In most of the existing mainstream action quality assessment methods, the score regression is performed on a complete action video to obtain the predicted score, which may prevent us from fully exploiting the multi-stage information in action video. In this paper, we attempt to divide a complete action video into clips according to the multiple phases it contains, and predict scores for each segment individually. In order to validate the effectiveness of the method, the FineDiving dataset is further divided into several action categories as the experimental dataset, and the improvement is applied to the mainstream USDL and CoRe methods. The proposed method achieves a significant performance enhancement in the metric of Spearman's correlation, which is commonly used in AQA tasks, thus validating the effectiveness of our proposed method.
Most existing medical event extraction methods have primarily adopted a simplex model based on either pattern matching or deep learning, which ignores the distribution characteristics of entities and events in the medical corpus. They have not categorized the granularity of event elements, leading to the poor generalization ability of the model. This paper proposes a diagnosis and treatment event extraction method in the Chinese language, fusing long short-level semantic dependency of the corpus, LSLSD, for solving these problems. LSLSD can effectively capture different levels of semantic information within and between event sentences in the electronic medical record (EMR) corpus. Moreover, the event arguments are divided into short word-level and long sentence-level, with the sequence annotation and pattern matching combined to realize multi-granularity argument recognition, as well as to improve the generalization ability of the model. Finally, this paper constructs a diagnosis and treatment event data set of Chinese EMRs by proposing a semi-automatic corpus labeling method, and an enormous number of experiment results show that LSLSD can improve the F1-value of event extraction task by 7.1% compared with the several strong baselines.
The Electronic Medical Record (EMR) contains a great deal of medical knowledge related to patients, which has been widely used in the construction of medical knowledge graphs. Previous studies mainly focus on the features based on surface semantics of EMRs for relation extraction, such as contextual feature, but the features of sentence structure in Chinese EMRs have been neglected. In this paper, a fusion dependency parsing-based relation extraction method is proposed. Specifically, this paper extends basic features with medical record feature and indicator feature that are applicable to Chinese EMRs. Furthermore, dependency syntactic features are introduced to analyse the dependency structure of sentences. Finally, the F1 value of relation extraction based on extended features is 4.87% higher than that of relation extraction based on basic features. And compared with the former, the F1 value of relation extraction based on fusion dependency parsing is increased by 4.39%. The results of experiments performed on a Chinese EMR data set show that the extended features and dependency parsing all contribute to the relation extraction.
1 September 2020 in Journal of Physics: Conference Series
Query expansion technology is a common method to solve the semantic deviation problem between questions and answers in the question answering system. In medical question answering field, the current semantic-based query expansion method has the following two shortcomings. (1) In the stage of obtaining candidate extended terms, only the medical entity’s concept in the query is used, but the medical entity’s association between questions and answers is ignored (2) During the extended terms filtering stage, the interference of negative medical entities to the mutual information is ignored. Given the above-mentioned deficiencies, this paper proposes a semantic query expansion method based on entity association in medical question answering. This method first combines the query intent with the inference association between medical entities to obtain candidate extended terms from the medical knowledge graph. Then filter candidate extended terms using the selection method based on negative medical entity recognition and mutual information and thus get the final extended terms according to the co-occurrence association between medical entities. Experimental results show that our proposed method outperforms the query expansion method based on the thesaurus which commonly used in the current medical question and answering system.