HPO2Vec+: Leveraging heterogeneous knowledge resources to enrich node embeddings for the Human Phenotype Ontology

Performance and clinical utility of a new supervised machine-learning pipeline in detecting rare ciliopathy patients based on deep phenotyping from electronic health records and semantic similarity

BACKGROUND

Rare diseases affect approximately 400 million people worldwide. Many of them suffer from delayed diagnosis. Among them, NPHP1-related renal ciliopathies need to be diagnosed as early as possible as potential treatments have been recently investigated with promising results. Our objective was to develop a supervised machine learning pipeline for the detection of NPHP1 ciliopathy patients from a large number of nephrology patients using electronic health records (EHRs).

METHODS AND RESULTS

We designed a pipeline combining a phenotyping module re-using unstructured EHR data, a semantic similarity module to address the phenotype dependence, a feature selection step to deal with high dimensionality, an undersampling step to address the class imbalance, and a classification step with multiple train-test split for the small number of rare cases. The pipeline was applied to thirty NPHP1 patients and 7231 controls and achieved good performances (sensitivity 86% with specificity 90%). A qualitative review of the EHRs of 40 misclassified controls showed that 25% had phenotypes belonging to the ciliopathy spectrum, which demonstrates the ability of our system to detect patients with similar conditions.

CONCLUSIONS

Our pipeline reached very encouraging performance scores for pre-diagnosing ciliopathy patients. The identified patients could then undergo genetic testing. The same data-driven approach can be adapted to other rare diseases facing underdiagnosis challenges.

Enriching representation learning using 53 million patient notes through human phenotype ontology embedding

The Human Phenotype Ontology (HPO) is a dictionary of >15,000 clinical phenotypic terms with defined semantic relationships, developed to standardize phenotypic analysis. Over the last decade, the HPO has been used to accelerate the implementation of precision medicine into clinical practice. In addition, recent research in representation learning, specifically in graph embedding, has led to notable progress in automated prediction via learned features. Here, we present a novel approach to phenotype representation by incorporating phenotypic frequencies based on 53 million full-text health care notes from >1.5 million individuals. We demonstrate the efficacy of our proposed phenotype embedding technique by comparing our work to existing phenotypic similarity-measuring methods. Using phenotype frequencies in our embedding technique, we are able to identify phenotypic similarities that surpass current computational models. Furthermore, our embedding technique exhibits a high degree of agreement with domain experts' judgment. By transforming complex and multidimensional phenotypes from the HPO format into vectors, our proposed method enables efficient representation of these phenotypes for downstream tasks that require deep phenotyping. This is demonstrated in a patient similarity analysis and can further be applied to disease trajectory and risk prediction.

FAIRification of health-related data using semantic web technologies in the Swiss Personalized Health Network

The Swiss Personalized Health Network (SPHN) is a government-funded initiative developing federated infrastructures for a responsible and efficient secondary use of health data for research purposes in compliance with the FAIR principles (Findable, Accessible, Interoperable and Reusable). We built a common standard infrastructure with a fit-for-purpose strategy to bring together health-related data and ease the work of both data providers to supply data in a standard manner and researchers by enhancing the quality of the collected data. As a result, the SPHN Resource Description Framework (RDF) schema was implemented together with a data ecosystem that encompasses data integration, validation tools, analysis helpers, training and documentation for representing health metadata and data in a consistent manner and reaching nationwide data interoperability goals. Data providers can now efficiently deliver several types of health data in a standardised and interoperable way while a high degree of flexibility is granted for the various demands of individual research projects. Researchers in Switzerland have access to FAIR health data for further use in RDF triplestores.

Defining the distance between diseases using SNOMED CT embeddings

Characterizing disease relationships is essential to biomedical research to understand disease etiology and improve clinical decision-making. Measurements of distance between disease pairs enable valuable research tasks, such as subgrouping patients and identifying common time courses of disease onset. Distance metrics developed in prior work focused on smaller, targeted disease sets. Distance metrics covering all diseases have not yet been defined, which limits the applications to a broader disease spectrum. Our current study defines disease distances for all disease pairs within the International Classification of Diseases, version 10 (ICD-10), the diagnostic classification system universally used in electronic health records. Our proposed distance is computed based on a biomedical ontology, SNOMED CT (Systemized Nomenclature of Medicine, Clinical Terms), which can also be viewed as a structured knowledge graph. We compared the knowledge graph-based metric to three other distance metrics based on the hierarchical structure of ICD, clinical comorbidity, and genetic correlation, to evaluate how each may capture similar or unique aspects of disease relationships. We show that our knowledge graph-based distance metric captures known phenotypic, clinical, and molecular characteristics at a finer granularity than the other three. With the continued growth of using electronic health records data for research, we believe that our distance metric will play an important role in subgrouping patients for precision health, and enabling individualized disease prevention and treatments.

