Distantly Supervised Biomedical Relation Extraction Via Negative Learning and Noisy Student Self-Training.

Biomedical relation extraction aims to identify underlying relationships among entities, such as gene associations and drug interactions, within biomedical texts. Despite advancements in relation extraction in general knowledge domains, the scarcity of labeled training data remains a significant challenge in the biomedical field. This paper provides a novel approach for biomedical relation extraction that leverages a noisy student self-training strategy combined with negative learning. This method addresses the challenge of data insufficiency by utilizing distantly supervised data to generate high-quality labeled samples. Negative learning, as opposed to traditional positive learning, offers a more robust mechanism to discern and relabel noisy samples, preventing model overfitting. The integration of these techniques ensures enhanced noise reduction and relabeling capabilities, leading to improved performance even with noisy datasets. Experimental results demonstrate the effectiveness of the proposed framework in mitigating the impact of noisy data and outperforming existing benchmarks.

A comparative study of elective nodal irradiation and involved field irradiation in elderly patients with advanced esophageal cancer.

Objective: This comparative study aimed to explore the feasibility of involved field irradiation (IFI) in the radiotherapy of elderly patients with advanced esophageal cancer, compared with elective nodal irradiation (ENI). Methods: A total of 245 elderly patients (age ≥70 years) with advanced esophageal cancer, who received radiotherapy in our department from January 2014 to December 2020, were divided into the ENI group (n=111) and the IFI group (n=134). Clinical efficacy, toxicities, survival rates, treatment failures, and multifactorial survival analyses were conducted for both groups. Results: The ENI group and the IFI group showed no significant differences in terms of short-term efficacy (91.9% vs 91.0%, P=0.814), 1-year overall survival (OS) (81.1% vs 74.6%, P=0.228), 2-year OS (22.5% vs 25.4%, P= 0.603), 1-year progression-free survival (PFS) (56.8% vs 51.5%, P= 0.198), 2-year PFS (8.1% vs 9.0%, P=0.814), regional failures (38.7% vs 31.3%, P=0.226), and distant metastasis (21.6% vs 14.9%, P=0.174). The median overall survival (OS) was 19 months in the ENI group and 18 months in the IFI group (Log-rankχ 2 = 0.012, P=0.913). The median progression-free survival (PFS) was 13 months in the ENI group and 11 months in the IFI group (Log-rankχ 2 = 1.834, P=0.176). There were no significant statistical differences in both OS and PFS (P>0.05). The incidence of grade ≥3 radiation pneumonia and grade ≥3 radiation esophagitis in the IFI group was 8.2% and 11.2%, respectively, which were significantly lower than those in the ENI group (17.1%, P=0.034; 21.6%, P=0.026). Univariate analysis revealed that age, gender, T stage, N stage, and synchronous chemotherapy were factors affecting prognosis. Multivariate analysis showed that age, gender, T stage, and synchronous chemotherapy were independent prognostic factors, with hazard ratios of 1.227, 1.466, 2.441, and 2.714, and P values of <0.001, 0.006, <0.001, and<0.001, respectively. Conclusion: IFI is a suitable choice for elderly patients with advanced esophageal cancer, as it yields similar efficacy to ENI while reducing toxicities. Age, gender, T stage, and synchronous chemotherapy are independent prognostic factors for elderly patients with esophageal cancer.

Drug-drug interaction prediction with Wasserstein Adversarial Autoencoder-based knowledge graph embeddings.

An interaction between pharmacological agents can trigger unexpected adverse events. Capturing richer and more comprehensive information about drug-drug interactions (DDIs) is one of the key tasks in public health and drug development. Recently, several knowledge graph (KG) embedding approaches have received increasing attention in the DDI domain due to their capability of projecting drugs and interactions into a low-dimensional feature space for predicting links and classifying triplets. However, existing methods only apply a uniformly random mode to construct negative samples. As a consequence, these samples are often too simplistic to train an effective model. In this paper, we propose a new KG embedding framework by introducing adversarial autoencoders (AAEs) based on Wasserstein distances and Gumbel-Softmax relaxation for DDI tasks. In our framework, the autoencoder is employed to generate high-quality negative samples and the hidden vector of the autoencoder is regarded as a plausible drug candidate. Afterwards, the discriminator learns the embeddings of drugs and interactions based on both positive and negative triplets. Meanwhile, in order to solve vanishing gradient problems on the discrete representation-an inherent flaw in traditional generative models-we utilize the Gumbel-Softmax relaxation and the Wasserstein distance to train the embedding model steadily. We empirically evaluate our method on two tasks: link prediction and DDI classification. The experimental results show that our framework can attain significant improvements and noticeably outperform competitive baselines. Supplementary information: Supplementary data and code are available at https://github.com/dyf0631/AAE_FOR_KG.