Search

Objective. The use of standardized structured reports (SSR) and suitable terminologies like SNOMED-CT can enhance data retrieval and analysis, fostering large-scale studies and collaboration. However, the still large prevalence of narrative reports in our laboratories warrants alternative and automated labeling approaches. In this project, natural language processing (NLP) methods were used to associate SNOMED-CT codes to structured and unstructured reports from an Italian Digital Pathology Department. Methods. Two NLP-based automatic coding systems (support vector machine, SVM, and long-short term memory, LSTM) were trained and applied to a series of narrative reports. Results. The 1163 cases were tested with both algorithms, showing good performances in terms of accuracy, precision, recall, and F1 score, with SVM showing slightly better performances as compared to LSTM (0.84, 0.87, 0.83, 0.82 vs 0.83, 0.85, 0.83, 0.82, respectively). The integration of an explainability allowed identification of terms and groups of words of importance, enabling fine-tuning, balancing semantic meaning and model performance. Conclusions. AI tools allow the automatic SNOMED-CT labeling of the pathology archives, providing a retrospective fix to the large lack of organization of narrative reports.

[ES] En el futuro están previstos los primeros viajes espaciales tripulados a Marte, lo que supone un gran reto para los astronautas y su salud. Mientras que los riesgos para la salud aumentan debido a las condiciones a las que estarán expuestos los astronautas a bordo, las medidas de seguridad para combatir esos riesgos se vuelven menos eficientes debido al mayor tiempo de comunicación con la Tierra. Esta situación ha impulsado la idea de crear un sistema de apoyo a la toma de decisiones clínicas a bordo para dotar a la tripulación de autonomía a la hora de enfrentarse a una emergencia médica. Para que el sistema de apoyo a la toma de decisiones clínicas a bordo funcione como es debido, es necesaria la interoperabilidad entre todos los sistemas informáticos participantes y sus funciones específicas. Este sistema está compuesto por diferentes módulos que cooperan entre sí enviando e intercambiando información médica relacionada con la salud de los miembros de la tripulación. Cada módulo tiene una función específica y proporciona datos relevantes para evaluar la emergencia médica y dar a los astronautas instrucciones precisas para llevar a cabo. En este trabajo se diseña un estándar de interoperabilidad semántica que estructura la información a interoperar entre diferentes sistemas informáticos dotando a la información de significado clínico. Para diseñar el estándar de interoperabilidad semántica se ha utilizado OpenEHR para establecer un estándar inicial. En el estándar inicial se han incorporado arquetipos y SNOMED-CT, que proporciona a los datos médicos un vínculo con las terminologías clínicas, para organizar todos los datos clínicos. Tras poner a prueba el estándar creado, se ha observado la correcta estructuración de la información clínica vinculada a una terminología. Los resultados mostraron una estructura que facilita la comprensión de los datos clínicos y la interpretación de la información clínica enviada., [EN] In the future, the first manned space voyages to Mars are planned, which poses a major challenge in terms of astronauts and their health. While health risks increase due to the conditions astronauts will be exposed to on board, safety measures to combat those risks become less efficient due to the increased communication time with Earth. This situation has prompted the idea of creating an onboard clinical decision support system to provide the crew with autonomy when facing a medical emergency. For the onboard clinical decision support system to work accordingly, interoperability between every participating computer system and their specific functions is needed. This system is composed of different modules which cooperate with each other by sending and exchanging medical information related to the crew members’ health. Each module has a specific function and provides relevant data to assess the medical emergency and to give the astronauts precise instructions to carry out. In this work, a semantic interoperability standard is designed, which structures the information to be interoperated between different computer systems by endowing information with clinical meaning. To design the semantic interoperability standard, OpenEHR has been used to establish an initial standard. Archetypes and SNOMED-CT, which provides medical data with a link to clinical terminologies, have been incorporated into the initial standard to organize all clinical data. After putting the created standard to the test, the correct structuring of the clinical information linked to a terminology has been observed. The results showed a structure which makes the clinical data easier to understand and facilitates the interpretation of the clinical information sent.

