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.
AIMS OF THE REVIEW: To critically anayze the literature for the botany, traditional uses, phytochemistry, pharmacology, toxicity, and clinical trials of P. sarmentosum in order to provide a scientific consensus for further research and discovery of potential candidate drugs.
MATERIALS AND METHODS: The contents of this review were sourced from electronic databases including PubMed, SciFinder, Web of Science, Science Direct, Elsevier, Google Scholar, Chinese Knowledge On frastructure (CNKI), Wanfang, Chinese Scientific and Technological Periodical Database (VIP), Chinese Biomedical Database (CBM), Cochrane Controlled register of Clinical Trials, Clinical Trials. gov, and Chinese Clinical Trial Registry. Chinese medicine books published over the years were used to elucidate the traditional uses of P. sarmentosum and additional information was also collected from Yao Zhi website (https://db.yaozh.com/).
RESULTS: Phytochemical analyses of the chemical constituents of P. sarmentosum include essential oil, alkaloids, flavonoids, lignans, and steroids. The literature supports the ethnomedicinal uses of P. sarmentosum for the treatment of cold, gastritis, and rheumatoid joint pain, and further confirms its relatively new pharmacological activities, including anti-inflammatory, antineoplastic, and antipyretic activities. Other biological roles such as anti-osteoporosis, antibacterial, antidepressant, anti-atherosclerotic, and hypoglycemic activities have also been reported. However, the methodologies employed in individual studies are limited.
CONCLUSIONS: There is convincing evidence from both in vitro and in vivo studies supporting the traditional use of P. sarmentosum and it is imperative that natural bioactive compounds are examined further. More efforts should be focused on the pharmacodynamic constituents of P. sarmentosum to provide practical basis for quality control, and additional studies are needed to understand the mechanism of their action. Further studies on the comprehensive evaluation of medicinal quality and understandings of serum chemistry, multi-target network pharmacology, and molecular docking technology of P. sarmentosum are of great importance and should be considered.
METHODS: Head and neck cancer patients were recruited from July 2016 till March 2017 at National Cancer Institute, Ministry of Health, Malaysia. All study participants continued their standard oncology surveillance. Treatment group participants additionally received Chinese herbal treatment. The assessments included unstimulated salivary flow rate (USFR), stimulated salivary flow rate (SSFR), and QoL questionnaire.
RESULTS: Of 42 recruited participants, 28 were in the treatment group and 14 were in the control group. Participants were mainly Chinese (71.4%), stage III cancer (40.5%), and had nasopharynx cancer (76.2%). The commonly used single herbs were Wu Mei, San Qi, and Tian Hua Fen. Sha Shen Mai Dong Tang, Liu Wei Di Huang Wan, and Gan Lu Yin were the frequently prescribed herbal formulas. The baseline characteristics, USFR, SSFR, and QoL between control and treatment groups were comparable (p > 0.05). USFR between control and treatment groups were similar throughout the 6-month study period. SSFR for the treatment group significantly improved from 0.15 ± 0.28 ml/min (baseline) to 0.32 ± 0.22 ml/min (p = 0.04; at the 3rd month) and subsequently achieved 0.46 ± 0.23 ml/min (p = 0.001; at the 6th month). The treatment group had better QoL in terms of speech (p = 0.005), eating (p = 0.02), and head and neck pain (p = 0.04) at the 6th month.
CONCLUSION: Herbal treatment may improve xerostomia and QoL in post-radiotherapy head and cancer patients.
METHOD: Data was collected on 1085 CI recipients of as part of a prospective, longitudinal, observational, international, multi-centre, paediatric registry, initiated by Cochlear Ltd (Sydney, NSW, Australia). Outcome data from children (≤10 years old) implanted in routine practice was voluntarily entered into a central, externally hosted, e-platform. Collection occurred prior to initial device activation (baseline) and at six monthly follow-up intervals up to 24 months and then at 3 years post activation. Clinician reported baseline and follow up questionnaires and Categories of Auditory Performance version II (CAP-II) outcomes were collated. Self-reported evaluation forms and patient information were provided by the parent/caregiver/patient via the implant recipient baseline and follow up, Children Using Hearing Implants Quality of Life (CuHIQoL) and Speech Spatial Qualities (SSQ-P) Parents Version questionnaires.
RESULTS: Children were mainly bilaterally profoundly deaf, unilaterally implanted and used a contralateral hearing aid. Prior to implant 60% used signing or total communication as their main mode of communication. Mean age at implant was 3.2 ± 2.2 years (range 0-10 years). At baseline 8.6% were in mainstream education with no additional support and 82% had not yet entered school. After three years of implant use, 52% had entered mainstream education with no additional support and 38% had not yet entered school. In the sub-group of 141 children who were implanted at or after three years of age and were thus old enough to be in mainstream school at the three-year follow up, an even higher proportion (73%) were in mainstream education with no support. Quality of life scores for the child improved statistically significantly post implant compared to baseline and continued to improve significantly at each interval up to 3 years (p