Annotating and detecting phenotypic information for chronic obstructive pulmonary disease

PhenoRerank: A re-ranking model for phenotypic concept recognition pre-trained on human phenotype ontology

The study aims at developing a neural network model to improve the performance of Human Phenotype Ontology (HPO) concept recognition tools. We used the terms, definitions, and comments about the phenotypic concepts in the HPO database to train our model. The document to be analyzed is first split into sentences and annotated with a base method to generate candidate concepts. The sentences, along with the candidate concepts, are then fed into the pre-trained model for re-ranking. Our model comprises the pre-trained BlueBERT and a feature selection module, followed by a contrastive loss. We re-ranked the results generated by three robust HPO annotation tools and compared the performance against most of the existing approaches. The experimental results show that our model can improve the performance of the existing methods. Significantly, it boosted 3.0% and 5.6% in F1 score on the two evaluated datasets compared with the base methods. It removed more than 80% of the false positives predicted by the base methods, resulting in up to 18% improvement in precision. Our model utilizes the descriptive data in the ontology and the contextual information in the sentences for re-ranking. The results indicate that the additional information and the re-ranking model can significantly enhance the precision of HPO concept recognition compared with the base method.

Computed tomography-identified phenotypes of small airway obstructions in chronic obstructive pulmonary disease

ABSTRACT

Chronic obstructive pulmonary disease (COPD) is a heterogeneous disease characteristic of small airway inflammation, obstruction, and emphysema. It is well known that spirometry alone cannot differentiate each separate component. Computed tomography (CT) is widely used to determine the extent of emphysema and small airway involvement in COPD. Compared with the pulmonary function test, small airway CT phenotypes can accurately reflect disease severity in patients with COPD, which is conducive to improving the prognosis of this disease. CT measurement of central airway morphology has been applied in clinical, epidemiologic, and genetic investigations as an inference of the presence and severity of small airway disease. This review will focus on presenting the current knowledge and methodologies in chest CT that aid in identifying discrete COPD phenotypes.

Building a semantically annotated corpus for chronic disease complications using two document types

Narrative information in electronic health records (EHRs) contains a wealth of information related to patient health conditions. In addition, people use Twitter to express their experiences regarding personal health issues, such as medical complaints, symptoms, treatments, lifestyle, and other factors. Both genres of text include different types of health-related information concerning disease complications and risk factors. Knowing detailed information about controlling disease risk factors has a great impact on modifying these risks and subsequently preventing disease complications. Text-mining tools provide efficient solutions to extract and integrate vital information related to disease complications hidden in the large volume of the narrative text. However, the development of text-mining tools depends on the availability of an annotated corpus. In response, we have developed the PrevComp corpus, which is annotated with information relevant to the identification of disease complications, underlying risk factors, and prevention measures, in the context of the interaction between hypertension and diabetes. The corpus is unique and novel in terms of the very specific topic in the biomedical domain and as an integration of information from both EHRs and tweets collected from Twitter. The annotation scheme was designed with guidance by a domain expert, and two further domain experts performed the annotation, resulting in a high-quality annotation, with agreement rate F-scores as high as 0.60 and 0.75 for EHRs and tweets, respectively.