Machine learning approaches for electronic health records phenotyping: a methodical review

OBJECTIVE

Accurate and rapid phenotyping is a prerequisite to leveraging electronic health records for biomedical research. While early phenotyping relied on rule-based algorithms curated by experts, machine learning (ML) approaches have emerged as an alternative to improve scalability across phenotypes and healthcare settings. This study evaluates ML-based phenotyping with respect to (1) the data sources used, (2) the phenotypes considered, (3) the methods applied, and (4) the reporting and evaluation methods used.

MATERIALS AND METHODS

We searched PubMed and Web of Science for articles published between 2018 and 2022. After screening 850 articles, we recorded 37 variables on 100 studies.

RESULTS

Most studies utilized data from a single institution and included information in clinical notes. Although chronic conditions were most commonly considered, ML also enabled the characterization of nuanced phenotypes such as social determinants of health. Supervised deep learning was the most popular ML paradigm, while semi-supervised and weakly supervised learning were applied to expedite algorithm development and unsupervised learning to facilitate phenotype discovery. ML approaches did not uniformly outperform rule-based algorithms, but deep learning offered a marginal improvement over traditional ML for many conditions.

DISCUSSION

Despite the progress in ML-based phenotyping, most articles focused on binary phenotypes and few articles evaluated external validity or used multi-institution data. Study settings were infrequently reported and analytic code was rarely released.

CONCLUSION

Continued research in ML-based phenotyping is warranted, with emphasis on characterizing nuanced phenotypes, establishing reporting and evaluation standards, and developing methods to accommodate misclassified phenotypes due to algorithm errors in downstream applications.

Patient-Patient Similarity-Based Screening of a Clinical Data Warehouse to Support Ciliopathy Diagnosis

A timely diagnosis is a key challenge for many rare diseases. As an expanding group of rare and severe monogenic disorders with a broad spectrum of clinical manifestations, ciliopathies, notably renal ciliopathies, suffer from important underdiagnosis issues. Our objective is to develop an approach for screening large-scale clinical data warehouses and detecting patients with similar clinical manifestations to those from diagnosed ciliopathy patients. We expect that the top-ranked similar patients will benefit from genetic testing for an early diagnosis. The dependence and relatedness between phenotypes were taken into account in our similarity model through medical concept embedding. The relevance of each phenotype to each patient was also considered by adjusted aggregation of phenotype similarity into patient similarity. A ranking model based on the best-subtype-average similarity was proposed to address the phenotypic overlapping and heterogeneity of ciliopathies. Our results showed that using less than one-tenth of learning sources, our language and center specific embedding provided comparable or better performances than other existing medical concept embeddings. Combined with the best-subtype-average ranking model, our patient-patient similarity-based screening approach was demonstrated effective in two large scale unbalanced datasets containing approximately 10,000 and 60,000 controls with kidney manifestations in the clinical data warehouse (about 2 and 0.4% of prevalence, respectively). Our approach will offer the opportunity to identify candidate patients who could go through genetic testing for ciliopathy. Earlier diagnosis, before irreversible end-stage kidney disease, will enable these patients to benefit from appropriate follow-up and novel treatments that could alleviate kidney dysfunction.

CADA: phenotype-driven gene prioritization based on a case-enriched knowledge graph

Many rare syndromes can be well described and delineated from other disorders by a combination of characteristic symptoms. These phenotypic features are best documented with terms of the Human Phenotype Ontology (HPO), which are increasingly used in electronic health records (EHRs), too. Many algorithms that perform HPO-based gene prioritization have also been developed; however, the performance of many such tools suffers from an over-representation of atypical cases in the medical literature. This is certainly the case if the algorithm cannot handle features that occur with reduced frequency in a disorder. With Cada, we built a knowledge graph based on both case annotations and disorder annotations. Using network representation learning, we achieve gene prioritization by link prediction. Our results suggest that Cada exhibits superior performance particularly for patients that present with the pathognomonic findings of a disease. Additionally, information about the frequency of occurrence of a feature can readily be incorporated, when available. Crucial in the design of our approach is the use of the growing amount of phenotype–genotype information that diagnostic labs deposit in databases such as ClinVar. By this means, Cada is an ideal reference tool for differential diagnostics in rare disorders that can also be updated regularly.