[ES] En el futuro están previstos los primeros viajes espaciales tripulados a Marte, lo que supone un gran reto para los astronautas y su salud. Mientras que los riesgos para la salud aumentan debido a las condiciones a las que estarán expuestos los astronautas a bordo, las medidas de seguridad para combatir esos riesgos se vuelven menos eficientes debido al mayor tiempo de comunicación con la Tierra. Esta situación ha impulsado la idea de crear un sistema de apoyo a la toma de decisiones clínicas a bordo para dotar a la tripulación de autonomía a la hora de enfrentarse a una emergencia médica. Para que el sistema de apoyo a la toma de decisiones clínicas a bordo funcione como es debido, es necesaria la interoperabilidad entre todos los sistemas informáticos participantes y sus funciones específicas. Este sistema está compuesto por diferentes módulos que cooperan entre sí enviando e intercambiando información médica relacionada con la salud de los miembros de la tripulación. Cada módulo tiene una función específica y proporciona datos relevantes para evaluar la emergencia médica y dar a los astronautas instrucciones precisas para llevar a cabo. En este trabajo se diseña un estándar de interoperabilidad semántica que estructura la información a interoperar entre diferentes sistemas informáticos dotando a la información de significado clínico. Para diseñar el estándar de interoperabilidad semántica se ha utilizado OpenEHR para establecer un estándar inicial. En el estándar inicial se han incorporado arquetipos y SNOMED-CT, que proporciona a los datos médicos un vínculo con las terminologías clínicas, para organizar todos los datos clínicos. Tras poner a prueba el estándar creado, se ha observado la correcta estructuración de la información clínica vinculada a una terminología. Los resultados mostraron una estructura que facilita la comprensión de los datos clínicos y la interpretación de la información clínica enviada., [EN] In the future, the first manned space voyages to Mars are planned, which poses a major challenge in terms of astronauts and their health. While health risks increase due to the conditions astronauts will be exposed to on board, safety measures to combat those risks become less efficient due to the increased communication time with Earth. This situation has prompted the idea of creating an onboard clinical decision support system to provide the crew with autonomy when facing a medical emergency. For the onboard clinical decision support system to work accordingly, interoperability between every participating computer system and their specific functions is needed. This system is composed of different modules which cooperate with each other by sending and exchanging medical information related to the crew members’ health. Each module has a specific function and provides relevant data to assess the medical emergency and to give the astronauts precise instructions to carry out. In this work, a semantic interoperability standard is designed, which structures the information to be interoperated between different computer systems by endowing information with clinical meaning. To design the semantic interoperability standard, OpenEHR has been used to establish an initial standard. Archetypes and SNOMED-CT, which provides medical data with a link to clinical terminologies, have been incorporated into the initial standard to organize all clinical data. After putting the created standard to the test, the correct structuring of the clinical information linked to a terminology has been observed. The results showed a structure which makes the clinical data easier to understand and facilitates the interpretation of the clinical information sent.

Nous décrivons la collaboration tripartite terminologie/TALN/sciences de la santé développée dans le cadre du projet iMediate (Interoperability of Medical Data through Information Extraction and Term Encoding). Cette recherche vise à mettre au point des technologies propres à réaliser automatiquement une représentation structurée des données des patients à partir de textes hospitaliers non structurés et de données cliniques structurées. Le partenaire terminologique intervient en amont de la structuration de documents médicaux peu ou pas structurés. Son rôle est, par repérage des variations observables dans l’usage, d’enrichir les ressources terminologiques qui seront exploitées par les trois applications – annotateur, catégoriseur et moteur de recherche – développées en aval du projet. Les nomenclatures médicales ne suffisant pas à rendre compte de la variation rencontrée dans les textes médicaux libres, l’enrichissement terminologique consiste à relever les variations d’usage dans un corpus, et à les associer, par regroupement synonymique, aux entrées des nomenclatures officielles. La taille du corpus iMediate confinant au traitement de données massives (big data), une adaptation de la méthodologie classique d’exploitation semi- automatique de corpus a été nécessaire. Plutôt qu’un traitement séquentiel des unités terminologiques, les partenaires ont élaboré une méthodologie d’extraction et de validation des candidats termes par cycles d’amélioration continue des processus manuels et automatiques.

In this paper we present two tools for facing task 2 in CLEF eHealth 2016. The first one is a semantic tagger aiming to detect relevant entities in French medical documents, tagging them with their appropriate semantic class and normalizing them with the Semantic Groups codes defined in the UMLS. It is based on a distant learning approach that uses several SVM classifiers that are combined to give a single result. The second tool is based on a symbolic procedure to obtain the English translation of each medical term and looks for normalization information in public accessible resources., Peer Reviewed, Postprint (published version)

This paper presents a rule-based method for the detection and normalization of medical entities using SNOMED-CT which, although based on knowledge stored in terminological resources, allows some flexibility in order to account for the language variation typical of medical texts. Our system is based on the software Unitex and is one of the few to code French medical texts with SNOMED-CT concept identifiers. Our evaluation quantifies the benefits of such a flexible approach, but also emphasizes terminological resource shortcomings for the processing of medical reports written in French. Finally, our methodology is an interesting alternative to supervised training, as the extraction rules require limited development.

