receiver Operating Characteristic curve (ROC-AUC). During external evaluations, researchers' publications were clustered using Dirichlet Process Mixture Model (DPMM), recommended grants by our model were then aggregated per cluster through Recency Weight, and finally researchers were invited to provide ratings to recommendations to calculate Precision@k. For comparison, baseline recommenders using Okapi Best Matching (BM25), Term-Frequency Inverse Document Frequency (TF-IDF), doc2vec, and Naïve Bayes (NB) were also developed. Both internal and external evaluations (all metrics) revealed favorable performances of our proposed BERT-based recommender.

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

Objective: 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. Materials and methods: 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). Results: 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. Discussion: 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.

Classifying Cyber-Risky Clinical Notes by Employing Natural Language Processing.

Clinical notes, which can be embedded into electronic medical records, document patient care delivery and summarize interactions between healthcare providers and patients. These clinical notes directly inform patient care and can also indirectly inform research and quality/safety metrics, among other indirect metrics. Recently, some states within the United States of America require patients to have open access to their clinical notes to improve the exchange of patient information for patient care. Thus, developing methods to assess the cyber risks of clinical notes before sharing and exchanging data is critical. While existing natural language processing techniques are geared to de-identify clinical notes, to the best of our knowledge, few have focused on classifying sensitive-information risk, which is a fundamental step toward developing effective, widespread protection of patient health information. To bridge this gap, this research investigates methods for identifying security/privacy risks within clinical notes. The classification either can be used upstream to identify areas within notes that likely contain sensitive information or downstream to improve the identification of clinical notes that have not been entirely de-identified. We develop several models using unigram and word2vec features with different classifiers to categorize sentence risk. Experiments on i2b2 de-identification dataset show that the SVM classifier using word2vec features obtained a maximum F1-score of 0.792. Future research involves articulation and differentiation of risk in terms of different global regulatory requirements.

A content-based dataset recommendation system for researchers-a case study on Gene Expression Omnibus (GEO) repository.

It is a growing trend among researchers to make their data publicly available for experimental reproducibility and data reusability. Sharing data with fellow researchers helps in increasing the visibility of the work. On the other hand, there are researchers who are inhibited by the lack of data resources. To overcome this challenge, many repositories and knowledge bases have been established to date to ease data sharing. Further, in the past two decades, there has been an exponential increase in the number of datasets added to these dataset repositories. However, most of these repositories are domain-specific, and none of them can recommend datasets to researchers/users. Naturally, it is challenging for a researcher to keep track of all the relevant repositories for potential use. Thus, a dataset recommender system that recommends datasets to a researcher based on previous publications can enhance their productivity and expedite further research. This work adopts an information retrieval (IR) paradigm for dataset recommendation. We hypothesize that two fundamental differences exist between dataset recommendation and PubMed-style biomedical IR beyond the corpus. First, instead of keywords, the query is the researcher, embodied by his or her publications. Second, to filter the relevant datasets from non-relevant ones, researchers are better represented by a set of interests, as opposed to the entire body of their research. This second approach is implemented using a non-parametric clustering technique. These clusters are used to recommend datasets for each researcher using the cosine similarity between the vector representations of publication clusters and datasets. The maximum normalized discounted cumulative gain at 10 (NDCG@10), precision at 10 (p@10) partial and p@10 strict of 0.89, 0.78 and 0.61, respectively, were obtained using the proposed method after manual evaluation by five researchers. As per the best of our knowledge, this is the first study of its kind on content-based dataset recommendation. We hope that this system will further promote data sharing, offset the researchers' workload in identifying the right dataset and increase the reusability of biomedical datasets. Database URL: http://genestudy.org/recommends/#/.

Mental Health Severity Detection from Psychological Forum Data using Domain-Specific Unlabelled Data.

Mental health has become a growing concern in the medical field, yet remains difficult to study due to both privacy concerns and the lack of objectively quantifiable measurements (e.g., lab tests, physical exams). Instead, the data that is available for mental health is largely based on subjective accounts of a patient's experience, and thus typically is expressed exclusively in text. An important source of such data comes from online sources and directly from the patient, including many forms of social media. In this work, we utilize the datasets provided by the CLPsych shared tasks in 2016 and 2017, derived from online forum posts of ReachOut which have been manually classified according to mental health severity. We implemented an automated severity labeling system using different machine and deep learning algorithms. Our approach combines both supervised and semi-supervised embedding methods using corpus from ReachOut (both labeled and unlabelled) and WebMD (unlabelled). Metadata, syntactic, semantic, and embedding features were used to classify the posts into four categories (green, amber, red, and crisis). The developed systems outperformed other state-of-the-art systems developed on the ReachOut dataset and obtained the maximum micro- averaged F-scores of 0.86 and 0.80 for CLPsych 2016 and 2017 test datasets, respectively, using the above features.

An informatics research platform to make public gene expression time-course datasets reusable for more scientific discoveries.

The exponential growth of genomic/genetic data in the era of Big Data demands new solutions for making these data findable, accessible, interoperable and reusable. In this article, we present a web-based platform named Gene Expression Time-Course Research (GETc) Platform that enables the discovery and visualization of time-course gene expression data and analytical results from the NIH/NCBI-sponsored Gene Expression Omnibus (GEO). The analytical results are produced from an analytic pipeline based on the ordinary differential equation model. Furthermore, in order to extract scientific insights from these results and disseminate the scientific findings, close and efficient collaborations between domain-specific experts from biomedical and scientific fields and data scientists is required. Therefore, GETc provides several recommendation functions and tools to facilitate effective collaborations. GETc platform is a very useful tool for researchers from the biomedical genomics community to present and communicate large numbers of analysis results from GEO. It is generalizable and broadly applicable across different biomedical research areas. GETc is a user-friendly and efficient web-based platform freely accessible at http://genestudy.org/.

A Semantic Parsing Method for Mapping Clinical Questions to Logical Forms.

This paper presents a method for converting natural language questions about structured data in the electronic health record (EHR) into logical forms. The logical forms can then subsequently be converted to EHR-dependent structured queries. The natural language processing task, known as semantic parsing, has the potential to convert questions to logical forms with extremely high precision, resulting in a system that is usable and trusted by clinicians for real-time use in clinical settings. We propose a hybrid semantic parsing method, combining rule-based methods with a machine learning-based classifier. The overall semantic parsing precision on a set of 212 questions is 95.6%. The parser's rules furthermore allow it to "know what it does not know", enabling the system to indicate when unknown terms prevent it from understanding the question's full logical structure. When combined with a module for converting a logical form into an EHR-dependent query, this high-precision approach allows for a question answering system to provide a user with a single, verifiably correct answer.