PhenoTagger: a hybrid method for phenotype concept recognition using human phenotype ontology

MOTIVATION

Automatic phenotype concept recognition from unstructured text remains a challenging task in biomedical text mining research. Previous works that address the task typically use dictionary-based matching methods, which can achieve high precision but suffer from lower recall. Recently, machine learning-based methods have been proposed to identify biomedical concepts, which can recognize more unseen concept synonyms by automatic feature learning. However, most methods require large corpora of manually annotated data for model training, which is difficult to obtain due to the high cost of human annotation.

RESULTS

In this article, we propose PhenoTagger, a hybrid method that combines both dictionary and machine learning-based methods to recognize Human Phenotype Ontology (HPO) concepts in unstructured biomedical text. We first use all concepts and synonyms in HPO to construct a dictionary, which is then used to automatically build a distantly supervised training dataset for machine learning. Next, a cutting-edge deep learning model is trained to classify each candidate phrase (n-gram from input sentence) into a corresponding concept label. Finally, the dictionary and machine learning-based prediction results are combined for improved performance. Our method is validated with two HPO corpora, and the results show that PhenoTagger compares favorably to previous methods. In addition, to demonstrate the generalizability of our method, we retrained PhenoTagger using the disease ontology MEDIC for disease concept recognition to investigate the effect of training on different ontologies. Experimental results on the NCBI disease corpus show that PhenoTagger without requiring manually annotated training data achieves competitive performance as compared with state-of-the-art supervised methods.

AVAILABILITYAND IMPLEMENTATION

The source code, API information and data for PhenoTagger are freely available at https://github.com/ncbi-nlp/PhenoTagger.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Comparative effectiveness of medical concept embedding for feature engineering in phenotyping

Objective

Feature engineering is a major bottleneck in phenotyping. Properly learned medical concept embeddings (MCEs) capture the semantics of medical concepts, thus are useful for retrieving relevant medical features in phenotyping tasks. We compared the effectiveness of MCEs learned from knowledge graphs and electronic healthcare records (EHR) data in retrieving relevant medical features for phenotyping tasks.

Materials and Methods

We implemented 5 embedding methods including node2vec, singular value decomposition (SVD), LINE, skip-gram, and GloVe with 2 data sources: (1) knowledge graphs obtained from the observational medical outcomes partnership (OMOP) common data model; and (2) patient-level data obtained from the OMOP compatible electronic health records (EHR) from Columbia University Irving Medical Center (CUIMC). We used phenotypes with their relevant concepts developed and validated by the electronic medical records and genomics (eMERGE) network to evaluate the performance of learned MCEs in retrieving phenotype-relevant concepts. Hits@k% in retrieving phenotype-relevant concepts based on a single and multiple seed concept(s) was used to evaluate MCEs.

Results

Among all MCEs, MCEs learned by using node2vec with knowledge graphs showed the best performance. Of MCEs based on knowledge graphs and EHR data, MCEs learned by using node2vec with knowledge graphs and MCEs learned by using GloVe with EHR data outperforms other MCEs, respectively.

Conclusion

MCE enables scalable feature engineering tasks, thereby facilitating phenotyping. Based on current phenotyping practices, MCEs learned by using knowledge graphs constructed by hierarchical relationships among medical concepts outperformed MCEs learned by using EHR data.

Recommendations for patient similarity classes: results of the AMIA 2019 workshop on defining patient similarity

Defining patient-to-patient similarity is essential for the development of precision medicine in clinical care and research. Conceptually, the identification of similar patient cohorts appears straightforward; however, universally accepted definitions remain elusive. Simultaneously, an explosion of vendors and published algorithms have emerged and all provide varied levels of functionality in identifying patient similarity categories. To provide clarity and a common framework for patient similarity, a workshop at the American Medical Informatics Association 2019 Annual Meeting was convened. This workshop included invited discussants from academics, the biotechnology industry, the FDA, and private practice oncology groups. Drawing from a broad range of backgrounds, workshop participants were able to coalesce around 4 major patient similarity classes: (1) feature, (2) outcome, (3) exposure, and (4) mixed-class. This perspective expands into these 4 subtypes more critically and offers the medical informatics community a means of communicating their work on this important topic.

Computational analysis of 10,860 phenotypic annotations in individuals with SCN2A-related disorders

Purpose

Pathogenic variants in SCN2A cause a wide range of neurodevelopmental phenotypes. Reports of genotype–phenotype correlations are often anecdotal, and the available phenotypic data have not been systematically analyzed.

Methods

We extracted phenotypic information from primary descriptions of SCN2A-related disorders in the literature between 2001 and 2019, which we coded in Human Phenotype Ontology (HPO) terms. With higher-level phenotype terms inferred by the HPO structure, we assessed the frequencies of clinical features and investigated the association of these features with variant classes and locations within the Na_V1.2 protein.

