Displaying all 2 publications

Abstract:
Sort:
  1. Li Y, Hui L, Zou L, Li H, Xu L, Wang X, et al.
    JMIR Med Inform, 2022 Oct 20;10(10):e41136.
    PMID: 36264604 DOI: 10.2196/41136
    BACKGROUND: With the rapid expansion of biomedical literature, biomedical information extraction has attracted increasing attention from researchers. In particular, relation extraction between 2 entities is a long-term research topic.

    OBJECTIVE: This study aimed to perform 2 multiclass relation extraction tasks of Biomedical Natural Language Processing Workshop 2019 Open Shared Tasks: relation extraction of Bacteria-Biotope (BB-rel) task and binary relation extraction of plant seed development (SeeDev-binary) task. In essence, these 2 tasks are aimed at extracting the relation between annotated entity pairs from biomedical texts, which is a challenging problem.

    METHODS: Traditional research methods adopted feature- or kernel-based methods and achieved good performance. For these tasks, we propose a deep learning model based on a combination of several distributed features, such as domain-specific word embedding, part-of-speech embedding, entity-type embedding, distance embedding, and position embedding. The multi-head attention mechanism is used to extract the global semantic features of an entire sentence. Meanwhile, we introduced a dependency-type feature and the shortest dependency path connecting 2 candidate entities in the syntactic dependency graph to enrich the feature representation.

    RESULTS: Experiments show that our proposed model has excellent performance in biomedical relation extraction, achieving F1 scores of 65.56% and 38.04% on the test sets of the BB-rel and SeeDev-binary tasks. Especially in the SeeDev-binary task, the F1 score of our model is superior to that of other existing models and achieves state-of-the-art performance.

    CONCLUSIONS: We demonstrated that the multi-head attention mechanism can learn relevant syntactic and semantic features in different representation subspaces and different positions to extract comprehensive feature representation. Moreover, syntactic dependency features can improve the performance of the model by learning dependency relation between the entities in biomedical texts.

  2. Viecelli AK, Teixeira-Pinto A, Valks A, Baer R, Cherian R, Cippà PE, et al.
    BMC Nephrol, 2022 Nov 19;23(1):372.
    PMID: 36402958 DOI: 10.1186/s12882-022-02987-1
    BACKGROUND: A functioning vascular access (VA) is crucial to providing adequate hemodialysis (HD) and considered a critically important outcome by patients and healthcare professionals. A validated, patient-important outcome measure for VA function that can be easily measured in research and practice to harvest reliable and relevant evidence for informing patient-centered HD care is lacking. Vascular Access outcome measure for function: a vaLidation study In hemoDialysis (VALID) aims to assess the accuracy and feasibility of measuring a core outcome for VA function established by the international Standardized Outcomes in Nephrology (SONG) initiative.

    METHODS: VALID is a prospective, multi-center, multinational validation study that will assess the accuracy and feasibility of measuring VA function, defined as the need for interventions to enable and maintain the use of a VA for HD. The primary objective is to determine whether VA function can be measured accurately by clinical staff as part of routine clinical practice (Assessor 1) compared to the reference standard of documented VA procedures collected by a VA expert (Assessor 2) during a 6-month follow-up period. Secondary outcomes include feasibility and acceptability of measuring VA function and the time to, rate of, and type of VA interventions. An estimated 612 participants will be recruited from approximately 10 dialysis units of different size, type (home-, in-center and satellite), governance (private versus public), and location (rural versus urban) across Australia, Canada, Europe, and Malaysia. Validity will be measured by the sensitivity and specificity of the data acquisition process. The sensitivity corresponds to the proportion of correctly identified interventions by Assessor 1, among the interventions identified by Assessor 2 (reference standard). The feasibility of measuring VA function will be assessed by the average data collection time, data completeness, feasibility questionnaires and semi-structured interviews on key feasibility aspects with the assessors.

    DISCUSSION: Accuracy, acceptability, and feasibility of measuring VA function as part of routine clinical practice are required to facilitate global implementation of this core outcome across all HD trials. Global use of a standardized, patient-centered outcome measure for VA function in HD research will enhance the consistency and relevance of trial evidence to guide patient-centered care.

    TRIAL REGISTRATION: Clinicaltrials.gov: NCT03969225. Registered on 31st May 2019.

Related Terms
Filters
Contact Us

Please provide feedback to Administrator (afdal@afpm.org.my)

External Links