This article describes information retrieval, natural language processing and text mining of electronic patient record text, also called clinical text. Clinical text is written by physicians and nurses to document the health care process of the patient. First we describe some characteristics of clinical text, followed by the automatic preprocessing of the text that is necessary for making it usable for some applications. We also describe some applications for clinicians including spelling and grammar checking, ICD-10 diagnosis code assignment, as well as other applications for hospital management such as ICD-10 diagnosis code validation and detection of adverse events such as hospital acquired infections. Part of the preprocessing makes the clinical text useful for faceted search, although clinical text already has some keys for performing faceted search such as gender, age, ICD-10 diagnosis codes, ATC drug codes, etc. Preprocessing makes use of ICD-10 codes and the SNOMED-CT textual descriptions. ICD-10 codes and SNOMED-CT are available in several languages and can be considered the modern Greek or Latin of medical language. The basic research presented here has its roots in the challenges described by the health care sector. These challenges have been partially solved in academia, and we believe the solutions will be adapted to the health care sector in real world applications.

Uno de los retos actuales de la informática médica es lograr la interoperabilidad semántica entre los sistemas de información de distintas instituciones sanitarias. La interoperabilidad completa permitirá que los sistemas intercambien y comprendan automáticamente la información de los pacientes y facilitará el acceso completo a la información de un paciente desde cualquier sistema o institución. Varios proyectos a nivel europeo han definido una hoja de ruta con recomendaciones y acciones encaminadas a lograr la interoperabilidad semántica de la Historia Clínica Electrónica (HCE), entre las que se incluyen: (i) el uso de modelos de datos clínicos estructurados (arquetipos) para capturar de forma ordenada y sistemática la información de pacientes en escenarios clínicos determinados y (ii) la integración de terminologías clínicas dentro de la HCE, principalmente mediante la creación de enlaces entre ítems de información clínica (presentes en los modelos de datos de la HCE) con conceptos estándar de las terminologías. La tesis se centra en el desarrollo de métodos avanzados para enlazar de forma automática la información clínica de la HCE, especialmente de arquetipos, con la terminología SNOMED-CT. También, se han desarrollado aplicaciones que demuestran las ventajas de integrar una terminología de referencia en repositorios de datos clínicos. La tesis demuestra que es factible automatizar el enlazado entre la información clínica de los arquetipos y SNOMED-CT, con una precisión y cobertura elevada, si se hace uso de la información contextual y estructural implícita en los arquetipos y en SNOMED-CT.

Advancement in medical therapies requires clinical research to test a therapy's efficacy and safety in humans, yet this process can be slow, inefficient, and error prone. Two key problems resulting in delays are: 1) the translation of research ideas into relevant, testable hypotheses, and 2) the enrollment of subjects in clinical research that test these hypotheses. A crucial need for computational support is indicated by several challenges: the complex nature of the medical field, the large number of possible health hypotheses, and the variety and size of clinical research data. Strong social and economic forces are changing the way healthcare works. Clinical research is evolving to emphasize increased efficiencies, productivity, and inclusion of health-consumer-centered outcomes. In the period of change, this research seeks to address these challenges through the design of a Health Research Exchange (HRE) that is a development and communication hub for health research hypotheses. The HRE improves upon existing health research databases, such as clinicaltrials.gov, by addressing the process of hypothesis formation. It seeks to facilitate health hypothesis development by non-medical professionals, enabling health research to become more relevant to health consumers.The HRE is centered on the innovative Molecular Hypothesis Model (MoHM), featuring structured representation and reasoning about health hypotheses. The MoHM &mdash based upon the principles of connectedness, modularity, and functional specificity &mdash decomposes hypotheses into reusable, function-specific domains connected to each other and to facts, relationships, and other hypotheses. Implemented in a Hybrid Hypergraph Description Logic (HDL), integrating SNOMED&ndashCT medical terminology with hypergraph and Boolean formula representations, the MoHM overlays functional domains specialized for efficient, scalable reasoning with hypergraph connectivity of concepts. The Molecular Hypothesis