Results

We identified 413 unrelated individuals and derived a total of 10,860 HPO terms with 562 unique terms. Protein-truncating variants were associated with autism and behavioral abnormalities. Missense variants were associated with neonatal onset, epileptic spasms, and seizures, regardless of type. Phenotypic similarity was identified in 8/62 recurrent SCN2A variants. Three independent principal components accounted for 33% of the phenotypic variance, allowing for separation of gain-of-function versus loss-of-function variants with good performance.

Conclusion

Our work shows that translating clinical features into a computable format using a standardized language allows for quantitative phenotype analysis, mapping the phenotypic landscape of SCN2A-related disorders in unprecedented detail and revealing genotype–phenotype correlations along a multidimensional spectrum.

Constructing co-occurrence network embeddings to assist association extraction for COVID-19 and other coronavirus infectious diseases

Objective

As coronavirus disease 2019 (COVID-19) started its rapid emergence and gradually transformed into an unprecedented pandemic, the need for having a knowledge repository for the disease became crucial. To address this issue, a new COVID-19 machine-readable dataset known as the COVID-19 Open Research Dataset (CORD-19) has been released. Based on this, our objective was to build a computable co-occurrence network embeddings to assist association detection among COVID-19–related biomedical entities.

Materials and Methods

Leveraging a Linked Data version of CORD-19 (ie, CORD-19-on-FHIR), we first utilized SPARQL to extract co-occurrences among chemicals, diseases, genes, and mutations and build a co-occurrence network. We then trained the representation of the derived co-occurrence network using node2vec with 4 edge embeddings operations (L1, L2, Average, and Hadamard). Six algorithms (decision tree, logistic regression, support vector machine, random forest, naïve Bayes, and multilayer perceptron) were applied to evaluate performance on link prediction. An unsupervised learning strategy was also developed incorporating the t-SNE (t-distributed stochastic neighbor embedding) and DBSCAN (density-based spatial clustering of applications with noise) algorithms for case studies.

Results

The random forest classifier showed the best performance on link prediction across different network embeddings. For edge embeddings generated using the Average operation, random forest achieved the optimal average precision of 0.97 along with a F1 score of 0.90. For unsupervised learning, 63 clusters were formed with silhouette score of 0.128. Significant associations were detected for 5 coronavirus infectious diseases in their corresponding subgroups.

Conclusions

In this study, we constructed COVID-19–centered co-occurrence network embeddings. Results indicated that the generated embeddings were able to extract significant associations for COVID-19 and coronavirus infectious diseases.

Clinical concept extraction: A methodology review

BACKGROUND

Concept extraction, a subdomain of natural language processing (NLP) with a focus on extracting concepts of interest, has been adopted to computationally extract clinical information from text for a wide range of applications ranging from clinical decision support to care quality improvement.

OBJECTIVES

In this literature review, we provide a methodology review of clinical concept extraction, aiming to catalog development processes, available methods and tools, and specific considerations when developing clinical concept extraction applications.

METHODS

Based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, a literature search was conducted for retrieving EHR-based information extraction articles written in English and published from January 2009 through June 2019 from Ovid MEDLINE In-Process & Other Non-Indexed Citations, Ovid MEDLINE, Ovid EMBASE, Scopus, Web of Science, and the ACM Digital Library.

RESULTS

A total of 6,686 publications were retrieved. After title and abstract screening, 228 publications were selected. The methods used for developing clinical concept extraction applications were discussed in this review.

Ontologies, Knowledge Representation, and Machine Learning for Translational Research: Recent Contributions

Objectives : To select, present, and summarize the most relevant papers published in 2018 and 2019 in the field of Ontologies and Knowledge Representation, with a particular focus on the intersection between Ontologies and Machine Learning.

Methods : A comprehensive review of the medical informatics literature was performed to select the most interesting papers published in 2018 and 2019 and that document the utility of ontologies for computational analysis, including machine learning.

Results : Fifteen articles were selected for inclusion in this survey paper. The chosen articles belong to three major themes: (i) the identification of phenotypic abnormalities in electronic health record (EHR) data using the Human Phenotype Ontology ; (ii) word and node embedding algorithms to supplement natural language processing (NLP) of EHRs and other medical texts; and (iii) hybrid ontology and NLP-based approaches to extracting structured and unstructured components of EHRs.

Conclusion : Unprecedented amounts of clinically relevant data are now available for clinical and research use. Machine learning is increasingly being applied to these data sources for predictive analytics, precision medicine, and differential diagnosis. Ontologies have become an essential component of software pipelines designed to extract, code, and analyze clinical information by machine learning algorithms. The intersection of machine learning and semantics is proving to be an innovative space in clinical research.