A hybrid method based on semi-supervised learning for relation extraction in Chinese EMRs.

BACKGROUND: Building a large-scale medical knowledge graphs needs to automatically extract the relations between entities from electronic medical records (EMRs) . The main challenges are the scarcity of available labeled corpus and the identification of complexity semantic relations in text of Chinese EMRs. A hybrid method based on semi-supervised learning is proposed to extract the medical entity relations from small-scale complex Chinese EMRs. METHODS: The semantic features of sentences are extracted by a residual network and the long dependent information is captured by bidirectional gated recurrent unit. Then the attention mechanism is used to assign weights for the extracted features respectively, and the output of two attention mechanisms is integrated for relation prediction. We adjusted the training process with manually annotated small-scale relational corpus and bootstrapping semi-supervised learning algorithm, and continuously expanded the datasets during the training process. RESULTS: We constructed a small corpus of Chinese EMRs relation extraction based on the EMR datasets released at the China Conference on Knowledge Graph and Semantic Computing. The experimental results show that the best F1-score of the proposed method on the overall relation categories reaches 89.78%, which is 13.07% higher than the baseline CNN.

Effects of body size and root to shoot ratio on foliar nutrient resorption efficiency in Amaranthus mangostanus.

PREMISE OF THE STUDY: Nutrient resorption is essential for plant nutrient conservation. Large-bodied plants potentially have large nutrient sink pools and high nutrient flux. Whether and how nutrient resorption can be regulated by plant size and biomass allocation are yet unknown. METHODS: Using the herbaceous plant Amaranthus mangostanus in greenhouse experiments for two consecutive years, we measured plant biomass, height, and stem diameter and calculated the root to shoot biomass ratio (R/S ratio) and nutrient resorption efficiency (NuRE) to assess the effects of plant body size and biomass allocation on NuRE. NuRE was calculated as the percentage reduction in leaf nutrient concentration from green leaf to senesced leaf. KEY RESULTS: NuRE increased with plant biomass, height, and stem diameter, suggesting that the individuals with larger bodies, which led to a larger nutrient pool, tended to resorb proportionally more nutrients from the senescing leaves. NuRE decreased with increasing root to shoot ratio, which might have reflected the nutrient acquisition trade-offs between resorption from the senescent leaves and absorption from the soil. Increased root biomass allocation increased the proportion of nutrient acquisition through absorption more than through resorption. CONCLUSIONS: This study presented the first experimental evidence of how NuRE is linked to plant size (indicated by biomass, height, and stem diameter) and biomass allocation, suggesting that nutrient acquisition could be modulated by the size of the nutrient sink pool and its partitioning in plants, which can improve our understanding of a conservation mechanism for plant nutrients. The body size and root to shoot ratio effects might also partly explain previous inconsistent reports on the relationships between environmental nutrient availability and NuRE.