Advancement in medical therapies requires clinical research to test a therapy's efficacy and safety in humans, yet this process can be slow, inefficient, and error prone. Two key problems resulting in delays are: 1) the translation of research ideas into relevant, testable hypotheses, and 2) the enrollment of subjects in clinical research that test these hypotheses. A crucial need for computational support is indicated by several challenges: the complex nature of the medical field, the large number of possible health hypotheses, and the variety and size of clinical research data. Strong social and economic forces are changing the way healthcare works. Clinical research is evolving to emphasize increased efficiencies, productivity, and inclusion of health-consumer-centered outcomes. In the period of change, this research seeks to address these challenges through the design of a Health Research Exchange (HRE) that is a development and communication hub for health research hypotheses. The HRE improves upon existing health research databases, such as clinicaltrials.gov, by addressing the process of hypothesis formation. It seeks to facilitate health hypothesis development by non-medical professionals, enabling health research to become more relevant to health consumers.The HRE is centered on the innovative Molecular Hypothesis Model (MoHM), featuring structured representation and reasoning about health hypotheses. The MoHM &mdash based upon the principles of connectedness, modularity, and functional specificity &mdash decomposes hypotheses into reusable, function-specific domains connected to each other and to facts, relationships, and other hypotheses. Implemented in a Hybrid Hypergraph Description Logic (HDL), integrating SNOMED&ndashCT medical terminology with hypergraph and Boolean formula representations, the MoHM overlays functional domains specialized for efficient, scalable reasoning with hypergraph connectivity of concepts. The Molecular Hypothesis

Uno de los problemas actuales en el dominio de la salud es reutilizar y compartir la información clínica entre profesionales, ya que ésta se encuentra escrita usando terminologías específicas. Una posible solución es usar un recurso de conocimiento común sobre el que mapear la información existente. Nuestro objetivo es comprobar si la adición de conocimiento semántico superficial puede mejorar los mapeados establecidos. Para ello experimentamos con un conjunto de etiquetas de NANDA-I y con un conjunto de descripciones de SNOMED-CT en castellano. Los resultados obtenidos en los experimentos muestran que la inclusión de conocimiento semántico superficial mejora significativamente el mapeado léxico entre los dos recursos estudiados., In health domain, one of the current problems is the reusing and the sharing the clinical information between professionals, due to this information is written using specific terminologies. One possible solution is to use a common knowledge resource for mapping the existing information. In this paper, our aim is to analyze if the use of lexical similarity algorithms enriched with shallow semantic knowledge can improve these mappings. In order to achieve this, we experiment with a set of NANDA-I labels and a set of SNOMED-CT descriptions in Spanish. The results obtained show that the addition of shallow semantic knowledge significantly improves the lexical mapping between both studied resources.

Animal classification needs vary by use and application. The Linnaean taxonomy is an important animal classification scheme but does not portray key animal identifying information like sex, age group, physiologic stage, living environment and role in production systems such as farms. Ontologies are created and used for defining, organizing and classifying information in a domain to enable learning and sharing of information. This work develops an ontology of animal classes that form the basis for communication of animal identifying information among animal managers, medical professionals caring for animals and biomedical researchers involved in disciplines as diverse as wildlife ecology and dairy science. The Animals in Context Ontology (ACO) was created from an extension and subset of the Systematized Nomenclature of Medicine — Clinical Terms (SNOMED-CT). The principles of the Open Biological and Biomedical Ontologies (OBO) Foundry were followed and freely available tools were used. ACO includes normal development and physiologic animal classes as well animal classes where humans have assigned the animal's role. ACO is interoperable with and includes classes from other OBO Foundry ontologies such as the Gene Ontology (GO). Meeting many of the OBO Foundry principles was straightforward but difficulties were encountered with missing and problematic content in some of the OBO ontologies. Additions and corrections were submitted to four ontologies. Some information in ACO could not be represented formally because of inconsistency in husbandry practices. ACO classes are of interest to science, medicine and agriculture, and can connect information between animal and human systems to enable knowledge discovery.

Uno de los problemas actuales en el dominio de la salud es reutilizar y compartir la información clínica entre profesionales, ya que ésta se encuentra escrita usando terminologías específicas. Una posible solución es usar un recurso de conocimiento común sobre el que mapear la información existente. Nuestro objetivo es comprobar si la adición de conocimiento semántico superficial puede mejorar los mapeados establecidos. Para ello experimentamos con un conjunto de etiquetas de NANDA-I y con un conjunto de descripciones de SNOMED-CT en castellano. Los resultados obtenidos en los experimentos muestran que la inclusión de conocimiento semántico superficial mejora significativamente el mapeado léxico entre los dos recursos estudiados., In health domain, one of the current problems is the reusing and the sharing the clinical information between professionals, due to this information is written using specific terminologies. One possible solution is to use a common knowledge resource for mapping the existing information. In this paper, our aim is to analyze if the use of lexical similarity algorithms enriched with shallow semantic knowledge can improve these mappings. In order to achieve this, we experiment with a set of NANDA-I labels and a set of SNOMED-CT descriptions in Spanish. The results obtained show that the addition of shallow semantic knowledge significantly improves the lexical mapping between both studied resources.