No abstract

Background Word embedding technologies, a set of language modeling and feature learning techniques in natural language processing (NLP), are now used in a wide range of applications. However, no formal evaluation and comparison have been made on the ability of each of the 3 current most famous unsupervised implementations (Word2Vec, GloVe, and FastText) to keep track of the semantic similarities existing between words, when trained on the same dataset. Objective The aim of this study was to compare embedding methods trained on a corpus of French health-related documents produced in a professional context. The best method will then help us develop a new semantic annotator. Methods Unsupervised embedding models have been trained on 641,279 documents originating from the Rouen University Hospital. These data are not structured and cover a wide range of documents produced in a clinical setting (discharge summary, procedure reports, and prescriptions). In total, 4 rated evaluation tasks were defined (cosine similarity, odd one, analogy-based operations, and human formal evaluation) and applied on each model, as well as embedding visualization. Results Word2Vec had the highest score on 3 out of 4 rated tasks (analogy-based operations, odd one similarity, and human validation), particularly regarding the skip-gram architecture. Conclusions Although this implementation had the best rate for semantic properties conservation, each model has its own qualities and defects, such as the training time, which is very short for GloVe, or morphological similarity conservation observed with FastText. Models and test sets produced by this study will be the first to be publicly available through a graphical interface to help advance the French biomedical research.

Gastroparesis is defined as delayed gastric emptying (GE) associated with symptoms, including nausea, vomiting, postprandial fullness, early satiation and epigastric pain. 1 Gastroparesis may lead, in the most severe cases, to altered quality of life and nutritional impairment and ultimately to death. The epidemiology as well as the natural history of gastroparesis remains poorly investigated. A recent study evaluated the economic burden of gastroparesis between US$4000 and US$9000 per patient per year. 2 Two studies based on symptom

No abstract

No abstract

BackgroundThe Foundational Model of Anatomy (FMA) is the reference ontology regarding human anatomy. FMA vocabulary was integrated into the Health Multi Terminological Portal (HMTP) developed by CISMeF based on the CISMeF Information System which also includes 26 other terminologies and controlled vocabularies, mainly in French. However, FMA is primarily in English. In this context, the translation of FMA English terms into French could also be useful for searching and indexing French anatomy resources. Various studies have investigated automatic methods to assist the translation of medical terminologies or create multilingual medical vocabularies. The goal of this study was to facilitate the translation of FMA vocabulary into French.MethodsWe compare two types of approaches to translate the FMA terms into French. The first one is UMLS-based on the conceptual information of the UMLS metathesaurus. The second method is lexically-based on several Natural Language Processing (NLP) tools.ResultsThe UMLS-based approach produced a translation of 3,661 FMA terms into French whereas the lexical approach produced a translation of 3,129 FMA terms into French. A qualitative evaluation was made on 100 FMA terms translated by each method. For the UMLS-based approach, among the 100 translations, 52% were manually rated as "very good" and only 7% translations as "bad". For the lexical approach, among the 100 translations, 47% were rated as "very good" and 20% translations as "bad".ConclusionsOverall, a low rate of translations were demonstrated by the two methods. The two approaches permitted us to semi-automatically translate 3,776 FMA terms from English into French, this was to added to the existing 10,844 French FMA terms in the HMTP (4,436 FMA French terms and 6,408 FMA terms manually translated).

Background The huge amount of clinical, administrative, and demographic data recorded and maintained by hospitals can be consistently aggregated into health data warehouses with a uniform data model. In 2017, Rouen University Hospital (RUH) initiated the design of a semantic health data warehouse enabling both semantic description and retrieval of health information. Objective This study aimed to present a proof of concept of this semantic health data warehouse, based on the data of 250,000 patients from RUH, and to assess its ability to assist health professionals in prescreening eligible patients in a clinical trials context. Methods The semantic health data warehouse relies on 3 distinct semantic layers: (1) a terminology and ontology portal, (2) a semantic annotator, and (3) a semantic search engine and NoSQL (not only structured query language) layer to enhance data access performances. The system adopts an entity-centered vision that provides generic search capabilities able to express data requirements in terms of the whole set of interconnected conceptual entities that compose health information. Results We assessed the ability of the system to assist the search for 95 inclusion and exclusion criteria originating from 5 randomly chosen clinical trials from RUH. The system succeeded in fully automating 39% (29/74) of the criteria and was efficiently used as a prescreening tool for 73% (54/74) of them. Furthermore, the targeted sources of information and the search engine–related or data-related limitations that could explain the results for each criterion were also observed. Conclusions The entity-centered vision contrasts with the usual patient-centered vision adopted by existing systems. It enables more genericity in the information retrieval process. It also allows to fully exploit the semantic description of health information. Despite their semantic annotation, searching within clinical narratives remained the major challenge of the system. A finer annotation of the clinical texts and the addition of specific functionalities would significantly improve the results. The semantic aspect of the system combined with its generic entity-centered vision enables the processing of a large range of clinical questions. However, an important part of health information remains in clinical narratives, and we are currently investigating novel approaches (deep learning) to enhance the semantic annotation of those unstructured data.

Objective: The aim of this paper is to propose an extended translation of the MeSH thesaurus based on Wikipedia pages. Methods: A mapping was realized between each MeSH descriptor (preferred terms and synonyms) and corresponding Wikipedia pages. Results: A tool called “WikiMeSH” has been developed. Among the top 20 languages of this study, seven have currently no MeSH translations: Arabic, Catalan, Farsi (Iran), Mandarin Chinese, Korean, Serbian, and Ukrainian. For these seven languages, WikiMeSH is proposing a translation for 47% for Arabic to 34% for Serbian. Conclusion: WikiMeSH is an interesting tool to translate the MeSH thesaurus and other health terminologies and ontologies based on a mapping to Wikipedia pages.